Federal Motor Vehicle Safety Standards; Electronic Stability Control Systems for Heavy Vehicles

Federal RegisterMay 23, 2012

Ask Donna

What actually matters in this document.

Text

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Part 571

[Docket No. NHTSA-2012-0065]

RIN 2127-AK97

Federal Motor Vehicle Safety Standards; Electronic Stability Control Systems for Heavy Vehicles

AGENCY:

National Highway Traffic Safety Administration (NHTSA), Department of Transportation (DOT).

ACTION:

Notice of proposed rulemaking (NPRM).

SUMMARY:

This document proposes to establish a new Federal Motor Vehicle Safety Standard No. 136 to require electronic stability control (ESC) systems on truck tractors and certain buses with a gross vehicle weight rating of greater than 11,793 kilograms (26,000 pounds). ESC systems in truck tractors and large buses are designed to reduce untripped rollovers and mitigate severe understeer or oversteer conditions that lead to loss of control by using automatic computer-controlled braking and reducing engine torque output.

In 2012, we expect that about 26 percent of new truck tractors and 80 percent of new buses affected by this proposed rule will be equipped with ESC systems. We believe that ESC systems could prevent 40 to 56 percent of untripped rollover crashes and 14 percent of loss-of-control crashes. By requiring that ESC systems be installed on truck tractors and large buses, this proposal would prevent 1,807 to 2,329 crashes, 649 to 858 injuries, and 49 to 60 fatalities at less than $3 million per equivalent life saved, while generating positive net benefits.

DATES:

Comments:

Submit comments on or before August 21, 2012.

Public Hearing:

NHTSA will hold a public hearing in the summer of 2012. NHTSA will announce the date for the hearing in a supplemental

Federal Register

document. The agency will accept comments to the rulemaking at this hearing.

ADDRESSES:

You may submit comments electronically [identified by DOT Docket Number NHTSA-2012-0065] by visiting the following Web site

•

Federal eRulemaking Portal:

Go to

http://www.regulations.gov.

Follow the online instructions for submitting comments.

Alternatively, you can file comments using the following methods:

•

Mail:

Docket Management Facility: U.S. Department of Transportation, 1200 New Jersey Avenue SE., West Building Ground Floor, Room W12-140, Washington, DC 20590-0001

•

Hand Delivery or Courier:

West Building Ground Floor, Room W12-140, 1200 New Jersey Avenue SE., between 9 a.m. and 5 p.m. ET, Monday through Friday, except Federal holidays.

•

Fax:

(202) 493-2251

Instructions:

For detailed instructions on submitting comments and additional information on the rulemaking process, see the Public Participation heading of the

SUPPLEMENTARY INFORMATION

section of this document. Note that all comments received will be posted without change to

http://www.regulations.gov,

including any personal information provided. Please see the Privacy Act heading below.

Privacy Act:

Anyone is able to search the electronic form of all comments received into any of our dockets by the name of the individual submitting the comment (or signing the comment, if submitted on behalf of an association, business, labor union, etc.). You may review DOT's complete Privacy Act Statement in the

Federal Register

published on April 11, 2000 (65 FR 19477-78).

Docket:

For access to the docket to read background documents or comments received, go to

http://www.regulations.gov.

Follow the online instructions for accessing the dockets.

FOR FURTHER INFORMATION CONTACT:

For technical issues, you may contact George Soodoo, Office of Crash Avoidance Standards, by telephone at (202) 366-4931, and by fax at (202) 366-7002. For legal issues, you may contact David Jasinski, Office of the Chief Counsel, by telephone at (202) 366-2992, and by fax at (202) 366-3820. You may send mail to both of these officials at the National Highway Traffic Safety Administration, 1200 New Jersey Avenue SE., Washington, DC 20590.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive Summary

II. Safety Problem

A. Heavy Vehicle Crash Problem

B. Contributing Factors in Rollover and Loss-of-Control Crashes

C. NTSB Safety Recommendations

D. Motorcoach Safety Plan

E. International Regulation

III. Stability Control Technologies

A. Dynamics of a Rollover

B. Description of RSC System Functions

C. Description of ESC System Functions

D. How ESC Prevents Loss of Control

E. Situations in Which Stability Control Systems May Not Be Effective

F. Difference in Vehicle Dynamics Between Light Vehicles and Heavy Vehicles

IV. Research and Testing

A. UMTRI Study

B. Simulator Study

C. NHTSA Track Testing

1. Effects of Stability Control Systems—Phase I

2. Developing a Dynamic Test Maneuver and Performance Measure To Evaluate Roll Stability—Phase II

(a) Test Maneuver Development

(b) Performance Measure Development

3. Developing a Dynamic Test Maneuver and Performance Measure To Evaluate Yaw Stability—Phase III

(a) Test Maneuver Development

(b) Performance Measure Development

4. Large Bus Testing

D. Truck & Engine Manufacturers Association Testing

1. Slowly Increasing Steer Maneuver

2. Ramp Steer Maneuver

3. Sine With Dwell Maneuver

4. Ramp With Dwell Maneuver

5. Vehicle J Testing

(a) EMA Testing of Vehicle J

(b) NHTSA Testing of EMA's Vehicle J

E. Other Industry Research

1. Decreasing Radius Test

2. Lane Change on a Large Diameter Circle

3. Yaw Control Tests

V. Agency Proposal

A. NHTSA's Statutory Authority

B. Applicability

1. Vehicle Types

2. Retrofitting In-Service Truck Tractors, Trailers, and Buses

3. Exclusions From Stability Control Requirement

C. ESC System Capabilities

1. Choosing ESC vs. RSC

2. Definition of ESC

D. ESC Disablement

E. ESC Malfunction Detection, Telltale, and Activation Indicator

1. ESC Malfunction Detection

2. ESC Malfunction Telltale

3. ESC Activation Indicator

F. Performance Requirements and Compliance Testing

1. Characterization Test—SIS

2. Roll and Yaw Stability Test—SWD

(a) Roll Stability Performance

(b) Yaw Stability Performance

3. Alternative Test Maneuvers Considered

(a) Characterization Maneuver

(b) Roll Stability Test Maneuvers

(d) Lack of an Understeer Test

4. ESC Malfunction Test

5. Test Instrumentation and Equipment

(a) Outriggers

(b) Automated Steering Machine

(d) Control Trailer

(e) Sensors

6. Test Conditions

(a) Ambient Conditions

(b) Road Test Surface

(d) Tires

(e) Mass Estimation Drive Cycle

(f) Brake Conditioning

7. Data Filtering and Post Processing

G. Compliance Dates and Implementation Schedule

VI. Benefits and Costs

A. System Effectiveness

B. Target Crash Population

C. Benefits Estimate

D. Cost Estimate

E. Cost Effectiveness

F. Comparison of Regulatory Alternatives

VII. Public Participation

VIII. Regulatory Analyses and Notices

A. Executive Order 12866, Executive Order 13563, and DOT Regulatory Policies and Procedures

B. Regulatory Flexibility Act

C. Executive Order 13132 (Federalism)

D. Executive Order 12988 (Civil Justice Reform)

E. Protection of Children From Environmental Health and Safety Risks

F. Paperwork Reduction Act

G. National Technology Transfer and Advancement Act

H. Unfunded Mandates Reform Act

I. National Environmental Policy Act

J. Plain Language

K. Regulatory Identifier Number (RIN)

L. Privacy Act

I. Executive Summary

The agency proposes to reduce rollover and loss of directional control of truck tractors and large buses by establishing a new standard, Federal Motor Vehicle Safety Standard (FMVSS) No. 136,

Electronic Stability Control Systems for Heavy Vehicles.

The standard would require truck tractors and certain buses

with a gross vehicle weight rating (GVWR) of greater than 11,793 kilograms (26,000 pounds) to be equipped with an electronic stability control (ESC) system that meets the equipment and performance criteria of the standard. ESC systems use engine torque control and computer-controlled braking of individual wheels to assist the driver in maintaining control of the vehicle and maintaining its heading in situations in which the vehicle is becoming roll unstable (

i.e.,

wheel lift potentially leading to rollover) or experiencing loss of control (

i.e.,

deviation from driver's intended path due to understeer, oversteer, trailer swing or any other yaw motion leading to directional loss of control). In such situations, intervention by the ESC system can assist the driver in maintaining control of the vehicle, thereby preventing fatalities and injuries associated with vehicle rollover or collision. Based on the agency's estimates regarding the effectiveness of ESC systems, we believe that an ESC standard could annually prevent 1,807 to 2,329 crashes, 649 to 858 injuries, and 49 to 60 fatalities, while providing net economic benefits.

As explained later in this notice, the applicability of this proposed standard to buses would be similar to the applicability of NHTSA's proposal to require seat belts on certain buses. These buses would have 16 or more designated seating positions (including the driver), at least 2 rows of passenger seats that are rearward of the driver's seating position and forward-facing or can convert to forward-facing without the use of tools. As with the seat belt NPRM, this proposed rule would exclude school buses and urban transit buses sold for operation as a common carrier in urban transportation along a fixed route with frequent stops.

There have been two types of stability control systems developed for heavy vehicles. A roll stability control (RSC) system is designed to prevent rollover by decelerating the vehicle using braking and engine torque control. The other type of stability control system is ESC, which includes all of the functions of an RSC system plus the ability to mitigate severe oversteer or understeer by automatically applying brake force at selected wheel-ends to help maintain directional control of a vehicle. To date, ESC and RSC systems for heavy vehicles have been developed for air-braked vehicles. Truck tractors and buses covered by this proposed rule make up a large proportion of air-braked heavy vehicles and a large proportion of the heavy vehicles involved in both rollover crashes and total crashes. Based on information we have received to date, the agency has tentatively determined that ESC and RSC systems are not available for hydraulic-braked medium or heavy vehicles.

Since 2006, the agency has been involved in testing truck tractors and large buses with stability control systems. To evaluate these systems, NHTSA sponsored studies of crash data in order to examine the potential safety benefits of stability control systems. NHTSA and industry representatives separately evaluated data on dynamic test maneuvers. At the same time, the agency launched a three-phase testing program to improve its understanding of how stability control systems in truck tractors and buses work and to develop dynamic test maneuvers to challenge roll propensity and yaw stability. By combining the studies of the crash data with the testing data, the agency is able to evaluate the potential effectiveness of stability control systems for truck tractors and large buses.

As a result of the data analysis research, we have tentatively determined that ESC systems can be 28 to 36 percent effective in reducing first-event untripped rollovers and 14 percent effective in eliminating loss-of-control crashes caused by severe oversteer or understeer conditions.

As a result of the agency's testing program and the test data received from industry, the agency was able to develop reliable and repeatable test maneuvers that could demonstrate a stability control system's ability to prevent rollover and loss of directional control among the varied configurations of truck tractors and buses in the fleet.

See Wang, Jing-Shiam, “Effectiveness of Stability Control Systems for Truck Tractors” (January 2011) (DOT HS 811 437); Docket No. NHTSA-2010-0034-0043.

In order to realize these benefits, the agency is proposing to require new truck tractors and certain buses with a GVWR of greater than 11,793 kilograms (26,000 pounds) to be equipped with an ESC system. This proposal is made pursuant to the authority granted to NHTSA under the National Traffic and Motor Vehicle Safety Act (“Motor Vehicle Safety Act”). Under 49 U.S.C. Chapter 301, Motor Vehicle Safety (49 U.S.C. 30101 et seq.), the Secretary of Transportation is responsible for prescribing motor vehicle safety standards that are practicable, meet the need for motor vehicle safety, and are stated in objective terms. The responsibility for promulgation of Federal motor vehicle safety standards is delegated to NHTSA.

This proposal requires ESC system must meet both definitional criteria and performance requirements. It is necessary to include definitional criteria in the proposal and require compliance with them because developing separate performance tests to cover the wide array of possible operating ranges, roadways, and environmental conditions would be impractical. The definitional criteria are consistent with those recommended by SAE International and used by the United Nations (UN) Economic Commission for Europe (ECE), and similar to the definition of ESC in FMVSS No. 126, the agency's stability control standard for light vehicles. This definition would describe an ESC system as one that would enhance the roll and yaw stability of a vehicle using a computer-controlled system that can receive inputs such as the vehicle's lateral acceleration and yaw rate, and use the information to apply brakes individually, including trailer brakes, and modulate engine torque.

The proposal requires that the system be able to detect a malfunction and provide a driver with notification of a malfunction by means of a telltale. This requirement would be similar to the malfunction detection and telltale requirements for light vehicles in FMVSS No. 126. An ESC system on/off switch is allowed for light vehicles; however, there is no provision in this proposal for allowing an ESC system to be deactivated. For truck tractors and large buses, we do not believe such controls are necessary.

After considering and evaluating several test maneuvers, the agency is proposing to use two test maneuvers for performance testing: The slowly increasing steer (SIS) maneuver and the sine with dwell (SWD) maneuver. The SIS maneuver is a characterization maneuver used to determine the relationship between a vehicle's steering wheel angle and the lateral acceleration. This test serves both to normalize the severity of the SWD maneuver and to ensure that the system has the ability to reduce engine torque. The SIS maneuver is performed by driving at a constant speed of 48 km/h (30 mph), and then increasing the steering wheel angle at a constant rate of 13.5 degrees per second until ESC system activation occurs. Using linear regression followed by extrapolation, the steering wheel angle that would produce a lateral acceleration of 0.5g is determined.

Using the steering wheel angle derived from the SIS maneuver, the agency would conduct the sine with dwell maneuver. The SWD test maneuver challenges both roll and yaw stability by subjecting the vehicle to a sinusoidal input. To conduct the SWD maneuver, the vehicle is accelerated to 72 km/h (45 mph) and then turned in a clockwise or counterclockwise direction to reach a set steering wheel angle in 0.5 seconds. The steering wheel is then turned in the opposite direction until the same steering wheel angle is reached in the opposite direction in one second. The steering wheel is then held at that steering wheel angle for one second, and then the steering wheel angle returned to zero degrees within 0.5 seconds. This maneuver would be repeated for two series of test runs (first in the counterclockwise direction and then in the clockwise direction) at several target steering wheel angles from 30 to 130 percent of the angle derived in the SIS maneuver.

The lateral acceleration, yaw rate, and engine torque data from the test runs would be measured, recorded, and processed to determine the four performance metrics: Lateral acceleration ratio (LAR), yaw rate ratio (YRR), lateral displacement, and engine torque reduction. The LAR and YRR metrics would be used to ensure that the system reduces lateral acceleration and yaw rate, respectively, after an aggressive steering input, thereby preventing rollover and loss of control, respectively. These two metrics can effectively measure what NHTSA's testing has found to be the threshold of stability. The lateral displacement metric would be used to ensure that the stability control system is not set to intervene solely by making the vehicle nonresponsive to driver input. The engine torque reduction metric would be used to ensure that the system has the capability to automatically reduce engine torque in response to high lateral acceleration and yaw rate conditions. The manner in which the data would be filtered and processed is described in this proposal.

The agency considered several test maneuvers based on its own work and that of industry. In particular, the agency's initial research focused on a ramp steer maneuver (RSM) for evaluating roll stability. In that maneuver, a vehicle is driven at a constant speed and a steering wheel input that is based on the steering wheel angle derived from the SIS maneuver is input. The steering wheel angle is then held for a period of time before it is returned to zero. A stability control system would act to reduce lateral acceleration, and thereby wheel lift and roll instability, by applying selective braking. A vehicle without a stability control system would maintain high levels of lateral acceleration and potentially experience wheel lift or rollover.

The proposed rule also sets forth the test conditions that the agency would use to ensure safety and demonstrate sufficient performance. All vehicles would be tested using outriggers for the safety of the test driver. The agency would use an automated steering controller to ensure reproducible and repeatable test execution performance. Truck tractors would be tested with an unbraked control trailer to eliminate the effect of the trailer's brakes on testing. Because the agency tests new vehicles, the brakes would be conditioned, as they are in determining compliance with the air brake standard. The agency would also test to ensure that system malfunction is detected.

This proposed rule would take effect for most truck tractors and covered buses produced two years after publication of a final rule. We believe that this amount of lead time is necessary to ensure sufficient availability of stability control systems from suppliers of these systems and to complete necessary engineering on all vehicles. For three-axle tractors with one drive axle, tractors with four or more axles, and severe service tractors, we would provide two years additional lead time. We believe this additional time is necessary to develop, test, and equip these vehicles with ESC systems. Although the agency has statutory authority to require retrofitting of in-service truck tractors, trailers, and large buses, the agency is not proposing to do so, given the integrated aspects of a stability control system.

Based on the agency's effectiveness estimates, the adoption of this proposal would prevent 1,807 to 2,329 crashes per year resulting in 649 to 858 injuries and 49 to 60 fatalities. The proposal also would result in significant monetary savings as a result of prevention of property damage and travel delays.

Based on information obtained from manufacturers, the agency estimates that 26.2 percent of truck tractors manufactured in model year 2012 will be equipped with an ESC system and that 80 percent of covered buses manufactured in model year 2012 will be equipped with an ESC system. Information obtained from manufacturers indicates that the average unit cost of an ESC system is approximately $1,160. In addition, 16.5 percent of truck tractors manufactured in model year 2012 will be equipped with an RSC system. The incremental cost of installing an ESC system in place of an RSC system is estimated to be $520 per vehicle. Based upon the agency's estimates that 150,000 truck tractors and 2,200 buses covered by this proposed rule will be manufactured in 2012, the agency estimates that the total cost of this proposal would be approximately $113.6 million.

The agency believes that this proposal is cost effective. The net benefits of this proposal are estimated to range from $228 to $310 million at a 3 percent discount rate and from $155 to $222 million at a 7 percent discount rate. As a result, the net cost per equivalent live saved from this proposal ranges from $1.5 to $2.0 million at a 3 percent discount rate and from $2.0 to $2.6 million at a 7 percent discount rate. The costs and benefits of this proposal are summarized in Table 1.

Table 1—Estimated Annual Cost, Benefits, and Net Benefits of the Proposal

[In millions of 2010 dollars]

Costs

Injury benefits

Property damage and travel delay savings

Cost per equivalent live saved

Net benefits

At 3% Discount

$113.6

$328-405

$13.9-17.8

$1.5-2.0

$228-310

At 7% Discount

113.6

257-322

11.0-14.1

2.0-2.6

155-222

The agency considered two regulatory alternatives. First, the agency considered requiring truck tractors and large buses to be equipped with RSC systems. When compared to this proposal, RSC systems would result in slightly lower cost per equivalent life saved, but would produce net benefits that are lower than the net benefits from this proposal. This is because RSC systems are less effective at preventing rollover crashes and much less effective at preventing loss-of-control crashes. The second alterative considered was requiring trailers to be equipped with RSC systems. However, this alternative would save fewer than 10 lives at a very high cost per equivalent life saved and would provide negative net benefits.

The remainder of this notice will describe in detail the following: (1) The size of the safety problem to be addressed by this proposed rule; (2) how stability control systems work to prevent rollover and loss of control; (3) the research and testing separately conducted by NHTSA and industry to evaluate the potential effectiveness of a stability control requirement and to develop dynamic test maneuvers to challenge system performance; (4) the specifics of the agency's proposal, including equipment and performance criteria, compliance testing, and the implementation schedule; and (5) the benefits and costs of this proposal.

II. Safety Problem

A. Heavy Vehicle Crash Problem

The

Traffic Safety Facts 2009

reports that tractor trailer combination vehicles are involved in about 72 percent of the fatal crashes involving large trucks, annually.

According to FMCSA's

Large Truck and Bus Crash Facts 2008,

these vehicles had a fatal crash involvement rate of 1.92 crashes per 100 million vehicle miles traveled during 2007, whereas single unit trucks had a fatal crash involvement rate of 1.26 crashes per 100 million vehicle miles traveled.

Combination vehicles represent about 25 percent of large trucks registered but travel 63 percent of the large truck miles, annually. Traffic tie-ups resulting from loss-of-control and rollover crashes also contribute to in millions of dollars of lost productivity and excess energy consumption each year.

DOT HS 811 402,

available at http://www-nrd.nhtsa.dot.gov/Pubs/811402.pdf

(last accessed May 9, 2012).

FMCSA-RRA-10-043 (Mar. 2010),

available at http://www.fmcsa.dot.gov/facts-research/ltbcf2008/index-2008largetruckandbuscrashfacts.aspx

(last accessed May 9, 2012).

According to

Traffic Safety Facts

2009, the overall crash problem for tractor trailer combination vehicles is approximately 150,000 crashes, 29,000 of which involve injury. The overall crash problem for single-unit trucks is nearly as large—approximately 146,000 crashes, 24,000 of which are injury crashes. However, the fatal crash involvement for truck tractors is much higher. In 2009, there were 2,334 fatal combination truck crashes and 881 fatal single-unit truck crashes.

The rollover crash problem for combination trucks is much greater than for single-unit trucks. In 2009, there were approximately 7,000 crashes involving combination truck rollover and 3,000 crashes involving single-unit truck rollover. As a percentage of all crashes, combination trucks are involved in rollover crashes at twice the rate of single-unit trucks. Approximately 4.4 percent of all combination truck crashes were rollovers, but 2.2 percent of single-unit truck crashes were rollovers. Combination trucks were involved in 3,000 injury crashes and 268 fatal crashes, and single-unit trucks were involved in 2,000 injury crashes and 154 fatal crashes.

According to FMCSA's

Large Truck and Bus Crash Facts 2008,

cross-country intercity buses were involved in 19 of the 247 fatal bus crashes in 2008, which represented about 0.5 percent of the fatal crashes involving large trucks and buses, annually. The bus types presented in the crash data include school buses, intercity buses, cross-country buses, transit buses, and other buses. These buses had a fatal crash involvement rate of 3.47 crashes per 100 million vehicle miles traveled during 2008. From 1998 to 2008, cross-country intercity buses, on average, accounted for 12 percent of all buses involved in fatal crashes, whereas transit buses and school buses accounted for 35 percent and 40 percent, respectively, of all buses involved in fatal crashes. Most of the transit bus and school bus crashes are not rollover or loss-of-control crashes that ESC systems are capable of preventing. The remaining 13 percent of buses involved in fatal crashes were classified as other buses or unknown. Fatal rollover and loss-of-control crashes are a subset of these crashes.

There are many more fatalities in buses with a GVWR greater than 11,793 kg (26,000 lb) compared to buses with a GVWR between 4,536 kg and 11,793 kg (10,000 lb and 26,000 lb).

In the 10-year period between 1999 and 2008, there were 34 fatalities on buses with a GVWR between 4,536 kg and 11,793 kg (10,000 lb and 26,000 lb) compared to 254 fatalities on buses with a GVWR greater than 11,793 kg (26,000 lb). Among buses with a GVWR of greater than 11,793 kg (26,000 lb), over 70 percent of the fatalities were cross-country intercity bus occupants.

This data was taken from the FARS database and was presented in the NPRM that would require seat belts on certain buses. See 75 FR 50,958, 50,917 (Aug. 18, 2010).

Furthermore, the size of the rollover crash problem for cross-country intercity buses is greater than in other buses. According to FARS data from 1999 to 2008, there were 97 occupant fatalities as a result of rollover events on cross-country intercity buses with a GVWR of greater than 11,793 kg (26,000 lb), which represents 52 percent of cross-country intercity bus fatalities.

In comparison, rollover crashes were responsible for 21 occupant fatalities on other buses with a GVWR of greater than 11,793 kg (26,000 lb) and 9 occupant fatalities on all buses with a GVWR between 4,536 kg and 11,793 kg (10,000 lb and 26,000 lb). That is, 95 percent of bus occupant rollover fatalities on buses over 4,536 kg (10,000 lb) were occupants on buses with a GVWR of over 11,793 kg (26,000 lb).

See U.S. Department of Transportation Motorcoach Safety Action Plan, DOT HS 811 177, at 13 (Nov. 2009),

available at http://www.fmcsa.dot.gov/documents/safety-security/MotorcoachSafetyActionPlan_finalreport-508.pdf

(last accessed May 9, 2012).

B. Contributing Factors in Rollover and Loss-of-Control Crashes

Many factors related to heavy vehicle operation, as well as factors related to roadway design and road surface properties, can cause heavy vehicles to become yaw unstable or to roll. Listed below are several real-world situations in which stability control systems may prevent or lessen the severity of such crashes.

•

Speed too high to negotiate a curve

—The entry speed of vehicle is too high to safely negotiate a curve. When the lateral acceleration of a vehicle during a steering maneuver exceeds the vehicle's roll or yaw stability threshold, a rollover or loss of control is initiated. Curves can present both roll and yaw instability issues to these types of vehicles due to varying heights of loads (low versus high, empty versus full) and road surface friction levels (

e.g.,

wet, dry, icy, snowy).

•

Sudden steering maneuvers to avoid a crash

—The driver makes an abrupt steering maneuver, such as a single- or double-lane-change maneuver, or attempts to perform an off-road recovery maneuver, generating a lateral acceleration that is sufficiently high to cause roll or yaw instability. Maneuvering a vehicle on off-road, unpaved surfaces such as grass or gravel may require a larger steering input (larger wheel slip angle) to achieve a given vehicle response, and this can lead to a large increase in lateral acceleration once the vehicle returns to the paved surface. This increase in lateral acceleration can cause the vehicle to exceed its roll or yaw stability threshold.

•

Loading conditions

—The vehicle yaw due to severe over-steering is more likely to occur when a vehicle is in a lightly loaded condition and has a lower center of gravity height than it would have when fully loaded. Heavy vehicle rollovers are much more likely to occur when the vehicle is in a fully loaded condition, which results in a high center of gravity for the vehicle. Cargo placed off-center in the trailer may result in the vehicle being less stable in one direction than in the other. It is also possible that improperly secured cargo can shift while the vehicle is negotiating a curve, thereby reducing roll or yaw stability. Sloshing can occur in tankers transporting liquid bulk cargoes, which is of particular concern when the tank is partially full because the vehicle may experience significantly reduced roll stability during certain maneuvers.

•

Road surface conditions

—The road surface condition can also play a role in the loss of control a vehicle experiences. On a dry, high-friction asphalt or concrete surface, a tractor trailer combination vehicle executing a severe turning maneuver is likely to experience a high lateral acceleration, which may lead to roll or yaw instability. A similar maneuver performed on a wet or slippery road surface is not as likely to experience the high lateral acceleration because of less available tire traction. Hence, the result is more likely to be vehicle yaw instability than vehicle roll instability.

•

Road design configuration

—Some drivers may misjudge the curvature of ramps and not brake sufficiently to negotiate the curve safely. This includes ramps with decreasing radius curves as well as curves and ramps with improper signage. A decrease in super-elevation (banking) at the end of a ramp where it merges with the roadway causes an increase in vehicle lateral acceleration, which may increase even more if the driver accelerates the vehicle in preparation to merge.

C. NTSB Safety Recommendations

The National Transportation Safety Board (NTSB) has issued several safety recommendations relevant to ESC systems on heavy and other vehicles. One is H-08-15, which addresses ESC systems and collision warning systems with active braking on commercial vehicles. Recommendations H-11-07 and H-11-08 specifically address stability control systems on commercial motor vehicles and buses with a GVWR above 10,000 pounds. Two other safety recommendations, H-01-06 and H-01-07, relate to adaptive cruise control and collision warning systems on commercial vehicles, and are indirectly related to ESC on heavy vehicles because all these technologies require the ability to apply brakes without driver input.

• H-08-15: Determine whether equipping commercial vehicles with collision warning systems with active braking

and electronic stability control systems will reduce commercial vehicle accidents. If these technologies are determined to be effective in reducing accidents, require their use on commercial vehicles.

Active braking involves using the vehicle's brakes to maintain a certain, preset distance between vehicles.

• H-11-07: Develop stability control system performance standards for all commercial motor vehicles and buses with a gross vehicle weight rating greater than 10,000 pounds, regardless of whether the vehicles are equipped with a hydraulic or pneumatic brake system.

• H-11-08: Once the performance standards from Safety Recommendation H-11-07 have been developed, require the installation of stability control systems on all newly manufactured commercial vehicles with a GVWR greater than 10,000 pounds.

D. Motorcoach Safety Plan

In November 2009, the U.S. Department of Transportation Motorcoach Safety Action Plan was issued.

Among other things, the Motorcoach Safety Action Plan includes an action item for NHTSA to assess the safety benefits for stability control on large buses and develop objective performance standards for these systems.

Consistent with that plan, NHTSA made a decision to pursue a stability control requirement for large buses.

See

supra,

note 6.

Id.

at 28-29.

In March 2011, NHTSA issued its latest Vehicle Safety and Fuel Economy Rulemaking and Research Priority Plan (Priority Plan).

The Priority Plan describes the agency plans for rulemaking and research for calendar years 2011 to 2013. The Priority Plan includes stability control on truck tractors and large buses, and states that the agency plans to develop test procedures for a Federal motor vehicle safety standard on stability control for truck tractors, with the countermeasures of roll stability control and electronic stability control, which are aimed at addressing rollover and loss-of-control crashes.

See Docket No. NHTSA-2009-0108-0032.

E. International Regulation

The United Nations (UN) Economic Commission for Europe (ECE) Regulation 13, Uniform Provisions Concerning the Approval of Vehicles of Categories M, N and O with Regard to

Braking,

has been amended to include Annex 21,

Special Requirements for Vehicles Equipped with a Vehicle Stability Function.

Annex 21's requirements apply to trucks with a GVWR greater than 3,500 kg (7,716 lb), buses with a seating capacity of 10 or more (including the driver), and trailers with a GVWR greater than 3,500 kg (7,716 lb). Trucks and buses are required to be equipped with a stability system that includes rollover control and directional control, while trailers are required to have a stability system that includes only rollover control. The directional control function must be demonstrated in one of eight tests, and the rollover control function must be demonstrated in one of two tests. For

compliance purposes, the ECE regulation requires a road test to be performed with the function enabled and disabled, or as an alternative accepts results from a computer simulation. No test procedure or pass/fail criterion is included in the regulation, but it is left to the discretion of the Type Approval Testing Authority in agreement with the vehicle manufacturer to show that the system is functional. The implementation date of Annex 21 is 2012 for most vehicles, with a phase-in based on the vehicle type.

III. Stability Control Technologies

A. Dynamics of a Rollover

Whenever a vehicle is steered, the lateral forces that result from the steering input lead to one of the following results: (1) Vehicle maintains directional control; (2) vehicle loses directional control due to severe understeer or plowing out; (3) vehicle loses directional control due to severe oversteer or spinning out; or (4) vehicle experiences roll instability and rolls over.

A turning maneuver initiated by the driver's steering input results in a vehicle response that can be broken down into two phases. Phase 1 is the yaw response that occurs when the front wheels are turned. As the steering wheel is turned, the displacement of the front wheels generates a slip angle at the front wheels and a lateral force is generated. That lateral force leads to vehicle rotation, and the vehicle starts rotating about its center of gravity.

This rotation leads to Phase 2. In Phase 2, the vehicle's yaw causes the rear wheels to experience a slip angle. That causes a lateral force to be generated at the rear tires, which leads to vehicle rotation. All of these actions establish a steady-state turn in which lateral acceleration and yaw rate are constant.

In combination vehicles, which typically consist of a tractor towing a semi-trailer, an additional phase is the turning response of the trailer. Once the tractor begins to achieve a yaw and lateral acceleration response, the trailer begins to yaw as well. This leads to the trailer's tires developing slip angles and producing lateral forces at the trailer tires. Thus, there is a slight delay in the turning response of the trailer when compared to the turning response of the tractor.

If the lateral forces generated at either the front or the rear wheels exceed the friction limits between the road surface and the tires, the result will be a vehicle loss-of-control in the form of severe understeer (loss of traction at the steer tires) or severe oversteer (loss of traction at the rear tires). In a combination vehicle, a loss of traction at the trailer wheels would result in the trailer swinging out of its intended path. However, if the lateral forces generated at the tires result in a vehicle lateral acceleration that exceeds the rollover threshold of the vehicle, then rollover will result.

Lateral acceleration is the primary cause of rollovers. Figure 1 depicts a simplified rollover condition. As shown, when the lateral force (

i.e.,

lateral acceleration) is sufficient large and exceeds the roll stability threshold of the tractor-trailer combination vehicle, the vehicle will roll over. Many factors related to the drivers' maneuvers, heavy vehicle loading conditions, vehicle handling characteristics, roadway design, and road surface properties would result in various lateral accelerations and influences on the rollover propensity of a vehicle. For example, given other factors are equal, a vehicle entering a curve at a higher speed is more likely to roll than a vehicle entering the curve at a lower speed. Also, transporting a high center of gravity (CG) load would increase the rollover probability more than transporting a relatively lower CG load.

EP23MY12.002

Stability control technologies help a driver maintain directional control and help to reduce roll instability. Two types of heavy vehicle stability control technologies have been developed. One such technology is roll stability control or RSC, which is designed to help prevent on-road, untripped rollovers by automatically decelerating the vehicle using brakes and engine control. The other technology is electronic stability control, or ESC,

which is designed to

assist the driver in mitigating severe oversteer or understeer conditions by automatically applying selective brakes to help the driver maintain directional control of the vehicle. On heavy vehicles, ESC also includes the RSC function described above.

In light vehicles, the term ESC generally describes a system that helps the driver maintain directional control and typically does not include the RSC function because these vehicles are much less prone to untripped rollover.

B. Description of RSC System Functions

Currently, RSC systems are available for air-braked tractors with a GVWR of greater than 11,793 kilograms (26,000 pounds) and for trailers. A tractor-based RSC system consists of an electronic control unit (ECU) that is mounted on a vehicle and continually monitors the vehicle's speed and lateral acceleration based on an accelerometer, and estimates vehicle mass based on engine torque information.

The ECU continuously estimates the roll stability threshold of the vehicle, which is the lateral acceleration above which a combination vehicle will roll over. When the vehicle's lateral acceleration approaches the roll stability threshold, the RSC system intervenes. Depending on how quickly the vehicle is approaching the estimated rollover threshold, the RSC system intervenes by one or more of the following actions: Decreasing engine power, using engine braking, applying the tractor's drive-axle brakes, or applying the trailer's brakes. When RSC systems apply the trailer's brakes, they use a pulse modulation protocol to prevent wheel lockup because tractor stability control systems cannot currently detect whether or not the trailer is equipped with ABS. Some RSC systems also use a steering wheel angle sensor, which allows the system to identify potential roll instability events earlier.

RSC systems are not presently available for large buses.

An RSC system can reduce rollovers, but is not designed to help to maintain directional control of a truck tractor. Nevertheless, RSC systems may provide some additional ability to maintain directional control in some scenarios, such as in a low-center-of-gravity scenario, where an increase in a lateral acceleration may lead to yaw instability rather than roll instability.

In comparison, a trailer-based RSC system has an ECU mounted on the trailer, which typically monitors the trailer's wheel speeds, the trailer's suspension to estimate the trailer's loading condition, and the trailer's lateral acceleration. When a high lateral acceleration that is likely to cause the trailer to rollover is detected, the ECU commands application of the trailer brakes to slow the combination vehicle. In this case, the trailer brakes on the outside wheels can be applied with full pressure since the ECU can directly monitor the trailer wheels for braking-related lockup. The system modulates the brake pressure as needed to achieve maximum braking force without locking the wheels. However, a trailer-based RSC system can only apply the trailer brakes to slow a combination vehicle, whereas a tractor-based RSC system can apply brakes on both the tractor and trailer.

C. Description of ESC System Functions

Currently, ESC systems are available for heavy vehicles, including truck tractors and buses, equipped with air brakes. An ESC system incorporates all of the inputs of an RSC system. In addition, an ESC system monitors steering wheel angle and yaw rate of the vehicle.

These system inputs are monitored by the system's ECU, which estimates when the vehicle's directional response begins to deviate from the driver's steering command, either by oversteer or understeer. An ESC system intervenes to restore directional control by taking one or more of the following actions: Decreasing engine power, using engine braking, selectively applying the brakes on the truck tractor to create a counter-yaw moment to turn the vehicle back to its steered direction, or applying the brakes on the trailer. An ESC system enhances the RSC functions because it has the added information from the steering wheel angle and yaw rate sensors, as well as more braking power because of its additional capability to apply the tractor's steer axle brakes.

Because ESC systems must monitor steering inputs from the tractor, ESC systems are not available for trailers.

This is a design strategy to avoid the unintended consequences of applying the brakes on the steering axle without knowing where the driver is steering the vehicle.

D. How ESC Prevents Loss of Control

Like an RSC system, an ESC system has a lateral acceleration sensor. However, it also has two additional sensors to monitor a vehicle for loss of directional control, which may result due to either understeer or oversteer. The first additional sensor is a steering wheel angle sensor, which senses the intended direction of a vehicle. The other is a yaw rate sensor, which measures the actual turning movement of the vehicle. When a discrepancy between the intended and actual headings of the vehicle occurs, it is because the vehicle is in either an understeering (plowing out) or an oversteering (spinning out) condition. The ESC system responds to such a discrepancy by automatically intervening and applying brake torque selectively at individual wheel ends on the tractor, by reducing engine torque output to the drive axle wheels, or by both means. If only the wheel ends at one corner of the vehicle are braked, the uneven brake force will create a correcting yaw moment that causes the vehicle's heading to change. An ESC system also has the capability to reduce the engine torque output to the drive wheels, which effectively reduces the vehicle speed and helps the wheels to regain traction. This means of intervention by the ESC system may occur separate from or simultaneous with the automatic brake application at selective wheel ends. An ESC system is further differentiated from an RSC system in that it has the ability to selectively apply the front steer axle brakes while the RSC system does not incorporate this feature.

Figure 2 illustrates the oversteering and understeering conditions. While Figure 2 may suggest that a particular vehicle loses control due to either oversteer or understeer, it is quite possible that a vehicle could require both understeering and oversteering interventions during progressive phases of a complex crash avoidance maneuver such as a double lane change.

EP23MY12.003

Oversteering.

The right side of Figure 2 shows that the truck tractor whose driver has lost directional control during an attempt to drive around a right curve. The rear wheels of the tractor have exceeded the limits of road traction. As a result, the rear of the tractor is beginning to slide. This would lead a vehicle without an ESC system to spin out. If the tractor is towing a trailer, as the tractor in the figure is, this would result in a jackknife crash. In such a crash, the tractor spins and may make physical contact with the side of the trailer. The oversteering tractor in this figure is considered to be yaw-unstable because the tractor rotation occurs without a corresponding increase in steering wheel angle by the driver. In a vehicle equipped with ESC, the system immediately detects that the vehicle's heading is changing more quickly than appropriate for the driver's intended path (

i.e.,

the yaw rate is too high). To counter the leftward rotation of the vehicle, it momentarily applies the right front brake, thus creating a rightward (clockwise) counter-rotational force and turning the heading of the vehicle back to the correct path. It will also cut engine power to gently slow the vehicle and, if necessary, apply additional brakes (while maintaining the uneven brake force to create the necessary yaw moment). The action happens quickly so that the driver does not perceive the need for steering corrections.

Understeering.

The left side of Figure 2 shows a truck tractor whose driver has lost directional control during an attempt to drive around a right curve, except that in this case, it is the front wheels that have exceeded the limits of road traction. As a result, the tractor is sliding at the front (“plowing out”). Such a vehicle is considered to be yaw-stable because no increase in tractor rotation occurs when the driver increases the steering wheel angle. However, the driver has lost directional control of the tractor. In this situation, the ESC system rapidly detects that the vehicle's heading is changing less quickly than appropriate for the driver's intended path (

i.e.,

the yaw rate is too low). In other words, the vehicle is not turning right sufficiently to remain on the right curve and is instead heading off to the left. The ESC system momentarily applies the right rear brake, creating a rightward rotational force, to turn the heading of the vehicle back to the correct path. Again, it will also cut engine power to gently slow the vehicle and, if necessary, apply additional brakes (while maintaining the uneven brake force to create the necessary yaw moment).

E. Situations in Which Stability Control Systems May Not Be Effective

A stability control system will not prevent all rollover and loss-of-control crashes. A stability control system has the capability to prevent many untripped on-road rollovers and first-event loss-of-control events. Nevertheless, there are real-world situations in which stability control systems may not be as effective in avoiding a potential crash. Such situations include:

• Off-road recovery maneuvers in which a vehicle departs the roadway and encounters an incline too steep to effectively maneuver the vehicle or an unpaved surface that significantly reduces the predictability of the vehicle's handling

• Entry speeds that are much too high for a curved roadway or entrance/exit ramp

• Cargo load shifts on the trailer during a steering maneuver

• Vehicle tripped by a curb or other roadside object or barrier

• Truck rollovers that are the result of collisions with other motor vehicles

• Inoperative antilock braking systems—the performance of stability control systems depends on the proper functioning of ABS

• Brakes that are out-of-adjustment or other defects or malfunctions in the ESC, RSC, or brake system.

• Maneuvers during tire tread separation or sudden tire deflation events.

F. Difference in Vehicle Dynamics Between Light Vehicles and Heavy Vehicles

On April 6, 2007, the agency published a final rule that established FMVSS No. 126,

Electronic Stability Control Systems,

which requires all passenger cars, multipurpose passenger vehicles, trucks and buses with a GVWR of 4,536 kg (10,000 lb) or less to be equipped with an electronic stability control system beginning in model year 2012.

The rule also requires a phase-in of 55 percent, 75 percent, and 95 percent of vehicles produced by each manufacturer during model years 2009, 2010, and 2011, respectively, to be equipped with a compliant ESC system. The system must be capable of applying brake torques individually at all four wheels, and must comply with the performance criteria established for stability and responsiveness when subjected to the sine with dwell steering maneuver test.

72 FR 17236.

For light vehicles, the focus of the FMVSS No. 126 is on addressing yaw instability, which can assist the driver in preventing the vehicle from leaving the roadway, thereby preventing fatalities and injuries associated with crashes involving tripped rollover, which often occur when light vehicles run off the road. The standard does not include any equipment or performance requirements for roll stability.

The dynamics of light vehicles and heavy vehicles differ in many respects. First, on light vehicles, the yaw stability threshold is typically lower than the roll stability threshold. This means that a light vehicle making a crash avoidance maneuver, such as a lane change on a dry road, is more likely to reach its yaw stability threshold and lose directional control before it reaches its roll stability threshold and rolls over. On a heavy

vehicle, however, the roll stability threshold is lower than the yaw stability threshold in most operating conditions, primarily because of its higher center of gravity height.

As a result, there is a greater propensity for a heavy vehicle, particularly in a loaded condition, to roll during a severe crash avoidance maneuver or when negotiating a curve, than to become yaw unstable, as compared with light vehicles.

One instance where a heavy vehicle's yaw stability threshold might be higher than its roll stability threshold is in an unloaded condition on a low-friction road surface.

Second, a tractor-trailer combination unit is comprised of a power unit and one or more trailing units with one or more articulation points. In contrast, although a light vehicle may occasionally tow a trailer, a light vehicle is usually a single rigid unit. The tractor and the trailer have different center of gravity heights and different lateral acceleration threshold limits for rollover. A combination vehicle rollover frequently begins with the trailer where the rollover is initiated by trailer wheel lift. The trailer roll torque is transmitted to the tractor through the vehicles' articulation point, which subsequently leads to tractor rollover. In addition to the trailer's loading condition, the trailer rollover threshold is also related to the torsional stiffness of the trailer body. A trailer with a low torsional stiffness, such as a flatbed open trailer, would typically experience wheel lift earlier during a severe turning maneuver than a trailer with a high torsional stiffness, such as a van trailer. Hence, compared with a light vehicle, the roll dynamics of a tractor trailer combination vehicle is a more complex interaction of forces acting on the units in the combination, as influenced by the maneuver, the loading condition, and the roadway.

Unlike with light vehicles, there is a large range of loading scenarios possible for a given heavy vehicle, particularly for truck tractors towing trailers. A tractor-trailer combination vehicle can be operated empty, loaded to its maximum weight rating, or loaded anywhere in between the two extremes. The weight of a fully loaded combination vehicle is generally more than double that of the vehicle with an empty trailer. Furthermore, the load's center of gravity height can vary over a large range, which can have substantial effects on the dynamics of a combination vehicle.

Third, due to greater length, mass, and mass moments of inertia of heavy vehicles, they respond more slowly to steering inputs than do light vehicles. The longer wheelbase of a heavy vehicle, compared with a light vehicle, results in a slower response time, which gives the stability control system the opportunity to intervene and prevent rollovers.

Finally, the larger number of wheels on a heavy vehicle, as compared to a light vehicle, results in making heavy vehicles less likely to yaw on dry road surface conditions.

As a result of the differences in vehicle dynamics between light vehicles and heavy vehicles, the requirements in FMVSS No. 126 for light vehicle ESC systems cannot translate directly into requirements for heavy vehicles. Nevertheless, many requirements in FMVSS No. 126 are pertinent to heavy vehicles because they do not relate to any difference in vehicle dynamics between light vehicles and heavy vehicles. For example, the ESC system malfunction detection and telltale requirements already developed for light vehicles can be translated to heavy vehicles.

IV. Research and Testing

NHTSA has been studying ways to prevent untripped heavy vehicle rollovers for many years. In the mid-1990s, the agency sponsored the development of a prototype roll stability advisor (RSA) system that displayed information to the driver regarding the truck's roll stability threshold and the peak lateral acceleration achieved during cornering maneuvers. This was followed by a fleet operational test sponsored by the Federal Highway Administration, under the Department of Transportation's Intelligent Vehicle Initiative. The tractors were equipped with a RSA system using an engine retarder, which was an early configuration of an RSC system. As that test program was concluding, industry developers of stability control systems began to add tractor and trailer foundation braking capabilities to increase the effectiveness these systems.

In 2006, the agency initiated a test program at the Vehicle Research and Test Center (VRTC) to conduct track testing on RSC- and ESC-equipped tractors and semitrailers. The initial testing focused only on roll stability testing and provided comparative data on the performance of the different stability control systems in several test maneuvers. Subsequent testing focused on refining test maneuvers and developing performance metrics suitable for a safety standard. The agency studied a slowly increasing steer maneuver that would characterize a tractor's steering system and verify the ability of a tractor-based system to control engine torque. The agency also developed a ramp steer maneuver to evaluate the roll stability performance of a stability control system, and investigated a sine with dwell maneuver to evaluate both yaw and roll stability performance. In addition to tests conducted on combination unit trucks, the VRTC research program included testing of three large buses equipped with ESC using these test maneuvers. As part of the research at VRTC, the agency also developed data collection and analysis methods to characterize the performance of stability control systems.

NHTSA researchers began updating their vehicle dynamics simulation programs to include a stability control model, and coordinated with researchers at the National Advanced Driving Simulator (NADS) at the University of Iowa to add stability control modeling capability to their tractor trailer simulations. NHTSA sponsored a research program with the NADS to evaluate potential RSC and ESC effectiveness in several tractor-trailer driving scenarios involving potential rollover and loss of control, using sixty professional truck drivers who were recruited as test participants.

NHTSA purchased three tractors equipped with ESC or RSC systems for testing: A Freightliner 6x4

tractor that had ESC as a production option, a Sterling 4x2 tractor that had RSC as a production option, and a Volvo 6x4 tractor that had ESC included as standard equipment. NHTSA also obtained a RSC control unit that could be retrofitted on the Freightliner 6x4 tractor so that it could be comparatively tested with both ESC and RSC. The agency also purchased a Heil 9,200-gallon tanker semitrailer that was equipped with a trailer-based RSC system, and retrofitted a Fruehauf 53-foot van semitrailer with a trailer-based RSC system. NHTSA also obtained three large buses equipped with stability control systems: A 2007 MCI D4500 (MCI #1), a 2009 Prevost H3, and a second 2007 MCI D4500 (MCI #2). The MCI buses were equipped with a Meritor WABCO ESC system and the Prevost was equipped with a Bendix ESC system.

The 6x4 description for a tractor represents the total number of wheel positions (six) and the total number of wheel positions that are driven (four), which means that the vehicle has three axles with two of them being drive axles. Similarly, a 4x2 tractor has four wheel positions, two of which are driven, meaning that the vehicle has two axles, one of which is a drive axle.

Although the manufacturers of truck tractors and large buses and the suppliers of stability control systems have performed extensive development

work to bring these systems to the market, there are few sources of objective evaluations for testing on stability control systems in the public domain beyond the research programs described above. The agency coordinated with truck, bus, and stability control system manufacturers throughout the VRTC test program so that industry organizations had the opportunity to contribute additional test data and other relevant information on test maneuvers that the agency could consider for use during the research program. Potential maneuvers suggested by industry included a decreasing radius test from the Truck & Engine Manufacturers Association (EMA),

a sinusoidal steering maneuver and a ramp with dwell maneuver from Bendix, and a lane change maneuver (on a large diameter circle) from Volvo.

In late 2009, the EMA provided results from their tests of the ramp steer, sine with dwell, and ramp with dwell maneuvers to NHTSA. The agency evaluated these data from a measures-of-performance perspective. EMA provided data in December 2010 discussing additional testing with the sine with dwell, J-turn, and a wet-Jennite drive through maneuver. Additional details on these research programs are included in the sections below.

EMA was formerly known as the Truck Manufacturers Association (TMA). Many docket materials refer to EMA as TMA.

Presentations from briefings NHTSA had with EMA have been included in the docket. See Docket Nos. NHTSA-2010-0034-0025 through NHTSA-2010-0034-0031; Docket Nos. NHTSA-2010-0034-0041 and NHTSA-2010-0034-0042. Research notes provided by EMA, Bendix, and Volvo Trucks have also been included in the docket. See Docket Nos. NHTSA-2010-0034-0032 through NHTSA-2010-0034-0040.

A. UMTRI Study

NHTSA sponsored a research program with Meritor WABCO and the University of Michigan Transportation Research Institute (UMTRI) to examine the potential safety effectiveness of stability control systems for five-axle tractor-trailer combination vehicles. The systems investigated included both RSC and ESC.

The research results are provided in the report “Safety Benefits of Stability Control Systems for Tractor-Semitrailers.” A copy of this report has been included in the docket.

A similar study has been initiated with respect to straight trucks over 10,000 pounds GVWR.

DOT HS 811 205 (Oct. 2009), Docket No. NHTSA-2010-0034-0006

The objectives of the study were: (1) To use the Large Truck Crash Causation Study (LTCCS) to define typical pre-crash scenarios and identify factors associated with loss-of-control and rollover crashes for tractor-trailers; (2) to study the effectiveness of RSC and ESC in a range of realistic scenarios through hardware-in-the-loop simulation testing, and through case reviews by a panel of experts; (3) to apply the results of this research to generate national estimates from the Trucks Involved in Fatal Accidents (TIFA) and General Estimates System (GES) crash databases of the safety benefits of RSC and ESC in preventing tractor-trailer crashes; and (4) to review crash data from 2001 through 2007 from a large trucking fleet that had started purchasing RSC on all of its new tractors starting in 2004, to determine if there was an influence of this system on reducing crashes.

The LTCCS was a joint study undertaken by the Federal Motor Carrier Safety Administration (FMCSA) and NHTSA, based on a sample of 963 crashes between April 2001 and December 2003 with a reported injury or fatality involving 1,123 trucks with a GVWR over 10,000 pounds. The LTCCS crash data formed the backbone for this study because of the high quality and consistent detail contained in the case files. Included in the LTCCS are categorical data, comprehensive narrative descriptions of each crash, scene diagrams, and photographs of the vehicle and roadway from various angles. This information allowed the researchers to achieve a high level of understanding of the crash mechanics for particular cases. The LTCCS was used to help develop the crash scenarios for modeling (hardware-in-the-loop) performed as part of the engineering analyses for this stability control project. In addition, LTCCS cases of interest with respect to stability control systems were also reviewed by a panel of three experts (two from UMTRI and one from industry) to help estimate the safety benefits of RSC and ESC.

One method for assessing the safety benefits of vehicle technologies is to analyze crash datasets containing data on the safety performance of vehicles equipped with the subject technology. However, because the deployment of the stability control technologies for large trucks is still in its early stages, national crash databases do not yet have sufficient cases that can be used to evaluate the safety performance of stability control technology. Given this limitation, this study used an indirect method to estimate the safety performance of stability control technologies based on probable outcome estimates derived from hardware-in-the-loop simulation, field test experience, expert panel assessment, and crash data from trucking fleets.

UMTRI's study made several conclusions. First, identifying relevant loss-of-control and rollover crashes within the national databases proved a difficult task because the databases are developed for general use and this project required very precise definitions of loss-of-control and rollover (

e.g.,

tripped versus untripped). Relying on the general loss-of-control or rollover categories captures a wide range of crashes, many of which cannot be prevented by the stability control technology. Furthermore, many of the crashes involved vehicles that were not equipped with ABS. Because ABS is now mandatory for the target population of vehicles, the researchers had to factor in what effect the presence of ABS on the vehicle may have reduced the likelihood of or prevented the crash.

Second, the LTCCS was highly valuable in providing a greater level of detail concerning rollover and loss-of-control crashes, which was used to construct a number of relevant crash scenarios so that the technical potential of the candidate RSC and ESC technologies could be estimated systematically. However, the inability to determine with confidence if a vehicle lost control and the lack of detailed information on driver input and vehicle state placed limitations on the ability to assess the potential for stability control technologies to alter the outcome of a particular crash scenario. In contrast, for rollover crashes, it was clear that rollover occurred. Tire marks and road alignment provide strong evidence of the vehicle path and the point of instability.

Third, UMTRI concluded that ESC systems would provide more overall safety benefits than RSC systems. The difference between the estimated effectiveness of RSC and ESC varied among crash scenarios. ESC systems were slightly more effective at preventing rollovers than RSC systems and much more effective at preventing loss-of-control crashes.

Finally, the safety benefits estimates derived from this study were limited to five-axle tractor-trailer combination vehicles, which constitute a majority of the national tractor fleet. However, the study did not include benefits estimates for multi-trailer combinations or for tractors not towing a trailer.

B. Simulator Study

NHTSA sponsored a research study with the University of Iowa to study the effectiveness of heavy truck electronic stability control systems in reducing jackknife and rollover incidents using the NADS-1 National Advanced Driving Simulator. The NADS-1 is a high-fidelity, full motion driving simulator with a 360-degree visual display system

that is typically used for the study of driver behavior. Sixty professional truck drivers were recruited to participate in the study. The participants drove a typical tractor-semitrailer in five scenarios designed to have a high potential for rollover or jackknife. The study used the NADS heavy truck cab and vehicle dynamics model to simulate a typical 6x4 tractor-trailer combination vehicle in a baseline (ABS-only), RSC-equipped, and ESC-equipped configurations, using twenty truck drivers per configuration. The purpose of the study was to determine the effectiveness of both roll stability control and yaw stability control systems, to demonstrate driver behavior while using stability control systems, and to help NHTSA refine safety benefits estimates for heavy truck stability technologies.

The final report is available in the docket. “Heavy Truck ESC Effectiveness Study Using NADS” (DOT HS 811 233, November 2009), Docket No. NHTSA-2010-0034-0007.

The NADS truck model performance was compared with test track data from VRTC. The test maneuver used was a ramp steer maneuver with a steering wheel angle of 190 degrees and an angular steering rate of 175 degrees per second. The steering angle was held constant for five seconds after reaching 190 degrees, and then returned to zero. Steering inputs on the NADS were performed manually rather than by using an automated steering machine. The RSM was performed in the NADS to both the right and left directions to check for any simulation abnormalities, and was performed for the baseline, RSC, and ESC test conditions. Exact matching of values to the test track data was not possible because the NADS model was developed by simulating the braking properties of a Freightliner tractor while using the inertial properties of a Volvo tractor. Also, the NADS was modeled with rigid body tractor and trailer vehicle models that did not include the torsional chassis compliance that is a variable in actual vehicles. The result of the testing was that the NADS model tractor-semitrailer experienced wheel lift at slightly lower speeds in the RSM in all three conditions (baseline, RSC, and ESC) than in the VRTC track tests. An additional comparison of VRTC track test data and the NADS ESC model was performed for lane change maneuvers at 45 and 50 mph and showed that the NADS ESC system responses closely matched the responses of the actual test vehicle.

The maneuvering events used to assess the influence of ESC systems consisted of lane incursion from the left side on a snow-covered road and from the right side on a dry road surface, with each event necessitating a sudden lane change to avoid collision. These events provided a greater challenge for the stability control systems due to the aggressive steering and braking inputs by the drivers. Neither stability control system showed benefits in preventing rollover on the dry road surface. ESC systems did provide improved vehicle control on the snow-covered surface; however, two jackknife events still occurred with the ESC system. A large number of jackknife events occurred on the snow-covered surface with the RSC system (11 loss-of-control events in 20 runs) which may have been a result of the aggressive RSC braking strategy found in the model interfering with the driver's ability to maintain steering control of the tractor.

The NADS research study indicated that the RSC system showed a statistically significant benefit in preventing rollovers on both curves and exit ramps on dry, high-friction road surfaces. The tractors equipped with RSC and ESC systems showed a benefit over the baseline tractor in assisting drivers to avoid a jackknife on a low-friction road surface and a rollover on a high-friction road surface when encountering a directional change due roadway geometry. However, in several instances the ESC system was found to activate at abnormally high levels of lateral acceleration in a curve with a high-friction road surface. Although the reason for this was not determined, there may have been problems with the mass estimation algorithm or vehicle parameter inaccuracies in the model.

C. NHTSA Track Testing

NHTSA researchers at VRTC in East Liberty, Ohio, initiated a test program in 2006 to evaluate the performance of stability control systems under controlled conditions on a test track, and to develop objective test procedures and measures of performance that could form the basis of a new FMVSS. Researchers tested three truck tractors, all of which were equipped with an RSC or ESC system (one vehicle was tested with both an RSC and ESC system), one trailer equipped with a trailer-based RSC system, and three large buses equipped with an ESC system. Additionally, the agency tested five baseline semi-trailers not equipped with a stability control system, including an unbraked control trailer that is used to conduct tractor braking tests as prescribed by FMVSS No. 121,

Air brake systems.

The testing was conducted in three phases. Phase I research focused on understanding how stability control systems performed. Phase II research focused on the development of a dynamic test maneuver to evaluate the roll stability of tractor semitrailers and large buses. Phase III research focused on the development of a dynamic test maneuver to evaluate the yaw stability of truck tractors and large buses.

The Phase I and II research results are documented in the report “Tractor Semi-Trailer Stability Objective Performance Test Research—Roll Stability.”

The Phase III research results for truck tractors are documented in the report “Tractor Semitrailer Stability Objective Performance Test Research—Yaw Stability.”

The information provided in sections IV.C.1, IV.C.2, and IV.C.3 below is based on these two reports. The motorcoach research is documented in the report “Test Track Lateral Stability Performance of Motorcoaches Equipped with Electronic Stability Control Systems.”

The information in section IV.C.4 is based on this report.

DOT HS 811 467 (May 2011), Docket No. NHTSA-2010-0034-0009. Results from Phase I are also summarized in the paper “NHTSA's Class 8 Truck-Tractor Stability Control Test Track Effectiveness” (ESV 2009. Paper No. 09-0552). Docket No. NHTSA-2010-0034-0008.

Docket No. NHTSA-2010-0034-0046.

Docket No. NHTSA-2010-0034-0045.

1. Effects of Stability Control Systems—Phase I

The test vehicles used in Phase I included a 2006 Freightliner 6x4 tractor equipped with air disc brakes and a Meritor WABCO ESC system as factory-installed options, a 2006 Volvo 6x4 tractor with S-cam drum brakes and a Bendix ESC system included as standard equipment, and a 2000 Fruehauf 53-foot van trailer that was retrofitted with a Meritor WABCO trailer-based RSC system. Tests were conducted by enabling and disabling the stability control systems on the tractor and the trailer to compare the individual performance of each system, evaluate the performance of the combined tractor and trailer stability control systems, establish the baseline performance of each tractor-trailer combination without any stability control system. All tests were conducted with the tractor connected to the trailer, in either the unloaded condition (lightly loaded vehicle weight (LLVW)) or loaded to a 80,000 pound combination weight with the ballast located to produce either a low or high center of gravity height (low CG or high CG) loading condition. During testing, all

combination vehicles were equipped with outriggers.

The first test maneuver evaluated in Phase I was a constant radius circle test (either a 150 foot or a 200 foot radius) conducted on dry pavement. In this constant radius circle test, the driver maintained the vehicle on the curved path while slowly increasing the vehicle speed until the stability control system activated, wheel lift occurred, or the tractor experienced a severe understeer condition.

With the stability control systems disabled, no cases of wheel lift were observed under the LLVW or low CG condition. Under these load conditions, both tractors went into a severe understeer condition. The LLVW tractor did not reach a velocity greater than 40 mph and the low CG tractor did not reach a velocity greater than 34 mph. However, in the high CG condition with the tractor ESC systems disabled, wheel lift occurred in every test that resulted in a lateral acceleration greater than 0.45g at 30 mph.

With the tractor ESC systems enabled, the performance of the two ESC-equipped vehicles improved during the constant radius tests. Both ESC systems limited the maximum lateral acceleration of the tractor by reducing the engine output torque and prevented wheel lift and severe tractor understeer with the different loads tested. With ESC systems enabled, both tractors tested allowed higher maximum lateral accelerations for the LLVW condition compared to the low CG and high CG conditions. There was little difference in peak lateral acceleration for the low CG and high CG conditions.

The trailer-based RSC system limited the maximum lateral acceleration by applying the trailer brakes, which mitigated wheel lift and understeer with the different loads tested. The maximum lateral acceleration of both tractors was limited by the trailer RSC system to below 0.50g for the LLVW condition, 0.40g to 0.50g for the low CG condition, and 0.35g to 0.40g for the high CG condition.

When both tractor- and trailer-based stability control systems were enabled, results were similar to the results of the tractor-based stability control system for the low CG and high CG conditions. Under the LLVW condition, results were similar to the trailer-based RSC system values observed.

The second maneuver evaluated in Phase I was a J-turn, also conducted on dry pavement, in which the test driver accelerated the vehicle to a constant speed in a straight lane and then negotiated 180 degrees of arc along a 150-foot radius curve. The initial maneuver entrance speed was 20 mph and it was incrementally increased in subsequent runs, until a test termination condition was reached. The test terminated upon the occurrence of one of the following: The trailer outriggers making contact with the ground, indicating that wheel lift was occurring; the tractor experiencing a severe understeer condition; a stability control system brake activating; or the maneuver entry speed reaching 50 mph.

For both tractors in the baseline configuration (stability control disabled), trailer wheel lift occurred in all load combinations except for the Freightliner in the LLVW condition, which went into a severe understeer condition at a maneuver entry speed of 50 mph. For the Volvo in the LLVW load condition, trailer wheel lift was observed when the tractor's maximum lateral acceleration exceeded 0.75g at 48 mph. With stability control disabled in the low CG load condition, trailer wheel lift was observed when the tractor's maximum lateral acceleration was greater than 0.67g at 40 mph for the Freightliner and 0.60g at 38 mph for the Volvo. For the high CG load condition, trailer wheel lift was observed when the tractor's maximum lateral acceleration was approximately 0.45g at 33 mph for the Freightliner and 0.42g at 31 mph for the Volvo.

Tractor ESC systems limited the maximum lateral acceleration for both the tractor and the trailer. Wheel lift was not observed for the range of speeds evaluated. For both tractors tested in the low CG and high CG loading conditions, the tractor's ESC intervened at a speed that was well below the speed that would produce trailer wheel lift. With the trailer in the LLVW load condition, the tractor's maximum lateral acceleration was limited to approximately 0.60g for the Freightliner and the Volvo. With the trailer tested in either the low CG or high CG load conditions, the tractor's lateral acceleration was limited to 0.50g and 0.40g for the Freightliner and Volvo respectively.

The trailer-based RSC system also improved the baseline vehicle's roll stability in the J-turn maneuver. For the LLVW load condition, the trailer-based RSC system activated at speeds similar to those of the tractor-based systems. For the low CG and high CG load conditions, the tractor-based systems activated at approximately a 3 mph lower speed than the trailer-based RSC system. With both systems enabled, the tractor-based system activated and mitigated the roll propensity before the trailer RSC system activated.

The third maneuver evaluated in Phase I was a double-lane-change maneuver, in which the test driver accelerated the vehicle up to a constant speed on a dry road surface and then negotiated a lane change maneuver followed by a return to the original lane within physical boundaries (gates) marked by cones. The maneuver entry speed was incrementally increased in subsequent test runs. Although the top speed in this maneuver was intended to be limited to 50 mph for safety reasons, the test driver performed runs at speeds as high as 51 mph.

In the baseline configuration, both tractors completed the maneuver at 50 mph without wheel lift or yaw instability in the LLVW and the low CG loading conditions. In the high CG loading condition, the Freightliner experienced trailer wheel lift at a maneuver entry speed of 41 mph and the Volvo experienced trailer wheel lift at a maneuver entry speed of 45 mph.

With the ESC system, the Freightliner's stability control system was observed to limit peak lateral acceleration to approximately 0.50g, which prevented trailer wheel lift in the high CG load condition for tests performed up to 50 mph. Tests performed at 51 mph resulted in trailer wheel lift. The Volvo's stability control system limited the tractor's maximum lateral acceleration to approximately 0.40g and prevented trailer wheel lift for the high CG condition up to a maximum test speed of 51 mph.

With only a trailer-based RSC system, trailer wheel lift was observed during the high CG load condition when the system was overdriven at 41mph when tested with the Freightliner, which represented no improvement over the baseline condition. Trailer wheel lift was observed at 50 mph when tested with the Volvo, which represented a 5 mph improvement over the baseline condition. When tested with this maneuver in the high CG load condition, the trailer-based RSC system activated the trailer brakes at entrance speeds of 30 and 33 mph for the Freightliner and Volvo, respectively.

All stability control systems tested improved the roll stability of the vehicle over the baseline condition. For each maneuver, the tractor-based stability control systems were able to mitigate trailer wheel lift at the same or higher maneuver entrance speeds than trailer-based systems. The trailer-based RSC system was typically able to mitigate trailer wheel lift at a higher maneuver entry speed than the baseline condition, with the exception of the double-lane-change maneuver with one of the tractors. In the tests with both tractor-

based ESC systems and trailer-based RSC systems enabled, the tractor-based ESC system was often found to be the first system to intervene to reduce wheel lift or understeer.

Based on the results of Phase I, the agency determined that a performance test based on the J-turn was suitable to evaluate tractor and trailer stability control systems. The J-turn maneuver generates a sufficient amount of lateral acceleration to provide a challenging test at reasonable test speeds. The J-turn maneuver is also more representative of the real-world conditions, such as curved off-ramp, that could generate untripped rollover. Because the results from Phase I showed that tractor-based stability control systems increased the roll stability by a larger margin than trailer-based RSC systems, NHTSA concluded that Phase II research should focus on tractor-based stability control systems.

2. Developing a Dynamic Test Maneuver and Performance Measure To Evaluate Roll Stability—Phase II

(a) Test Maneuver Development

The researchers at VRTC conducted Phase II to develop test methods that could evaluate stability control system performance objectively and measures of performance that would ensure that a stability control system could prevent rollover effectively. After Phase I test results demonstrated that a test driver's steering input variation could affect test outcome, an automated steering machine was used for subsequent research. The testing focused on tractor-based stability control systems that were determined to be most effective in preventing rollovers from the Phase I research.

Both the Freightliner and Volvo 6x4 tractors equipped with an ESC system from Phase I were tested, and an RSC electronic control unit was also obtained for the Freightliner. A Sterling 4x2 equipped with a Meritor WABCO RSC system was also tested in Phase II. In addition to the Fruehauf 53-foot van trailer used in Phase I (its trailer-based RSC system was disabled throughout the Phase II testing), five additional trailers were tested, including a second 53-foot van trailer, two 48-foot flatbed trailers, a 9200-gallon tanker trailer, and a 28-foot flatbed trailer which is used as a control trailer in FMVSS No. 121 brake system testing.

The first maneuver evaluated in Phase II was a slowly increasing steer maneuver. The SIS maneuver has been used by the agency and the industry to determine the unique dynamic characteristics of each vehicle. This maneuver is included in the FMVSS No. 126 test procedure for ESC systems on light vehicles. The maneuver provides the steering wheel angle to lateral acceleration relationship for each vehicle, accounting for the differences in steering gear ratios, suspension systems and wheelbases among vehicles. It also normalizes test conditions to account for variations in test conditions, such as road surface friction. The steering wheel angle derived from the SIS test was used to program the automated steering machine for the ramp steer maneuver discussed below.

To initiate the SIS maneuver, the test driver accelerated the vehicle to a constant speed of 30 mph on a dry road surface. The driver then activated the steering machine to input a steadily increasing steering wheel angle up to 270 degrees at a rate of 13.5 degrees per second. The test driver manually maintained constant speed using the accelerator pedal while the tractor's path radius steadily decreased and the tractor's lateral acceleration steadily increased. The SIS maneuvers were conducted with the tractor in the bobtail condition (no trailer attached). The SIS maneuver also demonstrated that tractor-based stability control systems are capable of detecting a high lateral acceleration condition and intervening by reducing the engine output torque.

The SIS maneuver was used to determine the steering wheel angle projected to generate 0.5g of lateral acceleration when traveling at 30 mph. This value varied depending on characteristics of the tractor such as its wheelbase and steering ratio. For tractors, that steering wheel angle and lateral acceleration data was found to have a linear relationship at the lateral acceleration values between 0.05 and 0.3g. Over this range of data a linear regression method followed by linear extrapolation was used to estimate the steering wheel angle at 0.5g lateral acceleration for each SIS maneuver. The final steering wheel angle was then calculated by averaging the values from tests conducted while turning to the left and while turning to the right. The resulting calculated steering wheel angles were 193 degrees for the Freightliner, 199 degrees for the Volvo, and 162 degrees for the Sterling. This indicates that the Sterling, which was a 4x2 configuration, had a higher steering wheel gain than the other tractors which were 6x4 configurations.

The SIS testing was repeated for the three tractors throughout the test program to determine the consistency of the steering wheel angle calculations and the test speeds. The resulting standard deviations in steering wheel angle were 2.5 degrees for the Sterling, 7.4 degrees for the Freightliner, and 10.2 degrees for the Volvo, although the replacement of the tires on the Volvo may have contributed to an increase in steering wheel angle during one of the repeat tests. The tractor speed at the beginning of the SIS steering input ranged from 29.6 to 32.2 mph for all of the tests.

After the SIS testing, tests were conducted using a ramp steer maneuver to assess the roll stability of tractor-trailer combinations and the effectiveness of both types of tractor-based stability control systems. The RSM was derived from and is similar to the J-turn maneuver, but instead of the driver controlling the steering wheel to follow a fixed path, the steering controller turns the steering wheel to an angle determined from the results of the SIS test. One advantage of the RSM over the J-turn maneuver is that the RSM uses a steering machine, which allows for a more consistent and repeatable steering input.

To conduct the RSM, the test driver accelerated the vehicle to a constant speed of one to two mph above the target maneuver entry speed on a dry surface and then released the throttle and de-clutched the engine. Once the vehicle coasted down to the desired maneuver entry speed, the automated steering controller initiated a steering input, at a constant rate of 175 degrees per second, up to the steering wheel angle that was derived for the tractor in the SIS test. Once the steering wheel angle was reached (the end of ramp input), it was held constant for five seconds, and then the controller returned the steering wheel angle back to zero at a steering rate of 175 degrees per second. The initial maneuver entry speed was 20 mph and it was incrementally increased in subsequent runs until a test termination condition was met. The termination conditions were as follows: Two inches of wheel lift occurring at either the tractor drive wheels or the trailer wheels; the tractor reaching a severe oversteer condition (safety cables were installed to limit the tractor-trailer articulation angle for testing safety); or the maneuver entry speed reached 50 mph without a roll or yaw instability condition. Although the intent of the RSM was to evaluate combination vehicle roll stability, testing with the trailers in the unloaded condition resulted in several occurrences of tractor yaw instability.

For all of the RSM tests, each tractor was tested with all six trailers and the trailers were either unloaded, or loaded to a high CG, on-highway combination

weight appropriate for the number of axles on the combination vehicle. For the flatbed and van trailers, the load ballast was placed on 24-inch high tables to produce a high CG height, and the tanker trailer was loaded with water.

The purpose of the RSM test is not to cause a rollover, but to create a high lateral acceleration condition to demonstrate that a stability control system has the capability to reduce the likelihood of a rollover. Typically, wheel lift occurred first at the trailer wheels although the flatbed trailer combinations had tractor drive wheel lift occurring first or in unison with the trailer wheels. In the RSM tests with the stability control system disabled and the trailer in the high CG condition, wheel lift occurred at entry speeds between 25 and 31 mph for all combinations of tractors and trailers. The peak tractor lateral acceleration at wheel lift was in the range of 0.45 to 0.50g, showing that the high CG loading condition was representative of fully loaded tractor-trailers with a medium density cargo.

Tractor-based stability control systems applied the foundation brakes on the tractor and trailer, which reduced the vehicle speed and lateral acceleration during the RSM. The entry speed at which wheel lift was first visible improved to between 31 and 42 mph for three of the four tractors tested (Freightliner RSC, Freightliner ESC, and Volvo ESC).

In tests with the trailer brakes disabled, the entry speed at which wheel lift was detected was between 29 and 41 mph, which showed that the contribution of trailer braking to prevent wheel lift was evident, but that it was relatively small in comparison to the deceleration resulting from tractor braking. The Sterling tractor equipped with an RSC system had wheel lift with three of the trailers at the same speed as with the stability control system disabled, and with the other three trailers at speeds between two and four mph over the disabled test condition. In all of the RSM tests, the Sterling tractor's RSC system was not as effective at mitigating wheel lift for this maneuver.

The results indicated that, in general, the ESC systems provided a higher level of deceleration compared to the RSC systems and typically had the higher maneuver entry speeds prior to wheel lift. However, there were individual trailer combinations in which the RSC system performed as well or slightly better than the ESC system on the Freightliner. We believe the better performance by the RSC system in some tests is attributable to the RSC system having a more aggressive braking strategy than the ESC system tested.

The RSM was then performed with each of the six trailers in the unloaded condition, with the tractor stability control system enabled with the trailer brakes disabled. Tests were not conducted with the systems disabled. The initial maneuver entry speed was 20 mph and was incrementally increased in subsequent runs until the speed reached 50 mph, severe oversteer occurred, or wheel lift occurred. The tractors with ESC systems enabled were able to complete all but one of the RSM tests up to 50 mph without any tractor instability or wheel lift. The Volvo tractor towing the empty tanker trailer resulted in wheel lift of the tractor drive wheels and the trailer wheels at a speed of 47 mph.

In comparison, most of the tests with the tractors equipped with RSC systems towing unloaded trailers resulted in severe tractor oversteer, with the tractor-trailer articulation angle typically reaching the limits allowed by the safety cables. This occurred at speeds between 35 and 39 mph for the Freightliner 6x4 tractor and between 34 and 42 mph for the Sterling 4x2 tractor. However, both of these tractors were able to complete the RSM up to 50 mph when coupled to the unloaded 28-foot control trailer, and the Freightliner reached 50 mph without wheel lift or severe understeer when coupled to the unloaded tanker trailer.

In summary, the goal of the Phase II research was to develop a test maneuver to challenge the roll propensity of a truck tractor. The RSM is similar in test severity to the J-turn and demonstrates that the stability control systems are able to mitigate wheel lift in most cases that occurred when the stability control systems were disabled. In the high CG load condition, the ESC systems were observed to mitigate wheel lift at or above the speed observed with RSC-equipped vehicles, with the exception of a few instances with the Freightliner's ESC system. When tested with the unloaded test trailer, substantial improvements in tractor yaw stability were evident in the tractors equipped with ESC systems during RSM tests.

(b) Performance Measure Development

NHTSA's Phase II testing also examined possible performance measures to evaluate roll stability. In situations where the vehicle's stability limits are approached in a gradual manner, engine/power unit control can improve stability in these situations. However, in situations where stability limits of the vehicle are approached rapidly, application of the vehicle's foundation brakes may be a more appropriate means of improving stability.

The agency investigated four measures for development as metrics for engine/power unit control. They were truck tractor speed, truck tractor lateral acceleration, truck tractor longitudinal acceleration, and actual engine torque and driver requested engine torque.

The forward speed of a truck tractor appears to be directly related to the lateral forces generated during an untripped rollover. Test data from four different vehicles with stability control enabled indicated that forward speed was reduced from the target maneuver entrance speed of 30 mph. However, due to the nature of the roll maneuver, it is possible for the vehicle to lose traction on the inside wheels, which results in a reduction in vehicle speed but does not necessarily enhance vehicle stability.

Lateral acceleration was a possible measure of performance because of its direct relationship in producing the forces associated with untripped rollover. Data from four different tractors with the stability control system enabled indicate that each combination of tractor and stability control system had a different lateral limit that the system has allowed. This shows that the control strategy used by the manufacturer is different depending on the vehicle and system used. One strategy allows the vehicle to build lateral acceleration to a set threshold level and then allows that level to be maintained throughout the maneuver. The other strategy allows lateral acceleration to build and then the stability control system reduces the lateral acceleration. Both of these strategies were observed to increase lateral stability. Because the lateral acceleration limits were different for vehicles using these control strategies, lateral acceleration alone was not found to be a good measure for stability control performance.

Longitudinal acceleration of a vehicle is reduced when a vehicle's stability control system is enabled and is directly related to a reduction in forward speed. On the four vehicles tested, the stability control activation had measurable differences in longitudinal acceleration, but had similar disadvantages to forward speed in being used as a performance metric.

Engine torque measures were observed to be a direct way to determine ESC activation during the SIS tests. Engine torque refers to two different measures. The first relates to the torque output from the engine and is expressed

as a percentage of maximum engine output. The second relates to the throttle pedal used by the driver to control engine torque output. This value is also expressed as a percentage of maximum engine output and is referred to as the “driver requested torque.” During normal operation the “driver requested torque” and “engine torque” measures were observed to be equal to each other. However, during ESC activation when engine control intervened, the two measures were observed to be separate. In every case, the “engine torque” was much less than the “driver requested torque” and continued to reduce until vehicle stability was regained. After careful review of the data the torque separation activity was confirmed for all the SIS test series in which stability control was enabled for each vehicle. This led the agency to conclude that this measure was a good candidate for further analysis and development as a measure of performance for truck tractors equipped with a stability control system.

The engine torque data analysis was based on the test driver attempting to maintain a constant vehicle speed at the point of stability control engine torque intervention by making a substantial increase in driver-requested engine torque. For the four vehicles tested, the driver requested engine torque after stability control intervention was between 60 percent and 100 percent of engine output whereas the engine torque output after stability control intervention ranged from zero to 60 percent. The analysis of engine torque differentials was limited to the first four seconds after stability control engine torque intervention since none of the SC systems were observed to make substantial reapplications of engine torque output during this initial time- frame. On two vehicles engine torque interventions reduced engine output torque to zero during the first four seconds, and both systems allowed engine torque to be momentarily reapplied to over 50 percent of engine torque output. The Volvo had the highest engine torque output during the first four seconds after intervention, which ranged from 23 percent to 18 percent of maximum engine torque.

The agency also investigated several other measures for development for foundation braking in rollover tests because stability control systems were observed to improve the vehicle's roll stability by applying the foundation brakes. The measures investigated were wheel lift, lateral acceleration, lateral acceleration ratio, trailer lateral acceleration ratio, and trailer roll angle ratio.

Wheel lift is a direct measure of performance with minimal calculations needed to determine its value. The measure is simple and directly represents the pre-crash condition that immediately precedes a rollover. If wheel lift can be prevented, a rollover cannot occur. For our research, wheel lift was considered to occur upon two inches of lift for the tractor drive axle wheels or the trailer wheels. Wheel lift does not always indicate that rollover is imminent, particularly because certain suspension designs will lift a wheel during hard cornering. We estimated the vehicle speed that produced wheel lift during the ramp steer maneuver and found that between 29 mph and 32 mph, there is a high probability of wheel lift occurring on the combination vehicles tested. Given that only four different truck tractors and six different test trailers were used, we believed that the data may not be sufficient to assess the real world service of tractors with ESC expected to function with different trailers having different torsional stiffness and loads.

Using lateral acceleration as a performance metric is based on the principle that a tractor-trailer combination vehicle with a high center of gravity that achieves a certain level of lateral acceleration would roll over. Tests performed on the Freightliner in combination with all trailers configured with a high-CG load, at a mean entrance speed of 28 mph generated a lateral acceleration. The data showed that using tractor maximum lateral acceleration as a performance criteria would not discriminate between vehicles equipped with stability control and those without it. However, it did show that a ratio-based metric could be more appropriate for such a performance metric.

Lateral acceleration ratio is calculated by dividing the tractor's lateral acceleration at a given time interval by the measured lateral acceleration at the end of ramp input, which is the end of the steering maneuver and the point near which the vehicle experiences its peak lateral acceleration. The LAR was plotted at five equal one-second intervals for several truck tractors and test trailers. The plots indicated sharp decreases in LAR caused by activation of the stability control system.

A similar ratio metric for trailers, trailer lateral acceleration ratio, also showed the ability to discriminate between vehicles with stability control systems and those without. A third ratio metric was considered, trailer roll angle ratio based on a test trailer roll angle, but it did not clearly discriminate between vehicles with stability control systems and those without.

3. Developing a Dynamic Test Maneuver and Performance Measure To Evaluate Yaw Stability—Phase III

(a) Test Maneuver Development

The purpose of the Phase III research was to develop maneuvers to evaluate the yaw stability performance of stability control systems on tractors. Although we have examined several maneuvers to evaluate yaw stability, two maneuvers were fully investigated because other maneuvers were not able to provide a consistent, repeatable performance test. We fully considered a sine with dwell test maneuver that is similar to the test maneuver used in FMVSS No. 126 for light vehicles; and a half-sine with dwell (HSWD) test maneuver. The steering inputs for the SWD and HSWD maneuvers are depicted in the figures below, and as discussed in additional detail, variations on the steering wheel angle, the frequency of the sine wave (cycles per second, Hz), and the dwell time were evaluated for both maneuvers. A steering machine was used to achieve consistent steering wheel inputs for these maneuvers.

EP23MY12.004

The test vehicles used in Phase III included: A 2006 Freightliner 6x4, which was tested with both ESC and RSC systems; a 2006 Volvo 6x4 tractor with an ESC system; and a Sterling 4x2 tractor equipped with an RSC system. Although most of the testing was performed using the 28-foot flatbed control trailer, each tractor was also tested with a 53-foot Strick van trailer, a 48-foot Fontaine spread axle flatbed trailer, and a 9600-gallon Heil tanker trailer. Tests were conducted with the trailer brakes both enabled and disabled.

Two tractor loading conditions were used for both the SWD and HSWD testing. Each tractor was tested in the bobtail condition (no trailer attached) and using a trailer loaded over the fifth wheel so that the tractor drive axle(s) was loaded to 60 percent of its gross axle weight rating (GAWR). The yaw instability that occurred in the RSM testing showed that the unloaded 28-foot control trailer was too light to produce yaw instability. Therefore, additional weight was added for these tests. Testing was conducted on two test surfaces: A high-friction dry road surface and a slippery wet Jennite road surface.

Additional SIS tests were performed, similar to the bobtail SIS tests described in Phase II, conducted with each tractor coupled to the 28-foot control trailer and loaded to the 60 percent GAWR condition. The steering wheel angles from these tests were 197 degrees for the Freightliner with ESC, 200 degrees for the Freightliner with RSC, 200 degrees for the Volvo, and 153 degrees for the Sterling. The average tractor lateral acceleration at engine torque intervention in the SIS tests was 0.40g for the Freightliner with ESC, 0.34g for the Freightliner with RSC, 0.35g for the Volvo, and 0.4g for the Sterling.

For the SWD and the HSWD test maneuvers, the maneuver entrance speed for the bobtail tractor tests was 50 mph, and for the tests at 60 percent GAWR the entry speed was 45 mph. The driver accelerated the test vehicle up to a speed slightly over the desired speed in a straight lane, then released the throttle and de-clutched the engine. Once the vehicle coasted down to the desired speed, the automated steering machine initiated either the sinusoidal or half-sine steering input, at a specified test frequency as described below (

e.g.,

0.3 Hz, 0.5 Hz, etc.), with the steering wheel angle held constant during the dwell, as depicted in the figures. Two dwell times were evaluated as described below, 0.5 and 1.0 second. The initial test run began with a steering wheel angle equal to 30 percent of the angle determined from an SIS test. The test severity was increased in subsequent runs by increasing the steering wheel angle in 10 percentage point increments until reaching 130 percent of the SIS steering wheel angle. Thus, 11 test runs were needed to complete a test series. If severe oversteer or wheel lift greater than two inches was detected, then the test was repeated using the previous steering wheel angle in which the systems was observed to be stable. If the tractor-trailer was stable during the repeated run, additional tests were performed by increasing the steering wheel angle in 5 percent increments until instability was observed.

Tests were conducted on baseline tractors in the 60 percent GAWR condition on dry pavement to evaluate frequency and dwell time for the SWD and HSWD test maneuvers. Frequencies between 0.3 and 0.7 Hz were evaluated. A frequency of 0.5 Hz was found to require the lowest steering scalar to produce severe oversteer in the Sterling and Volvo tractors in the SWD maneuver, and 0.4 Hz was found to require the lowest steering scalar to produce severe oversteer in the Freightliner tractor (and 0.5 Hz was the second-most severe frequency for this tractor). A dwell time of 1.0 second was found to result in severe tractor oversteer at lower steering scalars. Thus the researchers selected a 0.5 Hz frequency and 1.0 second dwell time as the parameters for the SWD and HSWD maneuvers. However, the researchers also found that the SWD maneuver was less sensitive to differences in steering frequency compared to the HSWD maneuver.

In tests conducted with baseline tractors in the bobtail condition, no yaw instability occurred; however, in both the SWD and HSWD tests the Sterling tractor experienced wheel lift at the tractor drive wheels. Seventy test series were conducted on the baseline tractors in the 60 percent GAWR load condition, with fifteen of the series terminated due to roll instability and 28 due to severe tractor oversteer.

In tests conducted with the tractor stability control system enabled and in the 60 percent GAWR load condition, all of the tractors with an ESC system were able to complete the SWD maneuver at test scalars up to 130 percent. However, the tractors equipped with RSC systems experienced severe oversteer in 12 of 15 test series at the steering scalars of 120 and 130 percent. In tests conducted using the HSWD maneuver, the ESC-equipped tractors completed seven of eight test series without tractor yaw instability, and the RSC-equipped tractors experienced

severe oversteer at steering scalars ranging from 80 to 125 percent. In both test maneuvers, the RSC systems improved tractor yaw stability compared to the baseline tractor, but they could not maintain yaw stability at the higher steering scalars.

Additional SWD tests were conducted with the 53-foot van trailer and the 48-foot flatbed trailer using the 60 percent GAWR loading condition. In eight test series conducted with the tractor stability control systems enabled, seven were completed without wheel lift or tractor yaw instability, but the Sterling tractor equipped with an RSC system tested with the 48-foot flatbed reached a termination condition at a steering scalar of 105 percent. In tests with stability control enabled, all of the tractors coupled to the tanker trailer experienced wheel lift in the SWD maneuver at scalars between 60 and 95 percent.

SWD tests were also conducted on a low-friction wet Jennite surface using a lower maneuver entry speed of 30 mph. In the baseline condition with the tractor stability control systems disabled, 43 test series were conducted and a termination condition was reached in only four test series. Testing on the dry, high-friction surface was found to result in more yaw instabilities than the testing conducted on the low-friction, wet Jennite surface.

In summary, the purpose of Phase III research was to develop a maneuver to evaluate the yaw stability of a tractor trailer combination vehicle. VRTC researchers found that the SWD maneuver with a one-second dwell time based on a single cycle of steering input with a frequency of 0.5 Hz conducted on a high friction surface appropriately assessed the ability of an ESC system to improve yaw stability. From this maneuver, performance measure were investigated for lateral stability and responsiveness: the lateral acceleration ratio, which is directly correlated to roll stability and the yaw rate ratio, which the performance metric used in FMVSS No. 126 for light vehicle ESC systems and was found to be a direct performance measure of yaw stability. A responsiveness measure was also studied to evaluate the lateral displacement of a vehicle during SWD maneuvers.

(b) Performance Measure Development

Phase III of NHTSA's research also examined potential measures of yaw instability prevention performance. In light of the conclusion in Phase II that lateral acceleration ratio was a suitable metric to measure a stability control system's ability to prevent lateral acceleration, the agency examined a yaw rate ratio metric. The YRR expresses the lateral stability criteria for the sine with dwell test to measure how quickly the vehicle stops turning, or rotating about its vertical axis, after the steering wheel is returned to the straight-ahead position. Similar to the LAR, the YRR metric is the percent of peak yaw rate that is present at a designated time after completion of steer. This performance metric is identical to the metric used in the light vehicle ESC system performance requirement in FMVSS No. 126. Phase III research found that both LAR and YRR were capable of measuring stability during the SWD maneuver. However, while LAR was better at predicting roll instability, YRR was better at predicting yaw instability.

4. Large Bus Testing

Researchers at VRTC tested three large buses equipped with stability control systems: A 2007 MCI D4500 (MCI #1), a 2009 Prevost H3, and a second 2007 MCI D4500 (MCI #2). The MCI buses were equipped with a Meritor WABCO ESC system and the Prevost was equipped with a Bendix ESC system. RSC systems were not offered on large buses and, consequently, were not evaluated. All of the buses were equipped with air disc brakes. Both the MCI #1 and the MCI #2 had a GVWR of 48,000 lb and a wheelbase of 317 in., and the Prevost had a GVWR of 53,000 lb and a wheelbase of 317 in. Each of the buses had three axles: A steer axle, a drive axle, and a non-driven tag axle.

The MCI #1 was equipped with outriggers supplied by MCI and Meritor WABCO. The outriggers limited the use of higher maneuver entry speeds for tests without the ESC system enabled. At higher speeds, the lower support portion of the outrigger would dig into the test surface and influence the dynamics of the vehicle. Therefore, tests of the MCI #1 at higher speeds had no baseline performance to compare to.

The Prevost and MCI #2 buses were tested using NHTSA-designed outriggers. The outriggers designed for combination vehicles were adapted for installation on the mid-section of each bus, just in front of its drive axle and slightly behind its longitudinal center of gravity. Using these outriggers, the vehicles were able to complete testing for all speeds, with or without ESC enabled.

Each bus was tested using two primary simulated load conditions. The first condition was a lightly loaded vehicle weight (LLVW) that included the weight of the test instrumentation, outriggers, and driver. The second load condition, gross person occupancy weight (GPOW), included the LLVW weight plus the addition of 175-lb water dummies in each available passenger seat without exceeding the GVWR of the vehicle. This condition was used to represent a high CG load that a bus may experience while in service. A third loading condition was conducted with the Prevost, which added ballast to the cargo holds under the mid-section of the bus. This condition loaded the vehicle to its GVWR.

Test maneuvers that were conducted included the 150 ft. constant radius increasing velocity test, SIS, RSM, HSWD, and SWD. Tests were conducted using an automated steering machine, except for the constant radius maneuvers. The severity for each test maneuver was increased either by increasing vehicle speed or steering angle.

SIS maneuvers were conducted under both loading conditions, with ESC systems enabled and disabled, and in both left and right directions in order to characterize each vehicle. Initially, the maneuver was executed exactly as it was for the tractor testing. However it was observed that steering to a maximum steering wheel angle of 270 degrees generated barely over 0.3g of lateral acceleration. From this, it was clear that large buses have a larger steering ratio, and it would take a larger steering input to achieve the appropriate lateral acceleration levels. The steering wheel angle necessary to achieve 0.5g in the LLVW loading condition was 405 degrees for the MCI #1, 352 degrees for the Prevost, and 407 degrees for the MCI #2. In the GPOW loading condition, steering wheel angles were found to be 405 degrees for the MCI #1, 383 degrees for the Prevost, and 461 degrees for the MCI #2.

SIS tests were conducted at GPOW to evaluate the ability of the ESC system to reduce speed by limiting engine torque. For the three buses tested the average speed at activation for each SIS maneuver ranged between 29.8 and 30.6 mph. At four seconds following SC activation the average speed for each SIS had been reduced to 27.9 mph for the MCI #1, 26.5 mph for the Prevost, and 26.6 mph for the MCI #2. Without stability control enabled, speeds did not decrease. The average lateral acceleration for a test series observed at activation was 0.32g for MCI #1, 0.27g for the Prevost, 0.31g for MCI #2.

RSM testing was completed for each bus to evaluate their roll propensity while loaded in the LLVW and GPOW conditions. Tests were conducted using the same RSM protocol as the one developed for tractors. Using an

automated steering machine programmed with the steering wheel angle calculated from the SIS maneuver, tests were conducted with ESC systems enabled and disabled. The initial maneuver entry speed was 20 mph and was incrementally increased in subsequent runs until two inches of wheel lift occurred at any of the wheels, the vehicle went into a severe oversteer condition, or the entry speed reached 50 mph without a roll or yaw instability condition.

For RSM tests with ESC systems disabled and the buses loaded in the LLVW condition, wheel lift was observed in both MCI test vehicles at speeds of 41 to 45 mph, and no wheel lift was observed for tests with the Prevost for the speeds tested. When tested in the GPOW condition, wheel lift was observed at 35 to 39 mph for all vehicles tested.

For RSM tests with ESC systems enabled and the buses loaded in the LLVW condition, no instances of wheel lift were observed over the range of speeds tested. During tests in the GPOW condition wheel lift was not observed in either MCI over the range of speeds tested, but was observed in some of the Prevost tests at speeds between 42 and 48 mph.

Initial tests conducted with the Prevost demonstrated that the vehicle was able to complete the RSM at up to 48 mph without wheel lift for the GPOW condition. The Prevost was not tested to 50 mph because there was not enough test area to bring the vehicle up to this speed and allow the driver to recover safely if the test needed to be aborted. RSM tests under the same conditions were repeated less than a week later. During these tests, wheel lift greater than 2 inches was observed at speeds of 42 to 44 mph with ESC enabled. Upon further investigation when preparing to de-instrument the vehicle, a broken roll stabilizer bar was discovered. Researchers attributed the change in performance observed to the broken stabilizer bar.

SWD testing was completed for each bus to evaluate its yaw propensity while loaded in the LLVW and GPOW conditions. All tests were conducted with the ESC systems enabled and disabled. Using an automated steering machine, the SWD tests were run using steering frequencies of 0.3, 0.4, 0.5, and 0.6 Hz, dwell times of 0.5 and 1.0 seconds, and a maneuver entry speed of 45 mph. Test severity was increased by increasing the steering wheel angle by a scalar from 30 to 130 percent in 10 percent increments. A test series was terminated if the vehicle experienced wheel lift greater than 2 inches, the vehicle spun out, or the steering input reached a terminating scalar of 130 percent.

No instances of spinout were observed during this testing, but tests at higher steering wheel angles produced drift. Although the buses were yaw stable in the maneuvers, the test results demonstrated that the SWD maneuver was challenging the buses' roll propensity. Several SWD test series with the GPOW condition produced wheel lift when the ESC system was disabled. When the ESC systems were enabled, all vehicles were able to complete their series without exceeding either roll or yaw stability thresholds.

The SWD test data from the GPOW load condition were analyzed to determine a frequency and dwell time for a candidate performance maneuver. For all tests with ESC disabled, maneuvers with a 1.0-second dwell time required an equal or lower steering scalar (0 to 50 percent lower) to exceed a threshold of 6 degrees of yaw angle. As with the tractor testing, this suggested that the 1.0-second dwell time was more challenging to large buses because it required less steering to exceed the threshold.

Using only the 1.0-second dwell time tests, analysis to determine the optimum frequency for the SWD test was completed by evaluating the roll and yaw angles. Review of the test data indicated that the largest roll and yaw angles were produced in the maneuvers using 0.4 and 0.5 Hz frequencies.

The large buses were also tested using the HSWD maneuver. Like the SWD, the test results for the HSWD indicated that the longer dwell time was more challenging to stability. Unlike the SWD, the lower frequencies were observed to produce wheel lift at lower steering wheel angle scalars. Tests results from both the SWD and HSWD maneuvers indicated that both maneuvers generated dynamic responses from the vehicles. There were clear differences in lateral acceleration and yaw rate between test series conducted with ESC systems enabled compared to test series with ESC systems disabled. The data showed that ESC systems were reducing both rollover and spinout propensities. However, the SWD maneuver was favored over the HSWD maneuver because the SWD maneuver could be conducted in a smaller area, would be representative of a crash avoidance or lane change maneuver, and its use in FMVSS No. 126 accelerated performance measure research.

This research indicates that large buses equipped with ESC systems can use the same objective performance maneuver as was developed for tractors. Testing also indicates that the same performance measures can be used to assess lateral stability and responsiveness, but the performance measures must be tailored for the vehicle differences.

D. Truck & Engine Manufacturers Association Testing

The Truck & Engine Manufacturers Association (EMA) performed tests on ten tractors listed in the following table equipped with stability control systems using the three test maneuvers developed at VRTC.

Table 2—EMA Test Tractors Including Type, GVWR, and Wheelbase

Tractor configuration

(EMA Vehicle I.D.)

Stability control type

GVWR

(lb)

Wheelbase

(inches)

6x4 Typical Tractor (Vehicle A)

ESC

52,000

228

4x2 (Vehicle B)

ESC

32,000

140

4x2 (Vehicle C)

RSC with steering wheel angle sensor

34,700

152

6x4 Severe Service (Vehicle D)

ESC

66,000

220

6x4 w/Pusher Axle (Vehicle E)

ESC

86,000

270

8x6 Tridem Drive Axle (Vehicle F)

ESC

89,000

263

6x4 w/Pusher Axle (Vehicle G)

ESC

92,000

243

6x4 Severe Service (Vehicle H)

RSC

60,600

246

6x4 (Vehicle I)

ESC

52,000

232

6x4 (Vehicle J)

ESC

52,350

245

EMA provided its test data to the agency.

Although the tractors were not identified by make or model, EMA provided the configuration and weight ratings for each tractor. Eight tractors were subjected to the SIS and RSM to evaluate rollover prevention, and three tractors were subjected to the SWD maneuver, and the ramp with dwell (RWD) maneuver on a low-friction surface to evaluate yaw stability. Two of the tractors were equipped with RSC systems and seven tractors were equipped with ESC systems. EMA also submitted test data for several maneuvers in which the test parameters were varied. With the exception of Vehicle J, EMA did not submit baseline test data—that is, EMA submitted data only for maneuvers with ESC or RSC systems enabled.

Data from Vehicles A through I are included have been placed in the docket. Docket Nos. NHTSA-2010-0034-0011 through NHTSA-2010-0034-0021 and Docket No. NHTSA-2010-0034-0024. Vehicle J testing is discussed in detail in a later section.

1. Slowly Increasing Steer Maneuver

For all tractors, test data were provided for the SIS tests used to derive the steering wheel angle with each tractor in the bobtail condition. In the first SIS series conducted on eight of the tractors, three SIS tests were conducted in each direction on a dry road surface, and a best fit linear regression was used to project the steering wheel angle for a lateral acceleration of 0.5g. The average of the absolute value of each of the six runs was calculated for the final angle.

Compared to the steering wheel angles that were derived for the three VRTC tractors, a much wider range in SWA was seen among EMA's results. The steering wheel angles generally increased with the tractor's wheelbase from an angle of 126 degrees for the 140-inch wheelbase 4x2 to an angle of 291 degrees for the 270-inch wheelbase 6x4 with a pusher axle. For Vehicle H, EMA also provided data from direct measurement of the steering wheel angle from driving the tractor at 0.5g of lateral acceleration. This angle was 290 degrees, which is slightly larger than the calculated value of 281 degrees extrapolated from the SIS test data in the 0.05 to 0.30g operating region. The EMA data provided for these SIS tests did not include information on stability control engine torque reduction.

Additional SIS tests were conducted on three tractors that were to be subsequently tested using the SWD maneuver to evaluate tractor yaw stability. The SIS test conditions were identical to the prior SIS tests. A best fit linear regression was used to project the steering wheel angle for a lateral acceleration of 0.5g, and the average of the absolute value of each of the six runs was calculated for the final angle as in the prior SIS tests. Comparing these data to the prior SIS test results, Vehicle B, which had the smallest angle of 126 degrees in the prior SIS tests, showed a ten degree reduction of its angle in this test series. Vehicle G's angle was nearly identical (203 degrees in the first series vs. 205 degrees in the second series).

2. Ramp Steer Maneuver

For the RSM tests on eight tractors, the tractors were attached to a FMVSS No. 121 control trailer and were loaded to their GVWR by placing the ballast over the fifth wheel, with the ballast placed directly on the trailer deck resulting in a low center of gravity height. The weight on the FMVSS No. 121 control trailer's single axle ranged between 5,720 and 5,930 lb for all eight tractor tests, and the trailer brakes were not enabled. While the weight on the trailer axle is nominally 4,500 lb when the trailer is used for FMVSS No. 121 stopping distance tests, the increased weight in these RSM tests reflects the added weight of the outriggers installed on the trailer. In general, each of the tractors was loaded to its GVWR with the steer, drive, and auxiliary axles loaded to, or very close to, their respective GAWRs. The only exception was the 140-inch wheelbase 4x2 which only had 9,950 lb on the steer axle, although it was rated for 12,000 lb.

In the tests, the stability control systems automatically applied the tractor's foundation brakes to reduce speed and lateral acceleration. The initial vehicle deceleration generally coincided with the end of ramp steer input, indicating that the stability control systems were effective at reducing the lateral acceleration. The speed at wheel lift for EMA's tests ranged from 33 to 38 mph, as compared to 31 to 39 mph for the VRTC tests that used a similar unbraked trailer, but with a higher center of gravity loading condition and a higher overall vehicle test weight. Both 4x2 tractors tested by EMA experienced oversteer in addition to the wheel lift.

3. Sine With Dwell Maneuver

EMA provided test results for the SWD maneuver for four tractors equipped with ESC systems. The sinusoidal steering frequency used for testing was 0.5 Hz and the dwell time was one second. The amplitude of the steering wheel inputs started at 30 percent of the steering wheel angle derived from SIS testing, and in subsequent test runs was increased by 10 percent increments up to 130 percent of the steering angle. The SWD tests were conducted with two tractor loading conditions: Loaded to 60 percent of the drive axle(s) GAWR with the FMVSS No. 121 unbraked control trailer attached (loaded tests), and in the unloaded condition with no trailer attached (bobtail tests). The maneuver entrance speed was 45 mph and the test was conducted on dry pavement.

The results of the loaded tests for Vehicles G and I indicated that both tractors remained roll and yaw stable through the full range of testing, and there were no indications of tractor wheel lift in the test comments or the unprocessed data. The largest steering wheel angle produced the highest peak lateral acceleration, which occurred during the dwell portion of the maneuver for both tractors. Vehicle I reached approximately 0.75g and Vehicle G reached just under 0.6g. Although both tractors were close in wheelbase and tested with similar steering wheel angles, Vehicle G, tested with its liftable axle in the lowered position, was either less responsive in the SWD maneuver or its ESC performed slightly better than the ESC on Vehicle I. Both tractors had similar overall vehicle decelerations; however, the ESC on Vehicle G commanded higher steer axle braking pressures than the ESC on Vehicle I. Vehicle I appeared to have more lateral sliding in the maneuver, as its yaw rate decay was slower at the end of steering input.

Vehicle B (140-inch wheelbase 4x2) exhibited yaw instability in the SWD maneuver. This tractor had high lateral acceleration that was attained at lower steering wheel angles than for the 6x4 tractors. For example, the peak tractor lateral acceleration was already reaching 0.70g at 80 percent of the SIS-derived steering wheel angle, compared to Vehicle I which reached 0.60g and Vehicle G which reached 0.45g at this steering wheel angle scalar. The yaw rate decay after completion of steer was also much slower than for the 6x4 tractors, which appears to indicate that the vehicle was sliding much more and taking longer to return to the straight-ahead position. This is most evident in the testing at 130 percent of the SIS-derived steering wheel angle, in which the decay yaw rate decay was about 3.5 seconds.

The maneuver entrance speed was reduced to 30 mph in the bobtail SWD tests, which were conducted on a low-friction wet Jennite surface. The short wheelbase 4x2 tractor, Vehicle B, appeared to complete all of the test series without any observed instability

or control issues, and the peak tractor lateral acceleration was limited to approximately 0.3g in all tests. However, both 6x4 tractors (Vehicles G and I) appeared to have steering responsiveness issues that were particularly noticeable at higher steering wheel angles. At the reversal in steering wheel angle direction, the yaw rate and lateral acceleration response was delayed, indicating severe understeer. During the dwell portion of the maneuver at higher steering wheel angles, Vehicle I slowly built lateral acceleration up to 0.3g, while Vehicle G achieved similar but slightly lower acceleration levels. Vehicle G's yaw rate also was slower to respond at the completion of steer, taking as long as 2.5 seconds to decay to zero for the test conducted at the highest steering wheel angle tested.

4. Ramp With Dwell Maneuver

The three tractors equipped with ESC systems tested in the SWD maneuvers were also tested to the RWD maneuver. Once the initial steering wheel angle and test speed were attained, the steering machine increased the steering wheel angle to 180 degrees in one second, held that steering wheel angle constant for three seconds (the dwell portion of the maneuver), and then reduced the steering wheel angle to zero in one second. In subsequent RWD test runs, the steering wheel angle was increased in 90 degree increments up to 540 degrees.

The test results show that for Vehicles B and I, the steady-state lateral acceleration (prior to the ramp steer) was approximately 0.2g, and for Vehicle G the steady-state tractor lateral acceleration was approximately 0.1g. When the steering wheel angle was increased during the initial steering ramp input, the lateral acceleration and yaw rate increased slightly and in many of the test runs was then observed to drop off, indicating that the tractor was not responsive to the steering input. During the first two seconds of the steering dwell portion of the maneuver, the tractor lateral acceleration typically remained at 0.25g or less for all tests. During the last one second of the steering dwell, all of the test runs for Vehicles G and I showed steadily increasing lateral acceleration, as high as 0.5g, even as the steering wheel angle was reduced to zero. This indicates that the tractors were in a severe oversteer condition, and the agency speculates that the relatively high lateral acceleration may have been a result of the tractor running off of the low friction wet Jennite surface and onto a higher friction road surface. The test data show that this was always accompanied by braking on the steer axle, which is indicative of oversteer corrections being commanded by the ESC. Vehicle B had much less increase in lateral acceleration at the end of the maneuver and appeared to be under control. Late in the maneuver the commanded brake pressures for Vehicle B showed that both front and rear brake applications were made on the right side of the tractor, and the application pressures were nearly identical. Whether this is a data collection anomaly or stability control braking strategy is not certain, but Vehicle B was the vehicle that exhibited the least amount of oversteer.

The RWD test results demonstrated that the stability control systems on these tractors correctly identified the vehicle loss of control problems (severe oversteer and understeer) and took corrective action, including engine output torque intervention and commanding individual applications of the tractor's foundation brakes. However, the severity of the RWD test maneuver was sufficiently high to overdrive the capability of the stability control systems to mitigate severe understeer.

In summary, EMA provided test data for nine tractors each tested for the three maneuvers developed by NHTSA researchers. The nine tractors included a wider variety of tractor configurations than those tested by the agency, and included severe service tractors, tractors with auxiliary lift axles, a tridem drive axle tractor, and a very short wheelbase two-axle tractor. Slowly increasing steer vehicle characterization tests were conducted on all nine tractors (two with RSC and seven with ESC) in the bobtail condition and the test data were used to extrapolate the steering wheel angle that would provide 0.5g of lateral acceleration at 30 mph. These data produced a wider range of steering wheel angles than had been seen from the agency's tests on its three tractors, with the short wheelbase 4x2 having an angle of only 116 degrees, and a 6x4 tractor with a liftable pusher axle having the highest angle at 291 degrees.

EMA provided ramp steer maneuver test results for eight tractors that were loaded to their GVWRs using an unbraked 28-foot control trailer. Data were only provided for tests with the stability control system enabled, and the RSM was conducted up to speeds at which the system could successfully intervene. The range of speeds achieved at the point of overdriving the stability control systems was similar to the range of speeds from the VRTC RSM tests, although the loading conditions were slightly different. The two 4x2 tractors (one with RSC, and one with ESC) tested by EMA experienced oversteer and wheel lift, while the other tractors all experienced wheel lift.

SWD test results were provided for three tractors, each equipped with ESC, using a 0.5 Hz sinusoidal steering input frequency and a 1.0 second dwell time, and the tractors were tested in the bobtail condition and loaded to 60 percent of drive axle(s) GAWR. In the tests on dry pavement at a maneuver entrance speed of 45 mph, the typical 6x4 completed all tests, while the 6x4 equipped with a lift axle (tested in the lowered position) also completed all tests but appeared to be slower to respond to the steering inputs. The short wheelbase 4x2 tractor appeared to exhibit control problems and at the highest steering wheel angle tested. The sine with dwell tests on the three tractors in the bobtail condition were conducted on a low-friction wet Jennite test surface with a lower maneuver entrance speed of 30 mph. In these tests, the short wheelbase 4x2 tractor completed all tests, while the two 6x4 tractors appeared to experience severe understeer at the higher steering wheel angles tested.

5. Vehicle J Testing

(a) EMA Testing of Vehicle J

In December 2010, EMA provided testing data on a tenth vehicle they tested.

Vehicle J was intended to be representative of a typical 6x4 tractor, with a 245 inch wheelbase and a GVWR of 52,350 pounds. EMA subjected Vehicle J to four different test maneuvers: The slowly increasing steer test; the sine with dwell test; a J-turn maneuver, and a wet Jennite drive through test.

Vehicle J data provided to the agency has been placed in Docket No. NHTSA-2010-0034-0022 and Docket No. NHTSA-2010-0034-0023.

EMA first conducted the slowly increasing steer test maneuver with a steering controller on Vehicle J to determine the steering wheel angle that would produce a lateral acceleration of 0.5g. EMA conducted two series of test runs, one in each direction. A best fit linear regression was used to determine that the average steering angle on the six runs that would produce a lateral acceleration of 0.5g was 197 degrees. This value was used for subsequent testing.

EMA next conducted sine with dwell testing. EMA conducted two series of SWD tests—one with the ESC system on and one with the ESC system off. EMA equipped the vehicle with an FMVSS No. 121 control trailer and loaded the

vehicle so that the drive axles were loaded to 60 percent of the GAWR, which resulted in the vehicle being loaded to approximately 78.6 percent of its GVWR.

EMA provided data on six runs of the SWD maneuver. EMA conducted the test at scalars from 0.8 to 1.3 of the SIS-derived steering wheel angle. EMA also provided data on three runs of the SWD maneuver with the system deactivated. Those tests were conducted at scalars of 1.0 and 1.3, and 1.5.

Each test run with the system enabled showed a 20- to 25-mph reduction of speed during the test maneuver. In contrast, tests conducted with the system off indicated only limited speed reduction of less than five mph. This indicated that the ESC system acted to reduce vehicle speed.

Each test run with the system enabled conducted at scalars between 0.8 and 1.2 resulted in a peak lateral acceleration between 0.6g and 0.7g. The lateral acceleration then quickly dropped to zero within 0.3 to 0.4 seconds after the completion of the steer. Yaw rate during the dwell portion of the maneuver peaked at approximately 18 to 22 degrees per second, except at a scalar of 1.2 where yaw rate peaked at approximately 24 degrees per second) and showed a downward trend during the dwell, dropping by approximately five degrees per second. The yaw rate dropped to zero within 0.2 seconds after completion of steer. The vehicle's ESC system used selective braking to reduce the speed, lateral acceleration, and yaw rate responses.

With the system disabled, the test run at a scalar of 1.0 resulted in a peak lateral acceleration of approximately 0.8g. A 0.2g drop in lateral acceleration was observed at the beginning of the dwell portion of the maneuver followed by a sudden rise of the same amount, indicating possible oversteer. The lateral acceleration dropped to zero less quickly than in tests with the system on (approximately 0.5 seconds) after completion of steer. This was largely due to the drop in lateral acceleration starting later with the system off than with the system on. The yaw rate peaked at approximately 21 degrees per second. Unlike with the system on, there was not a clear drop in yaw rate during the dwell portion of the maneuver. The yaw rate also dropped to zero slower than in tests with the system off (approximately 0.25 seconds after completion of steer).

For test runs at steering wheel angle scalars of 1.3, the peak lateral acceleration was slightly lower with the system on (approximately 0.75g) in comparison to the test run with the system off (over 0.8g). Momentary variability in lateral acceleration was observed in both tests, indicating possible tractor instability. Again, with the system on, the lateral acceleration decayed faster at the completion of steer (approximately 0.4 seconds) than it did with the system off (over 0.6 seconds). This was largely due to the reduction in lateral acceleration starting later with the system off than with the system on. The yaw rate peaked for both tests at approximately 25 degrees per second. Again, however, the yaw rate decreased by approximately five degrees during the dwell portion of the maneuver with the system on while no clear decay was observed with the system off. Also, the yaw rate decreased to zero slower after completion of steer with the system off (0.25 seconds) than it did with the system on (less than 0.2 seconds).

EMA also submitted data on one SWD test run with the system off at a steering wheel angle scalar of 1.5. Peak lateral acceleration observed during this test run was nearly 0.9g. The lateral acceleration rate dropped to zero in slightly over 0.5 seconds after completion of steer. The yaw rate peaked at approximately 24 degrees per second. Unlike in runs with lower steering wheel angles, a reduction in yaw rate was observable during the dwell portion. However, that reduction was much sharper, occurring entirely within a 0.5 second period rather than throughout the entire 1.0 second dwell period. Like in prior tests, the yaw rate dropped to zero within approximately 0.25 seconds.

EMA's SWD maneuver test data from Vehicle J demonstrated that the ESC system activated to lower lateral acceleration and yaw rate during the SWD maneuver. However, even with the ESC system turned off, the lateral acceleration and yaw rates dropped relatively quickly at the end of the test maneuver, indicating that the vehicle did not become unstable during testing. Although EMA only provided test data from three runs with the system off compared to six runs with the system enabled, the runs with the system off did include a run with a steering wheel angle scalar of 1.5, which was higher than any run in NHTSA's testing, and no severe incidents of instability were observed.

EMA next conducted J-turn testing both with the system enabled and disabled. The test was conducted on a 150-foot fixed radius curve. The vehicle was tested with an FMVSS No. 121 control trailer and was loaded to the FMVSS No. 121 loading conditions. The tests were conducted at initial entry speeds of 30 to 36 mph, in increments of two mph.

In tests conducted with the ESC system enabled, system activation occurred at each test speed. The system commanded brake activations to reduce vehicle speed to 18 mph from initial speeds of 30 mph and 32 mph, down to 10 mph from an initial speed of 34 mph, and down to 6 mph at an initial speed of 36 mph. The vehicle was able to maintain the lane at all speeds tested. Lateral acceleration peaked at 0.4 to 0.5g at 30 and 32 mph and peaked at 0.6g at 34 mph and 36 mph. Yaw rate peaked at approximately 15 degrees per second at 30 and 32 mph and peaked at approximately 20 degrees per second at 34 mph and 36 mph. At the higher speeds tested, lateral acceleration and yaw rate were observed to drop coincident with speed.

With the system disabled, no reduction in speed during the maneuver was observed. Thus, lateral acceleration and yaw rates remained relatively constant throughout the maneuver. At test speeds of 30 and 32 mph, lateral acceleration peaked at approximately 0.55 to 0.65g and yaw rate peaked at approximately 20 degrees per second. At 34 mph, the lateral acceleration peaked at approximately 0.9g and the steering wheel angle necessary to maintain the lane decreased substantially. Yaw rate peaked at approximately 22 degrees per second and dropped to approximately 15 degrees per second, indicating the vehicle was starting to plow out. At 36 mph, the vehicle plowed out of the lane.

The fourth maneuver EMA performed on Vehicle J was a wet Jennite drive-through (WJDT) maneuver. This maneuver was intended to test yaw stability. The WJDT maneuver is identical to method for determining the maximum drive-through speed when testing vehicles for compliance with S5.3.6.1 of FMVSS No. 121. The vehicle is driven through a 500-foot radius curve with a wet surface having a peak coefficient of friction of approximately 0.5 at successively increasing speeds (up to 40 mph) to determine the maximum speed at which the vehicle can maintain the curve.

To conduct the FMVSS No. 121 stability and control during braking compliance test, the vehicle is driven at the lesser of 30 mph or 75 percent of the maximum drive-through speed. A full brake application is made and a vehicle must stop at least three times out of four within the 12-foot lane.

EMA performed this test with both the stability control system enabled and disabled in two load configurations. First, the vehicle was tested in the bobtail (unloaded) configuration.

Second, the vehicle was loaded to the FMVSS No. 121 test loading condition.

In the bobtail configuration with the ESC system enabled, test runs at 30 and 32 mph yielded no system activation. At 33 mph, system activation occurred as both engine torque reduction and selective braking to improve yaw stability occurred. As a result, the vehicle speed decreased to approximately 29 mph during the maneuver and the driver responded by rapidly straightening the steering wheel. Vehicle yaw rate peaked at approximately 10 degrees per second. A second run at 33 mph showed only brief system activation and a minimal reduction in speed. During two runs at 34 mph, ESC system intervention was again observed as torque reduction and selective braking reduced vehicle speed to 28 to 29 mph and the driver again responded by rapidly straightening the steering wheel. Yaw rate peaked at near 10 degrees per second and again, as the driver responded, decreased. During two runs at 35 mph, the vehicle was unable to maintain the lane due to understeer, despite system intervention.

In the bobtail configuration with the system disabled, at 32 mph, the driver had to adjust steering by adding steering input during both runs attempted at this speed, indicating substantial understeer. During two runs at 33 mph, the vehicle was unable to maintain the lane, despite large steering inputs from the driver.

In the loaded configuration with the ESC system enabled, system activation occurred at a speed of 30 mph, though only slight (1 to 2 mph) reduction in speed was observed. The driver had to increase his steering input, but there was no corresponding increase in yaw rate, indicating understeer. At 32 mph, both engine torque reduction and selective braking occurred to improve yaw stability occurred. As a result, the vehicle speed decreased to approximately 27 to 28 mph during the maneuver. At 34 mph, the ESC system intervened more substantially, resulting in a reduction of speed to approximately 26 mph. Nevertheless, the vehicle was able to maintain the lane. At 35 mph, the vehicle was unable to maintain the lane due to understeer, despite system intervention.

In the loaded configuration with the system disabled, understeer was observed at 32 mph, as evident by substantial increase in steering input by the driver; however, the vehicle was able to maintain the lane. At 33 mph, the vehicle was unable to maintain the lane.

The maximum drive through speed in both vehicle configurations was only 32 mph with the system off, compared to 34 mph with the system on. This demonstrates that an ESC system has some ability to mitigate understeer when navigating a curve on a low-friction surface, and allow the driver to maintain control at higher curve entrance speeds.

(b) NHTSA Testing of EMA's Vehicle J

At NHTSA's request, EMA provided Vehicle J to NHTSA for NHTSA to duplicate EMA's testing.

In particular, the agency was interested in the performance of Vehicle J during the sine with dwell maneuver. NHTSA's two 6x4 tractors that were tested in with the SWD represented the upper and lower size bounds of what would be considered a typical 6x4 tractor and both tractors could not maintain stability during a SWD maneuver with the ESC system disabled. Vehicle J's size is within the bounds of the two typical 6x4 tractors tested by NHTSA.

A copy of NHTSA's Vehicle J test data has been placed in the docket. Docket No. NHTSA-2010-0034-0044.

NHTSA conducted 20 test runs of Vehicle J in the SWD maneuver at steering wheel angle scalars of 0.4 to 1.3 of the SIS-derived steering wheel angle attached to VRTC's FMVSS No. 121-style control trailer. When tested with the ESC system disabled at a steering wheel angle scalar of 1.2, NHTSA was able to detect lateral instability that continued for almost two seconds after completion of the SWD maneuver.

NHTSA was able to conduct 19 test maneuvers with Vehicle J that did not result in substantial roll instability. NHTSA did not find any yaw instability in any of the 20 test maneuvers.

It was discovered that EMA conducted its testing of Vehicle J with a control trailer with different specifications than NHTSA used. NHTSA then attempted to duplicate EMA's Vehicle J's testing using the control trailer used by EMA.

The results of NHTSA's tests with EMA's control trailer were not meaningfully different than the results of EMA's testing. That is, there were no instances of substantial roll or yaw instability in 20 test runs conducted by NHTSA.

NHTSA's test data identifies the trailer used by EMA as a “Link” trailer and the trailer used by NHTSA as the “NHTSA” or “VRTC” trailer.

As a result of NHTSA's testing of Vehicle J, the agency discovered that there exist three areas of variability in FMVSS No. 121-style control trailers and loading which, while not necessarily relevant to FMVSS No. 121 testing, could affect the results of stability control system testing if the specifications for an FMVSS No. 121-style control trailer were simply carried over to a stability control standard. First, EMA's control trailer had a wider track width

than NHTSA's trailer, which made EMA's trailer, and thereby the combination vehicle, more stable during SWD testing. Second, EMA's control trailer had a lower deck height than NHTSA's trailer, which contributed to a lower center of gravity on EMA's trailer. Third, EMA loaded its trailer with steel for ballast, whereas NHTSA loaded its trailer with concrete for ballast, which also contributed to the lower center of gravity on EMA's trailer because steel would not have to be stacked as high to achieve a full load.

The track width is the distance between the centerlines of a vehicle's left and right tires. In vehicles with dual tires, the track width would be measured from between the dual tires on each side of the vehicle.

E. Other Industry Research

The SAE Truck and Bus Control Systems Task Force (renamed as the Truck and Bus Stability Control Committee) was formed in 2007 to facilitate information sharing among the industry and government regarding heavy vehicle stability control systems.

The information shared included proposed test maneuvers that could potentially be used to evaluate the performance of stability control systems. Although the Task Force has not published any formal documents describing these test maneuvers, the following provides an overview of the maneuvers that have been discussed.

See

http://www.sae.org/events/cve/presentations/2007truckbus.pdf

for an overview of the SAE Truck and Bus Council organizational chart.

1. Decreasing Radius Test

A decreasing radius test (DRT) was developed to evaluate the roll stability performance of a heavy vehicle stability control system.

With the DRT, the test conditions could also be adjusted to evaluate yaw stability as well. In the DRT, the vehicle is accelerated to a constant speed of 29 mph on a dry road surface, and an initial steering input is made to follow a curve with a 150-foot radius. Once the initial curve radius is achieved, the radius is linearly reduced to a radius of 90 feet as the vehicle negotiates 120 degrees of arc. Thus, it is similar to the J-turn maneuver. The speed of 29 mph was derived based on a vehicle dynamics simulation, which estimated that the maneuver would produce 0.3g of lateral acceleration during the initial steering input and this would steadily increase to 0.6g at the 90-foot radius curve.

See Docket No. NHTSA-2010-0034-0036.

Tests would be conducted in a loaded condition with the tractor coupled to a trailer and an unloaded condition in a

bobtail configuration. Because actual vehicle testing had not been conducted using this maneuver, pass/fail criteria have not yet been developed. Simulations of this test have been run using driver-controlled steering inputs; however, parameters could also be developed to conduct this maneuver using an automated steering controller.

2. Lane Change on a Large Diameter Circle

Volvo provided information on the Lane Change on a Large Diameter Circle (LC-LDC) maneuver that they have used to evaluate stability control system performance.

In this maneuver the vehicle is driven at a constant speed, just below the threshold speed for rollover or loss of control, around the inside lane of an 800-foot radius curve that has two lanes. The driver then drifts to the outside lane, and steers back into the inside lane. For rollover testing the asphalt road surface is dry and for yaw testing the surface is wet. The test can be conducted using a bobtail tractor, a tractor towing an FMVSS No. 121 control trailer, or a tractor towing any other type of trailer in a fully loaded condition. Volvo evaluated the roll stability performance during this maneuver based on whether the trailer outrigger made contact with the ground. Volvo considers this maneuver to be representative of certain highway segments that are encountered, and that the maneuver is severe enough to fully challenge a stability control system.

See Docket No. NHTSA-2010-0034-0042.

3. Yaw Control Tests

Bendix developed two yaw stability test maneuvers to evaluate the ability of stability control systems to prevent severe oversteer and understeer conditions. The first test maneuver is a Sinusoidal Steering Maneuver (SSM) to evaluate oversteer prevention.

The first step in this test is to identify the steering wheel angle that produces a tractor lateral acceleration of 0.5g at 30 mph on dry pavement with the tractor in the bobtail condition. Bendix recommended that this angle be derived by either a slowly increasing steer test (SIS test described in section IV.D.2 above) or an equation developed by Bendix for estimating the angle based on the tractor's wheelbase:

See Docket No. NHTSA-2010-0034-0037.

Steering Wheel Angle (δ) = (35.5 × (tractor wheelbase in meters)) + 30.94

The Sinusoidal Steering Maneuver test is then conducted with the tractor in the bobtail condition using a low-friction wet Jennite road surface (nominal peak friction coefficient of 0.5). The vehicle is driven at a constant speed of approximately 30 mph and, as a sinusoidal steering input is initiated (continuous left and right steering inputs using the steering wheel angle determined above), the driver increases the throttle position to request 100 percent of engine torque.

The second test maneuver developed by Bendix was the ramp with a dwell maneuver discussed in section IV.D.4 above.

The RWD maneuver is intended to evaluate understeer prevention, though oversteer can also occur during the maneuver. The RWD test is conducted with the tractor in the bobtail condition and using a wet Jennite road surface. The first step in this test is to characterize the vehicle's steering by conducting a series of drive-through speed evaluations at a constant speed on a 500-foot radius curve. Once the maximum constant travel speed is determined (typically between 28 and 32 mph, but not to exceed 35 mph), the steering wheel angle is measured for negotiating the curve at that speed. The RWD test maneuver speed is then conducted at the maximum drive-through speed. Bendix suggested that manual steering by a test driver or an automated steering machine could be used. Once the vehicle has been accelerated to the test maneuver speed, the speed is held constant by the driver and he inputs the drive-through steering wheel angle. After the vehicle reaches a constant lateral acceleration condition, the steering wheel angle is increased to 180 degrees in a period of one second. That increased angle is held constant for three seconds, and then the angle is reduced to zero in a period of one second. Subsequent test runs are conducted by increasing the steering wheel angle in increments of 90 degrees up to 540 degrees.

See Docket No. NHTSA-2010-0034-0038.

The RWD test performance measures would be based upon test data showing that the vehicle's stability control system successfully identified a vehicle control problem (understeer or oversteer) and intervened by reducing the engine torque output and commanding the application of individual foundation brakes in a manner that is suitable to mitigate the control problem. Bendix did not believe that vehicle yaw or path-following pass/fail criteria would be appropriate for this test maneuver.

Two maneuvers that the industry has developed to evaluate the performance of stability control systems, lane change on a large diameter circle and sinusoidal steering, can be used to demonstrate that a stability control system is capable of preventing a rollover or a yaw instability condition. The RWD maneuver may exceed the capabilities of stability control systems but provides brake application data that can be reviewed to determine if a stability control system provides the correct control responses to address a severe oversteer or understeer condition.

V. Agency Proposal

Based upon the foregoing research, the agency is proposing a new FMVSS to require ESC systems be installed on truck tractors and buses with a GVWR of greater than 11,793 kilograms (26,000 pounds).

There are several issues raised by this proposed rule on which the agency seeks public comment, each of which is discussed in detail in the following sections.

To distinguish this new FMVSS from the light vehicle ESC requirement in FMVSS No. 126, we are proposing to revise the title FMVSS No. 126 to reflect that it is applicable only to light vehicles.

A. NHTSA's Statutory Authority

NHTSA is proposing today's NPRM under the National Traffic and Motor Vehicle Safety Act (“Motor Vehicle Safety Act”). Under 49 U.S.C. Chapter 301, Motor Vehicle Safety (49 U.S.C. 30101 et seq.), the Secretary of Transportation is responsible for prescribing motor vehicle safety standards that are practicable, meet the need for motor vehicle safety, and are stated in objective terms. “Motor vehicle safety” is defined in the Motor Vehicle Safety Act as “the performance of a motor vehicle or motor vehicle equipment in a way that protects the public against unreasonable risk of accidents occurring because of the design, construction, or performance of a motor vehicle, and against unreasonable risk of death or injury in an accident, and includes nonoperational safety of a motor vehicle.” “Motor vehicle safety standard” means a minimum performance standard for motor vehicles or motor vehicle equipment. When prescribing such standards, the Secretary must consider all re

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.