New Car Assessment Program Final Decision Notice-Advanced Driver Assistance Systems and Roadmap

Federal RegisterDec 3, 2024

Ask Donna

What actually matters in this document.

Text

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

[Docket No. NHTSA-2024-0077]

New Car Assessment Program Final Decision Notice—Advanced Driver Assistance Systems and Roadmap

AGENCY:

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

ACTION:

Final decision notice.

SUMMARY:

This final decision notice adds four new advanced driver assistance systems (ADAS) technologies—blind spot warning (BSW), blind spot intervention (BSI), lane keeping assist (LKA), and pedestrian automatic emergency braking (PAEB)—to the New Car Assessment Program (NCAP) and enhances the performance evaluation of ADAS technologies currently in NCAP. The notice also finalizes a 10-year roadmap for updating NCAP through multiple phases for the period 2024 through 2033. This notice responds in part to the provisions in section 24213 of the Infrastructure, Investment, and Jobs Act.

DATES:

Decisions on planned changes to the New Car Assessment Program are effective for the 2026 model year.

FOR FURTHER INFORMATION CONTACT:

For technical issues, you may contact Ms. Taryn E. Rockwell, New Car Assessment Program, Office of Crashworthiness Standards (Telephone: (202) 366-1810). For legal issues, you may contact Ms. Sara R. Bennett, or Ms. Natasha D. Reed, Office of Chief Counsel (Telephone: (202) 366-2992). You may send mail to these officials at the National Highway Traffic Safety Administration, 1200 New Jersey Avenue SE, West Building, Washington, DC 20590-0001.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive Summary

II. Summary of Updates to NCAP and Roadmap for Future Updates

III. Background

IV. Updating Forward Collision Prevention Technologies

V. Adding Pedestrian Automatic Emergency Braking (PAEB) Technology

VI. Adding Blind Spot Technologies

VII. Updating Lane Keeping Technologies

VIII. Self-Reported Data

IX. NCAP Roadmap

X. Economic Analysis

XI. Appendix

I. Executive Summary

Since its launch in 1978, NHTSA's New Car Assessment Program (NCAP) has supported NHTSA's mission to reduce the number of fatalities and injuries that occur on U.S. roadways. NCAP, like many other NHTSA programs, has contributed to significant reductions in motor vehicle related crashes, fatalities, and injuries, with passenger vehicle occupant fatalities decreasing from 32,043 to 26,325 from 2001 to 2021.

Unfortunately, this reduction was not universal, with pedestrian fatalities increasing by 51 percent during the same timeframe, from 4,901 to 7,388.

Despite improvements in automotive safety since NCAP's implementation, far more work must be done to reduce the continued high toll to human life on our nation's roads. In response to this need, on March 9, 2022, NHTSA published a Request for Comments (RFC) notice outlining proposed NCAP updates.

Traffic Safety Facts 2021 “

A Compilation of Motor Vehicle Crash Data.

” U.S. Department of Transportation. National Highway Traffic Safety Administration. NHTSA acknowledges a recent increase in passenger vehicle occupant fatalities occurring during the COVID-19 pandemic. In 2019, 22,372 passenger vehicle occupants were killed in traffic crashes.

Traffic Safety Facts 2021 “

A Compilation of Motor Vehicle Crash Data.

” U.S. Department of Transportation. National Highway Traffic Safety Administration.

Docket No. NHTSA-2021-0002. 87 FR 13452 (March 9, 2022).

After careful consideration of all comments received and applicable regulatory considerations, this notice announces the Agency's decision to update NCAP with the enhanced evaluation of advanced driver assistance systems (ADAS) technologies currently in NCAP

and to add four new ADAS technologies to NCAP: blind spot warning (BSW), blind spot intervention (BSI), lane keeping assist (LKA),

and pedestrian automatic emergency braking (PAEB). This notice also establishes a 10-year roadmap for updating NCAP through a multi-phased approach, with RFC notices planned over the next several years. NHTSA will address comments received on program elements outside the scope of the March 2022 RFC notice in subsequent final decision notices as part of the multi-phase efforts to update NCAP over the next several years.

The ADAS technologies currently evaluated in NCAP are forward collision warning (FCW), lane departure warning (LDW), dynamic brake support (DBS), and crash imminent braking (CIB).

“LKS” was used for this technology in the March 2022 RFC. However, in this final decision notice, “LKA” is used instead to maintain consistency with other agency initiatives.

A. Legal and Policy Considerations

In finalizing its decisions for this notice, in addition to comments received, the Agency sought to address requirements from the 2015 Fixing America's Surface Transportation (FAST) Act,

the 2021 Bipartisan Infrastructure Law (BIL), enacted as the Infrastructure Investment and Jobs Act,

and the U.S. Department of Transportation's National Roadway Safety Strategy. The Agency also took into consideration its May 9, 2024, final rule for FMVSS No. 127, “Automatic Emergency Braking for Light Vehicles.”

These considerations are described below.

Public Law 114-94.

Public Law 117-58.

Docket No. NHTSA-2023-0021. 89 FR 39686 (May. 9, 2024).

1. 2015 Fixing America's Surface Transportation Act

This final decision notice serves as NHTSA's initial step in fulfilling section 24322 of the FAST Act, which directs the Agency to promulgate a rule ensuring the display of crash avoidance information next to crashworthiness information on window stickers that manufacturers place on motor vehicles.

The Agency is currently working to develop a crash avoidance rating system based on comments received in response to several rating system concepts discussed in the March 2022 RFC, and this notice finalizes additional crash avoidance technologies that will be included in the future crash avoidance rating system.

Section 24322 of the FAST Act, otherwise known as the “Safety Through Informed Consumers Act of 2015.”

2. 2021 Bipartisan Infrastructure Law

This notice also fulfills in part several mandates in section 24213 of the BIL, enacted on November 15, 2021 as the Infrastructure Investment and Jobs Act.

First, section 24213(a) requires NHTSA to “finalize the proceeding for which comments were requested” on December 16, 2015.

This final decision notice does so by adopting four new ADAS technologies discussed in the Agency's December 16, 2015 RFC notice,

thus finalizing that proceeding and notice.

Public Law 117-58.

Id.

at Section 24213(a); the notice referred to in the Bipartisan Infrastructure Law is 80 FR 78522 (Dec. 16, 2015).

Docket No. NHTSA-2015-0119. 80 FR 78591 (Dec. 16, 2015).

As communicated in the March 2022 RFC, while NHTSA is adopting a roadmap that includes aspects of the 2015 RFC, this notice is not an extension of the December 2015 notice.

Second, this notice addresses the Advanced Crash-Avoidance Technologies portion of section 24213(b) of the BIL, which directs the Secretary of the Department of

Transportation to “publish a notice, for the purposes of public comment, to establish a means for providing consumer information relating to advanced crash-avoidance technologies” within one year of enactment that includes an appropriate methodology for: (1) determining which advanced crash avoidance technologies should be included in the information, (2) developing performance test criteria for use by manufacturers in evaluating those technologies, (3) determining a distinct rating system involving each crash avoidance technology, and (4) updating overall vehicle ratings to incorporate the advanced crash avoidance technology ratings. This notice satisfies two of these four requirements by (1) adopting established criteria for determining which advanced crash avoidance technology

should be included as referenced and discussed in the March 9, 2022 RFC notice, and (2) finalizing test procedures and criteria to evaluate performance for each of these advanced crash avoidance technologies. Although the Agency is not yet implementing a rating system for individual crash avoidance technologies, it has sought comments in this regard and has detailed plans in its roadmap to finalize such ratings, along with an updated overall (

i.e.,

crashworthiness and crash avoidance) rating, in the near future.

This notice refers to advanced crash avoidance technology as ADAS technology.

Third, this notice addresses the Vulnerable Road User Safety portion of section 24213(b), which directs the Secretary to publish a notice meeting similar requirements to those mandated for advanced crash avoidance technologies “to establish a means for providing to consumers information relating to pedestrian, bicyclist, or other vulnerable road user safety technologies” within one year of enactment. By applying the established inclusion criteria in the adoption of PAEB technology and the applicable test procedures and evaluation criteria included in this notice, two of the four requirements for the Vulnerable Road User Safety portion of section 24213(b) will be met. NHTSA will fulfill the remaining requirements when it proposes and finalizes a new rating system for the crash avoidance technologies in NCAP.

Fourth, this final decision notice fulfills the requirements in section 24213(c) of the BIL. This section states that, within one year of the law's enactment, the Secretary of the Department of Transportation shall establish a roadmap, vetted through the public comment process, identifying and prioritizing safety opportunities and technologies that could be used in future roadmaps, establishing a plan for implementation of NCAP changes, and considering the benefits of consistency with other U.S. and international rating systems. Section 24213(c) further specifies that the roadmap shall span a term of ten years, with five-year mid-term and five-year long-term components. Further, it requires updates to the roadmap at least once every four years to reflect new Agency interests and diverse stakeholder input (garnered annually), and in consideration of opportunities to benefit from collaboration and/or harmonization with third-party safety rating programs. As will be discussed herein, the Agency is taking steps to harmonize with existing consumer information rating programs, where possible and when appropriate, both for this NCAP update and future initiatives included in the program's roadmap. The Agency's proposed roadmap includes phased updates, as mandated, and was made available for public comment as part of the March 2022 RFC notice. As all relevant comments received have been considered prior to this notice's finalization, the Agency has fulfilled the requirements of section 24213(c). Additional details for the mid-term and long-term five-year spans are available in the NCAP Roadmap section of this notice.

3. 2022 U.S. Department of Transportation National Roadway Safety Strategy (NRSS)

The U.S. Department of Transportation published the National Roadway Safety Strategy (NRSS) in January 2022.

The NRSS announced key planned departmental actions aimed at significantly reducing serious roadway injuries and deaths to reach the Department's long-term zero roadway fatalities goal. At the core of the NRSS is the Department-wide adoption of the Safe Systems Approach,

which focuses on building layers of protection to both prevent crashes from happening and minimize harm when crashes do occur.

U.S. Department of Transportation. (2020). “National Roadway Safety Strategy, Version 1.1.”

https://www.transportation.gov/sites/dot.gov/files/2022-02/USDOT-National-Roadway-Safety-Strategy.pdf.

https://www.transportation.gov/NRSS/SafeSystem.

With respect to NCAP, the NRSS supports program updates emphasizing safety features that protect people both inside and outside the vehicle. These safety features may incorporate consideration of pedestrian protection systems, better understanding of impacts to pedestrians (

e.g.,

specific considerations for children), and may include automatic emergency braking and lane keeping assistance to benefit bicyclists and pedestrians. The NCAP program also works to identify the most promising vehicle technologies to help achieve NRSS's safety goals, such as alcohol detection systems and driver distraction mitigation systems. In addition, the NRSS includes a 10-year roadmap for the program and lists as a key departmental action the initiation of rulemaking to update the vehicle Monroney label. As part of that process, the Agency may also consider including information on features that mitigate safety risks for people outside of the vehicle.

This final decision notice presents NHTSA's initial actions towards the implementation of this broad, multi-faceted safety strategy for NCAP that includes improved road safety for both motor vehicle occupants and people outside of the vehicle, including pedestrians and other vulnerable road users. Additionally, the 10-year roadmap for the program presents a plan for the incorporation of future safety technologies and provides a projected timeline for updating the Monroney label to include crash avoidance information.

Relatedly, NRSS lists the initiation of a new rulemaking to require automatic emergency braking and pedestrian automatic emergency braking on passenger vehicles as a key departmental action. In response to this action, NHTSA published a final rule on May 9, 2024, establishing a new Federal motor vehicle safety standard, FMVSS No. 127, “Automatic Emergency Braking for Light Vehicles.” Similar to the changes adopted by NCAP in this notice, this final rule aims to reduce the frequency and associated injury and fatalities of rear-end and pedestrian crashes. Manufacturers must comply with the final rule by September 1, 2029.

This final decision notice will upgrade NCAP to provide consumers with additional vehicle safety information on AEB and PAEB technologies to help them make more informed purchasing decisions. NHTSA will identify vehicles that are equipped with these recommended technologies and pass NHTSA's performance criteria by way of check marks on the NHTSA website starting with model year 2026

vehicles, as discussed in the following sections. Although the final rule and this decision on NCAP rely on the agency's separate authorities, NHTSA has sought to ensure that the revised test procedures for NCAP and the AEB final rule are compatible with one another, such that a manufacturer would be able to design a system that both received NCAP credit and would meet the requirements contained in the final rule.

NHTSA believes these collective efforts will lead to more rapid and complete market penetration of AEB and PAEB technologies.

Vehicles produced by small-volume manufacturers, final-stage manufacturers, and alterers must be equipped with a compliant AEB system by September 1, 2030.

See Appendix.

II. Summary of Updates to NCAP and Roadmap for Future Updates

A brief summary of the updates to NCAP included in this final decision notice is provided below, along with the finalized 10-year roadmap for future updates to NCAP.

Updates To Crash Imminent Braking (CIB), Dynamic Brake Support (DBS), and Forward Collision Warning (FCW) Evaluations

This notice modifies the existing test conditions, evaluation procedure, and performance criteria for crash imminent braking (CIB) and dynamic brake support (DBS) systems, subject to the same test scenarios currently used in NCAP.

An overview of the amended test scenarios (Lead Vehicle Stopped (LVS), Lead Vehicle Moving (LVM), and Lead Vehicle Decelerating (LVD)) and test conditions (subject vehicle (SV) speed, principal other vehicle (POV) speed, POV headway, and POV deceleration) required to receive passing credit for AEB systems (

i.e.,

CIB and DBS collectively) in NCAP is shown in Tables 1 and 2. NHTSA will test vehicles starting with the lowest test speed for a test scenario and incrementally increase test speed according to the test matrix in Tables 1 and 2, with only one trial

conducted per test condition. The passing criterion for a test trial is no contact between the subject vehicle and principal other vehicle. If the subject vehicle contacts the principal other vehicle during a test trial, the vehicle fails the assessed test condition and the AEB test overall, whether CIB or DBS. In the event of subject vehicle-to-principal other vehicle contact, testing will cease for the test condition, respective test scenario, the AEB test being performed (

i.e.,

CIB or DBS), and the AEB assessment overall.

NHTSA will also continue to conduct the false positive

test scenario currently used in NCAP, but has modified the test conditions and requirements for passing performance. This test scenario evaluates the propensity of a vehicle's DBS system to activate inappropriately in a non-critical driving scenario that would not present a safety risk to the vehicle's occupants. A vehicle must pass each of the 19 required CIB test conditions to obtain credit for CIB and must also separately pass each of the 17 required DBS test conditions to obtain credit for DBS.

CIB and DBS systems are collectively known as automatic emergency braking (AEB).

Trial or test trial is a test among a set of tests conducted under the same test conditions (including test speed) with the same subject vehicle.

In essence, because the Agency will provide an overall assessment for AEB performance, if a vehicle fails a trial run in the DBS test, testing will cease for the DBS assessment, and CIB assessments will not be conducted because the vehicle will have failed the AEB assessment overall.

For purposes of this document, NHTSA uses “false positive” and “false activation” interchangeably, and the Agency intends for them to refer to the same situations.

NHTSA is consolidating forward collision warning (FCW) testing to assess and evaluate FCW functionality during CIB and DBS testing in all test scenarios except NHTSA's false positive tests. For evaluations during CIB and DBS testing, the test vehicle must issue an FCW prior to the onset of automatic braking (as defined by the instant the subject vehicle deceleration reaches at least 0.15g) for the vehicle to pass each test trial run conducted as part of NCAP's CIB and DBS testing. If the required FCW is not issued prior to the onset of automatic braking imparted by CIB, the vehicle will fail the test trial and CIB/DBS assessment overall. NHTSA will conduct the AEB evaluation by (1) fully releasing the subject vehicle's accelerator pedal (at any rate) within 500 milliseconds (ms) after an FCW is issued (during CIB and DBS evaluations, and whether before or after automatic braking has begun), and (2) initiating manual (robotic) brake application at a time that corresponds to 1.0 ± 0.1 seconds after issuance of the required FCW signals (during DBS evaluations). A FCW must be presented to the vehicle operator via a minimum of two sensory modalities to receive credit in each of NCAP's CIB and DBS tests (except for the false positive test). A vehicle must present, at a minimum, an FCW comprised of visual and auditory signals. Finally, Revision G of the AB Dynamics (ABD) Global Vehicle Target (GVT) will be used as the principal other vehicle in NCAP testing instead of the currently used Strikable Surrogate Vehicle (SSV) test device. Other details of the test conditions and response to comments on updating CIB, DBS, and FCW evaluations are provided in relevant sections in this notice.

Table 1—Adopted CIB Test Scenarios and Conditions

Test no.

Test scenario

SV speed (kph (mph))

POV speed (kph (mph))

POV headway (m (ft.))

POV

deceleration (g)

Requirement to pass

LVS

40 (24.9)

n/a

No SV-to-POV contact during any test trial.

50 (31.1)

n/a

60 (37.3)

n/a

70 (43.5)

n/a

80 (49.7)

n/a

LVM

40 (24.9)

20 (12.4)

n/a

50 (31.1)

20 (12.4)

n/a

60 (37.3)

20 (12.4)

n/a

70 (43.5)

20 (12.4)

n/a

80 (49.7)

20 (12.4)

n/a

LVD

50 (31.1)

40 (131.2)

0.3

50 (31.1)

12 (39.4)

0.3

80 (49.7)

40 (131.2)

0.3

80 (49.7)

12 (39.4)

0.3

50 (31.1)

40 (131.2)

0.5

50 (31.1)

12 (39.4)

0.5

80 (49.7)

40 (131.2)

0.5

80 (49.7)

12 (39.4)

0.5

False Positive (STP)

80 (49.7)

n/a

SV peak deceleration <0.25g

Table 2—Adopted DBS Test Scenarios and Conditions

Test no.

Test scenario

SV speed (kph (mph))

POV speed (kph (mph))

POV headway (m (ft.))

POV

deceleration (g)

Requirement to pass

LVS

70 (43.5)

n/a

No SV-to-POV contact during any test trial.

80 (49.7)

n/a

90 (55.9)

n/a

100 (62.1)

n/a

LVM

70 (43.5)

20 (12.4)

n/a

80 (49.7)

20 (12.4)

n/a

90 (55.9)

20 (12.4)

n/a

100 (62.1)

20 (12.4)

n/a

LVD

50 (31.1)

40 (131.2)

0.3

50 (31.1)

12 (39.4)

0.3

80 (49.7)

40 (131.2)

0.3

80 (49.7)

12 (39.4)

0.3

50 (31.1)

40 (131.2)

0.5

50 (31.1)

12 (39.4)

0.5

80 (49.7)

40 (131.2)

0.5

80 (49.7)

12 (39.4)

0.5

False Positive (STP)

80 (49.7)

n/a

SV peak deceleration <0.25g over the baseline peak imparted by manual braking.

Adding Pedestrian Automatic Emergency Braking Evaluation

NHTSA is adding the evaluation of pedestrian automatic emergency braking (PAEB) to NCAP using four crossing test scenarios and two in-path test scenarios to evaluate PAEB in daylight and darkness lighting conditions with no overhead lights. For the crossing scenarios (S1), a walking adult or running child pedestrian mannequin crosses perpendicular to the vehicle's line of travel from either the driver's left or right side. For the in-path scenarios (S4), an adult pedestrian mannequin is slightly overlapped with the front of the vehicle and is either facing away while standing in front of the vehicle, or walking away from the vehicle, parallel to the flow of traffic.

The subject vehicle's lower beam headlamps will be used during all NCAP PAEB testing in dark lighting conditions, and the upper beam headlamps will not be engaged either manually or automatically by way of an advanced lighting system, such as adaptive driving beams, unless such a system cannot be deactivated. This requirement will apply even to those systems that are active by default when low beam headlamps are first engaged. The performance criterion for NCAP's PAEB tests will be no contact with the pedestrian mannequin. The 4activePA Adult and 4activePA Child pedestrian test mannequins (articulating mannequins) will be used for NCAP's PAEB evaluation.

NHTSA will test for each of the adopted PAEB test conditions at a minimum subject vehicle speed threshold of 10 kph (6.2 mph), increasing the subject vehicle speed in 10 kph (6.2 mph) increments until the maximum speed threshold is reached, so long as the test vehicle does not contact the pedestrian mannequin during each progressive speed tested. For test conditions S1a, S1b, S1e, S4a, and S4c, the Agency is adopting a maximum subject vehicle speed threshold of 60 kph (37.3 mph) for both daylight and darkness testing. For test condition S1d, NHTSA is adopting a maximum subject vehicle speed threshold of 60 kph (37.3 mph) for daylight testing and 40 kph (24.9 mph) for darkness testing. Should the subject vehicle contact the pedestrian mannequin during the initial run for any test speed, testing will cease for the test condition, respective test scenario, and PAEB testing overall for the particular lighting condition. Only one trial will be conducted per test condition and vehicles must pass all required tests (

i.e.,

no contact with pedestrian mannequin) to receive PAEB credit for the relevant lighting condition.

An overview of test scenarios and test parameters (pedestrian size, test speed, pedestrian motion, overlap, and obstruction) is provided in Tables 3 and 4.

Table 3—Adopted NCAP PAEB Daylight Test Conditions and Variants

Test condition

Size

Movement classification

Path

origin

Overlap

(%)

Obstruction

Test no.

Test speeds

(kph (mph))

Pedestrian

S4c

Adult (Facing Away)

Walk

Right

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S4a

Adult (Facing Away)

Stationary

Right

10 (6.2)

20 (12.4)

30 (18.6)

40 (24.9)

50 (31.1)

60 (37.3)

S1b

Adult

Walk

Right

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S1a

Adult

Walk

Right

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S1e

Adult

Run

Left

10 (6.2)

8 (5.0)

20 (12.4)

8 (5.0)

30 (18.6)

8 (5.0)

40 (24.9)

8 (5.0)

50 (31.1)

8 (5.0)

60 (37.3)

8 (5.0)

S1d

Child

Run

Right

Yes

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

Table 4—Adopted NCAP PAEB Darkness Test Conditions and Variants

Test condition

Size

Movement

classification

Path

origin

Overlap

(%)

Obstruction

Test no.

Test speeds (kph (mph))

Pedestrian

S4c

Adult (Facing Away)

Walk

Right

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S4a

Adult (Facing Away)

Stationary

Right

10 (6.2)

20 (12.4)

30 (18.6)

40 (24.9)

50 (31.1)

60 (37.3)

S1b

Adult

Walk

Right

50%

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S1a

Adult

Walk

Right

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

50 (31.1)

5 (3.1)

60 (37.3)

5 (3.1)

S1e

Adult

Run

Left

10 (6.2)

8 (5.0)

20 (12.4)

8 (5.0)

30 (18.6)

8 (5.0)

40 (24.9)

8 (5.0)

50 (31.1)

8 (5.0)

60 (37.3)

8 (5.0)

S1d

Child

Run

Right

Yes

10 (6.2)

5 (3.1)

20 (12.4)

5 (3.1)

30 (18.6)

5 (3.1)

40 (24.9)

5 (3.1)

* All darkness testing to occur without the use of overhead artificial lighting.

Adding Blind Spot Warning (BSW) and Blind Spot Intervention (BSI) Evaluation

This notice adds assessments for two blind spot technologies, BSW and BSI, to NCAP's crash avoidance program. Blind spot warning (BSW) and blind spot intervention (BSI) will be evaluated separately in individual tests conducted in daylight with the principal other vehicle on the left and right side of the subject vehicle, with the subject vehicle turn signal indicator activated and not activated. BSW will be evaluated using tests representing the Straight Lane Converge and Diverge and Straight Lane Pass-by scenarios,

using an actual vehicle (representing a high production mid-size passenger car) as the principal other vehicle. For tests where the turn signal is not activated, a visual warning signal in the side mirror or the A-pillar must be issued within a specified time as detailed in the BSW test procedure. For tests where the turn signal is activated, an additional warning modality (

i.e.,

a dual-modality warning) or an escalating visual warning signal (

e.g.,

switches from steady-burning to flashing) is required within the time specified in the BSW test procedure.

The two scenarios for assessing BSW were proposed in the March 2022 RFC notice and are described in a later section of this notice.

For the BSW Straight Lane Converge and Diverge scenario, the test speed for both the subject vehicle and principal other vehicle will be 72.4 kph (45.0 mph). For the BSW Straight Lane Pass-by scenario, NHTSA will conduct the lowest speed differential condition (subject vehicle/principal other vehicle speeds of 72.4/80.5 kph (45.0/50.0 mph)) first. If the subject vehicle issues a passing BSW during the run, the principal other vehicle speed will be incrementally increased by 8.0 kph (5.0 mph) and testing will continue with one run conducted per speed differential condition until a principal other vehicle speed of 104.6 kph (65.0 mph) is reached. Testing will then be repeated following a similar methodology for principal other vehicle movement on the opposite side of the subject vehicle. If, for any speed differential condition, the subject vehicle does not issue a passing BSW, NHTSA will discontinue BSW testing for that vehicle model. Only one trial per BSW test condition will be conducted. An overview of the test scenarios and test parameters for the BSW tests is presented in Table 5. To obtain credit for BSW, the vehicle must pass all 20 tests for BSW.

Table 5—Blind Spot Warning (BSW) Adopted Test Conditions

Test scenario

SV speed (kph (mph))

POV speed (kph (mph))

POV direction of

approach

Turn signal

Straight Lane

Converge and Diverge

72.4 (45)

Right

Enabled

Disabled

Left

Enabled

Disabled

Straight Lane Pass-by

72.4 (45)

80.5 (50)

Right

Enabled

Disabled

Left

Enabled

Disabled

88.5 (55)

Right

Enabled

Disabled

Left

Enabled

Disabled

96.6 (60)

Right

Enabled

Disabled

Left

Enabled

Disabled

104.6 (65)

Right

Enabled

Disabled

Left

Enabled

Disabled

BSI will be evaluated using tests representing two lane change scenarios (Subject Vehicle Lane Change with Constant Headway and Subject Vehicle Lane Change with Closing Headway) and one false positive scenario (Subject Vehicle Lane Change with Constant Headway False Positive Assessment),

using Revision G of the ABD GVT as the principal other vehicle. All BSI evaluations will be conducted with adaptive cruise control (ACC), lane centering assistance (LCA), and/or lane keeping assist (LKA) technologies (if equipped and if the systems can be disengaged) turned off.

These three scenarios for assessing BSI were proposed in the March 2022 RFC notice and are described in a later section of the notice.

For the BSI Subject Vehicle Lane Change with Constant Headway and the False Positive tests, the test speed for both the subject vehicle and principal other vehicle will be 72.4 kph (45.0 mph). For the BSI Subject Vehicle Lane Change with Closing Headway tests, the subject vehicle test speed will be 72.4 kph (45.0 mph) and the principal other vehicle speed will be 80.5 kph (50 mph). In these tests, after a short period of steady-state driving, the subject vehicle driver (

i.e.,

robot) initiates a lane change and follows an 800 m (2,625 ft.) radius curved path towards the principal other vehicles' travel lane. The subject vehicle driver then releases the steering wheel upon the subject vehicle exiting the curve so as to achieve a steady state lateral velocity of 0.7 ± 0.1 m/s (2.3 ± 0.3 ft./s) relative to the line separating the subject vehicle and principal other vehicle travel lanes. Each test scenario is conducted with turn signal enabled and disabled and for both left and right lane change directions.

To pass the Subject Vehicle Lane Change with Constant Headway and the Subject Vehicle Lane Change with Closing Headway tests, the BSI system must prevent any contact between the subject vehicle and the principal other vehicle. The subject vehicle BSI intervention must not cause a secondary departure on the opposite side of the lane. To pass a false positive test, the BSI system must not intervene. Only one trial per BSI test condition will be conducted. An overview of the test scenarios and test parameters for the BSI tests is presented in Table 6. To obtain credit for BSI, the vehicle must pass all 12 tests.

Table 6—Blind Spot Intervention (BSI) Adopted Test Conditions

Test scenario

SV speed

(kph (mph))

POV speed

(kph (mph))

Lane change

direction

Turn signal

SV Lane Change with Constant Headway

72.4 (45)

Left

Enabled.

Disabled.

Right

Enabled.

Disabled.

SV Lane Change with Closing Headway

72.4 (45)

80.5 (50)

Left

Enabled.

Disabled.

Right

Enabled.

Disabled.

SV Lane Change with Constant Headway, False Positive Assessment

72.4 (45)

Left

Enabled.

Disabled.

Right

Enabled.

Disabled.

Adding Lane Keeping Assist (LKA) and Enhancing Lane Departure Warning (LDW) Evaluation

NHTSA is adding the assessment of lane keeping assist (LKA) into NCAP and integrating the evaluation of lane departure warning (LDW) with the LKA evaluation. To evaluate a vehicle's LDW sensitivity and LKA intervention capabilities, NHTSA's testing includes the use of a single solid white lane line, dashed yellow lane line, or Botts' dots (raised pavement markers) on either the right or left side of the vehicle's travel lane, depending on testing direction. Additional tests will be conducted with two lane lines (solid yellow and dashed white lines, and dashed white and solid white lines) to evaluate a vehicle's ability to properly correct its heading to prevent a secondary lane departure after the initial intervention. For the LDW/LKA tests, the subject vehicle, traveling at a speed of 72.4 kph (45 mph), heads towards the lane line using an initial path defined by a 1,200 m (3,937 ft.) radius curve. Tests will be conducted by incrementing the lateral velocity of the subject vehicle's approach toward the lane line from 0.2 to 0.6 m/s (0.7 to 2.0 ft./s) in 0.1 m/s (0.3 ft./s) increments.

To pass the criteria of the LDW/LKA evaluation test, the subject vehicle must issue a visual signal when the lateral position of the vehicle, represented by a two-dimensional polygon, is within 0.75 m (2.5 ft.) of the inboard edge of the lane line and before the lane departure exceeds 0.3 m (1 ft.). The LKA intervention itself will serve as a secondary haptic alert component. Neither an LDW nor LKA intervention shall occur when a vehicle has not departed its lane and is farther than 0.75 m (2.5 ft.) from the inboard edge of the lane line. In addition, the visual warning signal and LKA intervention must be issued before the lane departure exceeds 0.3 m (1 ft.), and the visual alert must be issued prior to, or concurrent with, the start of the LKA intervention. Only one trial per test condition is conducted. An overview of the test scenarios and test parameters for the LDW/LKA tests is presented in Table 7. To obtain credit for LDW and LKA, the vehicle must pass all 50 tests performed during the LDW/LKA performance assessment.

Table 7—Lane Departure Warning (LDW)/Lane Keeping Assist (LKA) Adopted Test Conditions

Test scenario

Line type

Departure direction

Lateral

velocity

(m/s (ft./s))

Passing criteria

Maximum SV

excursion

(m (ft.))

LDW alert issued

(m (ft.))

Primary Departure

(Single Straight Lane Line)

Solid White

Left

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

−0.3 (−1.0)

0.75 to −0.3

(2.5 to −1.0).

Solid White

Right

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Dashed Yellow

Left

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Dashed Yellow

Right

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Raised Pavement Markers

Left

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Raised Pavement Markers

Right

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Secondary Departure

(Dual Straight Lane Line)

Solid Yellow (L)/Dashed White (R)

Left

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

−0.3 (−1.0)

0.75 to −0.3

(2.5 to −1.0).

Solid Yellow (L)/Dashed White (R)

Right

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Dashed White (L)/Solid White (R)

Left

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

Dashed White (L)/Solid White (R)

Right

0.2 (0.7)

0.3 (1.0)

0.4 (1.3)

0.5 (1.6)

0.6 (2.0)

NCAP Roadmap 2024-2033

NHTSA has developed a final roadmap to update NCAP through multiple phases from 2024 through 2033, with mid-term roadmap items spanning the period 2024-2028, and long-term items spanning the period 2024-2033. The NCAP roadmap includes four phases for each NCAP initiative, along with a completion milestone for each phase. The four phases are: (1) Research phase, if applicable, (2) Request for comment (RFC) phase, (3) Final decision phase, and (4) Implementation phase. NHTSA plans updates to NCAP in the following three safety programs: crashworthiness, crash avoidance, and vulnerable road user safety. A summary of the mid-term and long-term actions for this roadmap is presented in Tables 8 and 9, respectively. The timeframe shown for the research, RFC, and final decision phases is in calendar years. The start of the implementation phase is in the fourth quarter of the calendar year shown in the two tables. Note that the implementation phase starts with vehicle models of the following calendar year shown in Tables 8 and 9. NHTSA plans to update the NCAP roadmap approximately every four years, with timelines updated accordingly. Details of the NCAP roadmap are provided in the roadmap section of this notice.

Table 8—Roadmap for Mid-Term Upgrades to NCAP

[In calendar years]

Potential updates to NCAP

evaluations

Research phase

RFC phase

Final

decision phase

Implementation

phase start

in 4th quarter

Crash Avoidance Program:

Enhanced FCW, CIB, DBS

2023-2024

2025

LDW+LKA and BSW+BSI

2023-2024

2025

Rear Automatic Braking

2024

2025

2025-2026

2027

Crashworthiness Program:

THOR-50M in Frontal Crash Tests and HIII-05F * in Driver Position in Frontal Rigid Barrier Crash Test

2024

2024-2025

2025-2026

2027

Frontal Oblique Crash Test with THOR-50M

2024

2024-2025

2025-2026

2027

WorldSID-50M in Side Impact Tests, and SID-IIs ** Rib Deflections for Injury Risk Assessment

2024

2024-2025

2025-2026

2027

Vulnerable Road User (VRU) Safety Program:

PAEB (day and night-time)

2023-2024

2025

Crashworthiness Pedestrian Protection

2023-2024

2025

Unattended Child Alert System (Availability of Direct Sensing Technologies Noted in Safety Features Section on Ratings Webpage)

2024

Bicyclist and Motorcyclist AEB (along path scenarios)

2024-2025

2025

2025-2026

2027

Vehicle Safety Rating:

Rating System for Crash Avoidance Technologies

2024-2025

2027

Rating Systems for Crashworthiness, VRU Safety, and Overall Safety

2024-2025

2025-2026

2027

Monroney Label Rulemaking—Crash Avoidance, Crashworthiness, VRU Safety, and Overall Safety Ratings

2023-2024

2025

2025-2026

2027

* The advanced 5th percentile female frontal impact test dummy, THOR-05F, is currently under evaluation/refinement and is included in the long-term NCAP update in this roadmap. Until THOR-05F is completed and included in NCAP, NHTSA will use the current HIII-05F dummy in frontal crash tests.

** The advanced 5th percentile female side impact test dummy, WorldSID-05F, is currently under development and its use in NCAP will be considered in the long-term section of this roadmap. Until WorldSID-05F is included in NCAP, the SID-IIs will be used in NCAP along with thoracic and abdominal deflection measurements.

Table 9—Roadmap for Long-Term Upgrades to NCAP

[In calendar years]

Potential updates to NCAP

Evaluations

Research phase

RFC phase

Final

decision phase

Implementation

phase start

in 4th quarter

Crash Avoidance Program:

Headlighting System (Advanced Driving Beam, Semi-Automatic Beam Switching, and Lower Beam Headlamp)

2024-2026

2026-2027

2028

2030

AEB for Intersection Crash Scenarios

2025-2027

2028

2029

2031

Enhanced LKA (Higher Speed, Curved Road and/or Road Edge Detection Scenarios)

2024-2026

2027

2028

2030

Enhanced AEB (Speed and Additional Scenarios)

2026-2028

2029

2030

2032

Driver Monitoring Systems—Distracted/Drowsy Driving

2023-2027

2028

2029

2031

Intelligent Speed Assist

2024-2028

Crashworthiness Program:

THOR-05F in Frontal Crash Tests in Front and Rear Seating Positions

2023-2027

2027-2028

2028-2029

2031

WorldSID-05F in Side Impact Crash Tests

2023-2029

2029-2030

2030-2031

2033

VRU Safety Program:

Enhanced AEB for Bicyclists and Motorcyclists in Intersection Crashes

2025-2026

2027

2028

2030

BSW and BSI Evaluation for Bicyclists and Motorcyclists Crash Protection

2025-2026

2027

2028

2030

Crashworthiness Pedestrian Protection using aPLI *

2024-2025

2026

2027

2029

Enhanced PAEB (Speed and Additional Scenarios)

2026-2028

2029

2030

2032

Driver Visibility

2023-2027

* aPLI is the advanced pedestrian legform impactor. It assesses pedestrian injuries to the knee, upper leg, and lower leg in impacts with the front of vehicles.

III. Background

The National Highway Traffic Safety Administration's (NHTSA's) New Car Assessment Program (NCAP) supports the Agency's mission to reduce the number of fatalities and injuries that occur on U.S. roadways by providing important vehicle safety information to consumers to inform their purchasing decisions. The last major NCAP upgrade occurred on July 11, 2008, and took effect with model year 2011 vehicles.

That program update included the Agency's adoption of new frontal and side anthropomorphic test devices (crash test dummies) and associated injury criteria, a new oblique side pole test, and a new overall rating system combining the individual frontal, side, and rollover ratings. NHTSA also expanded NCAP to include assessment of three advanced driver assistance systems (ADAS) technologies: forward collision warning (FCW), lane departure warning (LDW), and electronic stability

control (ESC).

Through that expansion, the Agency began to identify which vehicles were equipped with these technologies and met specified performance requirements, making this information available on the NHTSA website. In November 2015, NHTSA also added crash imminent braking (CIB) and dynamic brake support (DBS) technologies (also known as automatic emergency braking, or AEB technology) to its ADAS assessments, with implementation beginning with model year 2018 vehicles.

73 FR 40016 (July 11, 2008).

ESC was removed from the Agency's list of recommended ADAS technologies through NCAP beginning in model year 2014 when the technology became mandated under Federal motor vehicle safety standard (FMVSS) No. 126, “Electronic stability control.” NHTSA also included rear video systems in its list of recommended technologies under NCAP from model years 2014 to 2017 and removed that technology from its list when it became mandated under FMVSS No. 111, “Rear visibility.”

80 FR 68604 (Nov. 5, 2015).

In December 2015, the Agency published a Request for Comments (RFC) notice with planned changes to the overall NCAP program. The notice sought comment on NCAP's potential use of enhanced tools and techniques to evaluate the safety of vehicles, generate star ratings, and encourage further vehicle safety developments.

The RFC notice also outlined planned changes for the crashworthiness, crash avoidance, and ratings categories. Many commenters responding to the December 2015 RFC notice stated it lacked sufficient detail and supporting information to allow for thorough review and comment. Commenters also expressed concern over test procedure repeatability and reproducibility based on the RFC notice's lack of detail, performance criteria, and non-standardized test devices. NHTSA hosted a public meeting in October 2018 to re-engage stakeholders and seek up-to-date input to help the Agency plan the future of NCAP.

80 FR 78521 (Dec. 16, 2015).

On March 9, 2022, NHTSA published an RFC notice proposing changes to NCAP in response to the comments received from the 2015 RFC and public meetings, which partially fulfills the Agency's obligations under the 2015 Fixing America's Surface Transportation (FAST) Act directive and recent mandates included in section 24213 of the November 2021 Bipartisan Infrastructure Law (BIL). The proposed changes include:

• Changes to test procedures and performance criteria, including an increase in stringency, for the four currently recommended ADAS technologies in NCAP (FCW, LDW, DBS, and CIB) to enhance evaluation of the systems' capabilities in current vehicle models, reduce test burden, and promote harmonization with other consumer information programs.

• The addition of four new ADAS technologies—blind spot warning (BSW), blind spot intervention (BSI), lane keeping assist (LKA), and pedestrian automatic emergency braking (PAEB)—to those currently recommended by NCAP and highlighted on the Agency's website. The Agency proposed to incorporate these four new ADAS technologies into NCAP because data indicates they satisfy NHTSA's four prerequisites for inclusion in the program: (1) a known safety need exists; (2) system designs (countermeasures) exist that can mitigate the safety problem; (3) existing or new system designs have the potential to improve safety; and (4) a performance-based objective test procedure exists that can assess system performance.

See NCAP Rating FAQ No. 07,

http://nhtsa.gov/ratings.

• A “roadmap” of the Agency's plans to update NCAP in phases over the next ten years, setting forth NHTSA's mid-term and long-term strategies for upgrading the program using a phased approach. The roadmap presents an

estimated

timeframe for the issuance of phased request for comment notices to incorporate various potential program components. However, NHTSA would only issue proposals to update the program as technologies mature and are considered ready for inclusion such that they meet the program's four prerequisites. Following each proposal, NHTSA will issue a final decision notice responding to the comments received and providing the Agency's decisions for that particular program update, as well as the lead time for implementation.

In addition to these proposed changes, the RFC notice proposed operational changes to streamline the information provided to consumers. Specifically, the Agency proposed a process for updating ADAS-related information provided to consumers to reflect changes made to vehicles during the middle of a model year. Further, although not explicitly proposed in the RFC notice, the Agency sought comment on the following topics to help guide future proposals:

• The Agency's plan to develop a new ADAS rating system for NCAP to provide consumers with improved data to compare and shop for vehicles and spur improved ADAS performance. NCAP currently assigns ratings to vehicles based on their performance in crashworthiness (

i.e.,

frontal and side impact) and rollover tests, but the program has no complementary rating system to differentiate performance among vehicles' crash avoidance systems. Instead, NCAP only recommends certain ADAS technologies to shoppers and identifies vehicles that offer the recommended technologies that pass the program's system performance criteria.

• Steps to include crash avoidance rating information on a vehicle's window sticker (

i.e.,

the Monroney label) at the point of sale, consistent with the 2015 FAST Act. The Agency noted that it is currently conducting consumer research to determine how best to present such information to maximize its effectiveness in informing vehicle purchasing decisions. NHTSA stated that it would consider the information gained from this research in conjunction with related comments received in response to the 2022 RFC to develop a proposal for a revised label, which will be detailed in a separate RFC notice.

• Expanding NCAP to include safety technologies that promote NHTSA's continuing efforts to combat unsafe driving or riding behaviors, such as speeding and drowsy, impaired, distracted, and unbelted driving, as well as safety technologies that may prevent unintentional human behavior that may result in injury or death, such as vehicular heatstroke.

• The Agency's ideas for updating several programmatic aspects of NCAP, including adding new crash test dummies, updating the rollover risk curve, and enhancing the presentation and dissemination of safety information provided to consumers to improve the program. More specifically, NHTSA requested comment on several potential ways to revise the 5-star safety ratings system, including adopting a points-based rating system concept, revising the baseline risk, and incorporating decimal or half-star ratings.

• Additional considerations that would allow NCAP to remain effective and relevant to improve vehicle safety, such as proposed updates to the NCAP website and the development of an NCAP database to modernize the operational aspects of the program, including a new vehicle information submission process for vehicle manufacturers.

Following publication of the March 2022 RFC notice, NHTSA received comments generally supportive of the Agency's proposal to adopt additional crash avoidance elements. Additional details of NHTSA's proposal for each of

the items listed above is provided in later sections.

Summary of General Comments on Proposed Updates to NCAP

NHTSA received over 4,000 comments in response to the March 2022 RFC notice.

The Agency received comments from a wide variety of commenters including safety advocacy groups, trade associations, vehicle manufacturers, suppliers and developers, government agencies and associations, test laboratories, an insurance company, a research/consulting firm, and individual members of the public. A summary of the commenters to the March 2022 RFC notice is provided in Table 10.

The March 2022 RFC notice requested comment on a number of topics, including emerging technologies, and potential future updates to NCAP, that have not been proposed, and thus are not addressed, in the present notice. Details of the comments received and the Agency's response to these comments will be provided in future RFC notices relevant to those topics.

Table 10—Commenters to the March 2022 NCAP RFC Notice

Category

Commenters

Safety Advocacy Groups

AAA Inc. (AAA), AARP, Advanced Mobility Group, Advocates for Highway and Auto Safety (Advocates), American Motorcyclist Association (AMA), Center for Auto Safety (CAS), Consortium for Constituents with Disabilities Transportation Task Force (CCD Transportation Task Force), Consumer Reports (CR), Insurance Institute for Highway Safety (IIHS), Kids and Car Safety (KAC), Families for Safe Streets (FSS), Intelligent Transportation Society of America (ITS America), Massachusetts Vision Zero Coalition, The League of American Bicyclists (The League), Vision Zero Network (VZN), and ZF Group.

Industry Trade Associations

Alliance for Automotive Innovation (Auto Innovators), Automotive Safety Council (ASC), Motor & Equipment Manufacturers Association (MEMA), Motorcycle Industry Council and Motorcycle Safety Foundation (MIC/MSF), National Automobile Dealers Association (NADA), Specialty Equipment Market Association (SEMA), The Lidar Coalition.

Vehicle Manufacturers

American Honda Motor Co., Inc. (Honda), BMW of North America, LLC (BMW), FCA US LLC (FCA), Ford Motor Company (Ford), General Motors (GM), Hyundai America Technical Center, Inc. (HATCI), Hyundai Motor North America (Hyundai), Mercedes-Benz, LLC, a division of Mercedes-Benz Automotive Group (Mercedes-Benz), North American Subaru, Inc. (Subaru), Rivian Automotive, LLC (Rivian), Tesla, Inc. (Tesla), Toyota Motor North America (Toyota).

Suppliers and Developers

Aptiv PLC (Aptiv), Bosch LLC (Bosch), DENSO Corporation (DENSO), Intel Corporation (Intel), Robert Vayyar, Velodyne Lidar, Inc. (Velodyne).

Government Agencies and Associations

American Association of State Highway and Transportation Officials (AASHTO), National Association of City Transportation Officials (NACTO), National Transportation Safety Board (NTSB), New York City Department of Transportation & New York City Department of Citywide Administrative Services, Vision Zero Task Force (NYC DOT/NYC DCAS, Vision Zero Task Force).

Test Laboratories

Applus IDIADA (IDIADA), Dynamic Research Inc. (DRI), Transportation Research Center, Inc. (TRC).

Insurance Companies

State Farm Insurance Companies (State Farm).

Research/Consulting Companies

Safe Streets Research & Consulting (Safe Streets).

General Public

Individuals.

Many commenters stated they support NHTSA's goal of providing comparative safety information to consumers through NCAP but encouraged the Agency to further leverage NCAP to ensure consumers have a comprehensive understanding of vehicle safety. Commenters also stated that more testing, rating, and information-sharing with consumers about the functionalities of advanced safety technologies via NCAP will promote the technologies' use in future vehicles and advance vehicle safety. The Alliance for Automotive Innovation (Auto Innovators) stated it supports NHTSA's efforts to modernize NCAP, noting that a key benefit of a well-developed and technically robust NCAP is the introduction of advanced safety technologies and performance evaluation through market incentives (objective ratings) in a structured and predictable manner via the development of testing procedures, evaluation metrics, and safety benefits. Auto Innovators stated that doing so will lead to more vehicles being equipped with new ADAS technologies, ultimately decreasing technology costs, and bolstering the case for potential regulation.

Many commenters, including those who have lost loved ones in automotive accidents, expressed support for the proposed NCAP changes but stated they do not go far enough, noting the U.S. program is behind other countries' NCAP programs in updating, implementing, and “standardizing” NCAP with proven safety technologies to save more lives. The Advocates for Highway and Auto Safety (Advocates), the Consumer Federation of America, and many others expressed concern that the approach described in the March 2022 RFC is not sufficiently specific and lacks firm commitments on key updates to the program. Further, the National Safety Council (NSC) stated that its data continues to suggest NHTSA is not doing enough to address roadway safety, with thousands of people dying in preventable crashes each year.

Several commenters pointed out that fatalities involving pedestrians and cyclists have been increasing at alarming rates and urged NHTSA to consider the safety of those outside of the vehicle. Many cyclist organizations stated that any NCAP updates should include cyclist AEB testing and tests aimed at increasing cyclist safety. One individual noted the more than 50 percent increase in annual pedestrian fatalities in the past decade, stating that safety innovations are benefiting those inside, not outside, of the vehicle. Many other individual commenters expressed concern for the safety of pedestrians, mobility device users, and cyclists amidst rising fatalities from increasingly large vehicles, suggesting that NHTSA should consider promoting technologies and performing tests to protect them.

Many commenters expressed support for the Agency's proposed inclusion of the four new ADAS technologies and for enhancing the evaluation of ADAS technologies currently in NCAP. However, some commenters argued the proposal did not go far enough, suggesting that ADAS technologies (including PAEB) should not just be part of the voluntary NCAP program but should be mandatory on new vehicles to reduce fatalities, especially in urban areas. Many commenters also requested that NHTSA harmonize test procedures and evaluations with other existing testing protocols like Euro NCAP. While commenters generally supported the proposed roadmap, some noted that it

lacked sufficient specificity on future timelines, dates, and required actions. Commenters requested periodic stakeholder engagement and collaboration for developing future NCAP roadmaps.

Several commenters also provided detailed responses to questions NHTSA posed in the March 2022 RFC notice to help guide its decisions. The following sections discuss in detail: (1) NHTSA's proposal for each technology, (2) the Agency's response to the comments received to the questions posed, and (3) final decisions on the proposed changes to NCAP specific to the technology.

IV. Updating Forward Collision Prevention Technologies

NHTSA is updating assessments for FCW and AEB technologies (

i.e.,

CIB and DBS) in NCAP's crash avoidance program. As discussed in NHTSA's March 2022 RFC notice, these technologies, designed to address forward collisions (rear-end crashes), have the potential to help prevent or mitigate rear-end pre-crash scenarios representing approximately 1.7 million crashes annually, or 29 percent of all crashes that currently occur on U.S. roadways.

These crashes result in 1,275 fatalities, on average, and 883,386 MAIS 1-5 injuries annually, representing 3.8 percent of all fatalities and 32 percent of all injuries, respectively.

FCW and AEB technologies have proven effective in reducing crashes, fatalities, and injuries. For instance, as discussed in the March 2022 RFC notice, the University of Michigan Transportation Research Institute (UMTRI) found that for 3.8 million model year 2013-2017 GM vehicles, camera-based FCW systems produced an estimated 21 percent reduction in rear-end striking crashes, while the AEB systems studied (which included a combination of camera-only, radar-only, and fused camera-radar systems) produced an estimated 46 percent reduction in the same crash type.

These findings align with a 2017 Insurance Institute for Highway Safety (IIHS) study,

which concluded that vehicles equipped with FCW and AEB showed a 50 percent reduction for the same crash type.

Wang, J.-S. (2019, March),

Target crash population for crash avoidance technologies in passenger vehicles

(Report No. DOT HS 812 653). Washington, DC: National Highway Traffic Safety Administration.

The Abbreviated Injury Scale (AIS) is a classification system for assessing impact injury severity. AIS ranks individual injuries by body region on a scale of 1 to 6 where 1=minor, 2=moderate, 3=serious, 4=severe, 5=critical, and 6=maximum (untreatable). MAIS represents the maximum injury severity, or AIS level, recorded for an occupant (

i.e.,

the highest single AIS for a person with one or more injuries).

Leslie, A.J., Kiefer, R.J., Meitzner, M.R., & Flannagan, C.A. (2019),

Analysis of the field effectiveness of General Motors production active safety and advanced headlighting systems,

The University of Michigan Transportation Research Institute and General Motors LLC. UMTRI-2019-6.

Cicchino, J.B. (2017, February), Effectiveness of forward collision warning and autonomous emergency braking systems in reducing front-to-rear crash rates,

Accident Analysis and Prevention,

2017 Feb;99(Pt A):142-152.

https://doi.org/10.1016/j.aap.2016.11.009.

Low-speed AEB showed a 43% reduction.

Until these technologies are standard equipment on all passenger vehicles, it is important for NCAP to continue to recommend FCW and AEB technologies to consumers and to inform them which vehicles have FCW and AEB technologies meeting NHTSA's performance criteria. Further, given recent increases in the penetration of FCW and AEB technologies in the vehicle fleet and improvements to sensors, now is an opportune time to increase the stringency of the current NCAP performance requirements for these technologies.

A. Forward Collision Warning (FCW)

FCW systems use forward-looking sensors (

e.g.,

radar, lidar, camera systems, or a combination thereof) that detect objects (

e.g.,

vehicles, pedestrians) in front of vehicles and issue an alert to the driver. An FCW system uses the sensors' input to determine the speed of the object in front of it and the distance between the vehicle and the object. If the sensing system determines that the closing distance and velocity

between the driver's vehicle and the object in front of it is such that a collision may be imminent, the warning system is designed to induce an immediate crash avoidance response by the vehicle operator. Warning systems in use today provide drivers with a visual display, such as an illuminated telltale on or near the instrument panel, an auditory signal (

e.g.,

buzzer or chime), and/or a haptic signal that provides tactile feedback (

e.g.,

rapid vibrations of the seat pan or steering wheel). These signals warn the driver of an impending collision so the driver may then manually intervene (

e.g.,

apply the vehicle's brakes or make an evasive steering maneuver) to avoid or mitigate a crash. FCW systems alone do not brake the vehicle.

Closing velocity is the rate at which two objects are getting closer to each other.

NHTSA added FCW systems to its NCAP ADAS evaluations in 2008 (with performance evaluations beginning with model year 2011 vehicles) because these systems met the Agency's four prerequisites for inclusion at the time.

In its March 2022 RFC notice, the Agency proposed to continue to include FCW assessments in NCAP, as it found FCW systems to be effective, well-accepted by consumers, and widely available in the current vehicle fleet. For example, in its 2017 study, IIHS

found that FCW systems reduced rear-end crashes by 27 percent. Further, in a 2019 survey of more than 57,000 Consumer Reports subscribers, 69 percent of vehicle owners reported they were satisfied with their vehicle's FCW technology.

Currently, manufacturer-submitted data collected by NHTSA indicates 91 percent of model year 2023 vehicles are equipped with FCW systems as standard equipment.

73 FR 40033 (July 11, 2008).

Cicchino, J.B. (2017, February), Effectiveness of forward collision warning and autonomous emergency braking systems in reducing front-to-rear crash rates,

Accident Analysis and Prevention,

2017 Feb;99(Pt A):142-152.

https://doi.org/10.1016/j.aap.2016.11.009.

Consumer Reports (2019, November),

Consumer Perception of ADAS, https://data.consumerreports.org/reports/consumer-perceptions-of-adas/.

NCAP's Current Forward Collision Warning Test Procedure

The Agency included FCW as a recommended technology in NCAP and conducted performance evaluations beginning with model year 2011 vehicles. The FCW test procedure adopted at that time is still in use in the Agency's testing today.

73 FR 40016 (July 11, 2008).

Currently, NCAP's FCW test procedure

consists of three scenarios that simulate the most frequent types of rear-end crashes. These include lead vehicle stopped (LVS), lead vehicle decelerating (LVD), and lead vehicle moving (LVM) scenarios. In each scenario, the vehicle being evaluated is called the subject vehicle (SV). The SV is driven by a professional driver who provides the necessary acceleration, braking, and steering inputs during the test. A production mid-size passenger car, designated as the principal other vehicle (POV) during testing, is positioned directly in front of the SV and is also driven by a professional driver. Time-to-collision (TTC) criteria are prescribed for each FCW scenario. The TTC for each scenario is calculated by considering the speed of the SV relative to the POV at the time of the FCW. If the FCW system fails to issue a warning within the required time during testing, the SV's professional test

driver brakes, or steers away, to avoid a collision with the POV. A short description of each test scenario and the requirements for a passing result based on the prescribed TTC is provided below:

National Highway Traffic Safety Administration. (2013, February).

Forward collision warning system confirmation test. https://www.regulations.gov.

Docket No. NHTSA-2006-26555-0134.

• LVS—The SV encounters a stopped POV directly in front of it on a straight road. The SV is moving at 72.4 kph (45 mph), and the POV is stationary. To pass this test, the SV must issue an FCW when the TTC is at least 2.1 s. See Figure 1 (below) for a scenario diagram.

EN03DE24.000

• LVD—The SV encounters a POV slowing with constant deceleration directly in front of it on a straight road. The SV and POV are both driven at 72.4 kph (45 mph) with an initial headway of 30.0 m (98.4 ft.). The POV then decelerates, braking at a constant deceleration of 0.3g in front of the SV. To pass this test, the SV must issue an FCW when the TTC is at least 2.4 s. See Figure 2 (below) for a scenario diagram.

EN03DE24.001

• LVM—The SV encounters a slower-moving POV directly in front of it on a straight road. The SV and POV are driven at constant speeds of 72.4 kph (45 mph) and 32.2 kph (20 mph), respectively. To pass this test, the SV must issue an FCW when the TTC is at least 2.0 s. See Figure 3 (below) for a scenario diagram.

EN03DE24.002

Each of these three scenarios is conducted up to seven times. To pass the NCAP FCW system performance tests, the SV must satisfy the respective TTC-based performance criteria for at least five out of seven trials

for each of the three test scenarios.

As noted in the Agency's 2015 AEB final decision notice (80 FR 68618 (Nov. 5, 2015)), a pass rate of five out of seven tests per scenario was adopted for NCAP's current FCW test protocol to provide the Agency with a way to encourage system robustness without precluding the proliferation of emerging technologies offering the potential for significant safety benefits.

B. Automatic Emergency Braking (AEB)

One limitation of FCW systems is that, although designed to warn the driver of an impending rear-end crash, they do not actively and automatically assist drivers with avoiding rear-end crashes or mitigating their severity. To address this, CIB and DBS (known collectively as AEB) both provide significant automatic braking of the vehicle.

DBS systems provide supplemental braking when sensors determine that driver-applied braking is insufficient to avoid an imminent crash. CIB systems provide automatic braking when forward-looking sensors indicate that a crash is imminent, and the driver has not braked.

Some FCW systems use haptic brake pulses to alert the driver of a crash-imminent driving situation, but they are not intended to effectively slow the vehicle.

Research has shown that active safety systems such as AEB provide greater safety benefits than the corresponding warning systems alone, such as FCW. In its 2019 study, UMTRI

found that

AEB systems produced an estimated 46 percent reduction in applicable rear-end crashes when combined with a forward collision warning, which alone showed only a 21 percent reduction.

Like FCW systems, AEB systems are also well-accepted by consumers and widely available in the current vehicle fleet. In Consumer Reports' 2019 subscriber survey, 81 percent of owners of vehicles equipped with AEB reported they were satisfied with AEB technology.

Currently, manufacturer-submitted data collected by NHTSA indicates approximately 91 percent of model year 2023 vehicles are equipped with AEB systems as standard equipment. For these reasons, in 2015, NHTSA added CIB and DBS technologies to its ADAS assessments starting with model year 2018 vehicles, and why the Agency also proposed to continue to include AEB assessments in NCAP in its March 2022 RFC notice.

Leslie, A.J., Kiefer, R.J., Meitzner, M.R., & Flannagan, C.A. (2019, September),

Analysis of the field effectiveness of General Motors production

active safety and advanced headlighting systems,

The University of Michigan Transportation Research Institute and General Motors LLC, UMTRI-2019-6.

The AEB systems studied by UMTRI consisted of camera-only, radar-only, and fused camera-radar AEB systems, the latter two systems of which also included adaptive cruise control functionality.

Consumer Reports, (2019, November),

Consumer Perceptions of ADAS, https://data.consumerreports.org/reports/consumer-perceptions-of-adas/.

Docket No. NHTSA-2021-0002. 87 FR 13452. March 9, 2022.

Dynamic Brake Support (DBS)

Like FCW (and CIB) systems, DBS systems employ forward-looking sensors to detect vehicles in the path directly ahead while simultaneously monitoring the operational state of the driver's vehicle (

e.g.,

speed, the relative speed of and distance to the lead vehicle, driver inputs of steering and braking). In response to an FCW or an imminent crash, a driver may initiate braking to avoid a rear-end crash. However, research suggests that a driver's brake application may not take full advantage of the vehicle's foundation braking system in cases where the driver is inattentive, receives an FCW, and re-engages in the driving task prior to automatic braking (

i.e.,

CIB) taking place. In situations where the driver's braking is insufficient to prevent a collision, DBS can automatically supplement the driver's braking action to prevent or mitigate the crash.

The NCAP DBS performance evaluations serve to ensure that the vehicle's supplemental braking is sufficient to augment the driver's manual brake application and avoid a collision with the lead vehicle in the tested driving situations. DBS testing also endeavors to ensure that a vehicle's automatic brake application (

i.e.,

CIB) is not suppressed in the event of a driver's manual brake application.

DBS systems differ from traditional brake assist systems used with the vehicle's foundation brakes. Whereas both systems rely on brake pedal application rate to determine whether supplemental braking is required, DBS has a lower activation threshold since it also uses information from forward-looking sensors to verify that more braking is needed.

NCAP's Current Dynamic Brake Support Test Procedure

NCAP's current DBS test procedure

consists of the same three rear-end pre-crash scenarios specified in the FCW system performance test procedure: LVS, LVD, and LVM. However, most of the test speed combinations specified in the DBS test procedure differ. The single exception is that the FCW and DBS test procedures both use an LVM test performed with SV and POV speeds of 72.4 and 32.2 kph (45 and 20 mph), respectively. The DBS performance assessment also includes a Steel Trench Plate (STP) false positive suppression test conducted at two test speeds. The false positive suppression test series evaluates the ability of a DBS system to differentiate a steel trench plate, often found on roadways, from an object presenting a genuine safety risk to the SV. Although STPs are large and metallic, they are designed to be driven over without risk of injury to drivers or vehicles. This fourth test scenario is used to evaluate the propensity of a vehicle's DBS system to activate inappropriately in this non-critical driving scenario that would not present a safety risk to the vehicle's occupants.

National Highway Traffic Safety Administration (2015, October),

Dynamic brake support performance evaluation confirmation test for the New Car Assessment Program, http://www.regulations.gov,

Docket No. NHTSA-2015-0006-0026.

Like NCAP's FCW tests, the vehicle subjected to the DBS test scenarios is termed the SV. However, unlike NCAP's FCW tests, the DBS test procedure uses a surrogate vehicle (

i.e.,

a realistic looking artificial vehicle) as the POV instead of an actual vehicle to limit the potential for damage to the SV and/or the test equipment in the event of a collision. Additionally, instead of driver- (human-) based inputs, like those required in NCAP's FCW tests, a programmable (robotic) brake controller is used to provide all SV brake pedal applications made during the DBS system performance evaluations.

The Strikeable Surrogate Vehicle (SSV) is the surrogate vehicle presently used as the POV by NCAP for the Agency's DBS testing. The SSV, developed by NHTSA for the purpose of track testing, appears as a “real” vehicle to the camera, radar, and lidar sensors used by existing AEB systems. The SSV system is comprised of (a) a shell,

which is a visually and dimensionally accurate representation of a compact passenger car; (b) a slider and load frame assembly to which the shell is attached, (c) a two-rail track on which the slider operates, (d) a road-based lateral restraint track, and (e) a tow vehicle, which pulls the SSV and its peripherals down the test track during the test where the POV (

i.e.,

SSV) must be in motion.

The shell is constructed from lightweight composite materials with favorable strength-to-weight characteristics, including carbon fiber, Kevlar®, phenolic, and Nomex honeycomb. It is also wrapped with a commercially available vinyl material to simulate paint on the body panels, rear bumper, and a tinted glass rear window. A foam bumper having a neoprene cover is attached to the rear of the SSV to reduce the peak forces realized immediately after an impact from a test vehicle occurs.

For the three test scenarios where braking is expected, the SV must provide enough supplemental braking to completely avoid contact with the SSV (

i.e.,

POV) to pass a trial run. In the case of the DBS false positive test scenario, the performance criterion is minimal to no activation for both test speeds.

51 52

A short description of each DBS system performance test scenario, and the requirements for a passing result, is provided below:

Minimal activation is defined as a peak SV deceleration attributable to DBS intervention that is less than or equal to 1.5 times the average of the deceleration recorded for the vehicle's foundation brake system alone during its approach to the STP. The 1.5 multiplier serves to provide some system flexibility, meaning a mild DBS intervention is acceptable, but one where the vehicle thinks it must respond to the STP as if it was a real vehicle is not.

Note that the March 2022 notice specified a multiplier of 1.25. This specification was in error.

• Lead Vehicle Stopped (LVS)—The SV encounters a stopped POV directly in front of it on a straight road. The SV is moving at 40.2 kph (25 mph) and the POV is stationary. The SV throttle is released within 500 ms after the SV issues an FCW, and the SV brake pedal is manually applied at a TTC of 1.1 s (

i.e.,

at a nominal headway of 12.2 m (40 ft.)). To pass this test, the SV must not contact the POV. See Figure 1 for a scenario diagram.

• Lead Vehicle Decelerating (LVD)—The SV encounters a POV slowing with constant deceleration directly in front of it on a straight road. The SV and POV are both driven at 56.3 kph (35 mph) with an initial headway of 13.8 m (45.3 ft.). The POV brakes are then applied to achieve a constant deceleration of 0.3g in front of the SV. The SV throttle is released within 500 ms after the SV issues an FCW, and the SV brake pedal

is manually applied at a TTC of 1.4 s (

i.e.,

at a nominal headway of 9.6 m (31.5 ft.)). To pass this test, the SV must not contact the POV. See Figure 2 for a scenario diagram.

• Lead Vehicle Moving (LVM)—The SV encounters a slower-moving POV directly in front of it on a straight road. In the first test, the SV and POV are driven on a straight road at a constant speed of 40.2 kph (25 mph) and 16.1 kph (10 mph), respectively. In the second test, the SV and POV are driven at a constant speed of 72.4 kph (45 mph) and 32.2 kph (20 mph), respectively. In both tests, the SV throttle is released within 500 ms after the SV issues an FCW, and the SV brake pedal is manually applied at a TTC of 1 s (

i.e.,

at a nominal headway of 6.7 m (22 ft.) in the first test, and 11.3 m (37 ft.) in the second test). To pass these tests, the SV must not contact the POV. See Figure 3 for a scenario diagram.

• Steel Trench Plate (STP) false positive suppression test—The SV is driven over a 2.4 m × 3.7 m × 25.4 mm (8 ft. × 12 ft. × 1 in.) steel trench plate at 40.2 kph (25 mph) and 72.4 kph (45 mph). If an FCW is issued, the SV throttle is released within 500 ms of the alert. If no FCW is issued by a TTC of 2.1 s, the SV throttle is released within 500 ms of a TTC of 2.1 s. In both instances, the SV brakes are applied at a TTC of 1.1 s (

i.e.,

at a nominal distance of 12.3 m (40 ft.) from the edge of the STP at 40.2 kph (25 mph), or 22.3 m (73 ft.) at 72.4 kph (45 mph)). To pass this test, the performance criterion is minimal to no activation, as defined previously. See Figure 4 (below) for a scenario diagram.

EN03DE24.003

Currently, to pass NCAP's DBS system performance criteria, the SV must pass at least five out of seven trials for each of the six test conditions.

2. Crash Imminent Braking (CIB)

If a driver does not manually apply the vehicle's brakes when a rear-end crash is imminent, CIB systems, using the same forward-looking sensors as DBS systems, apply the vehicle's brakes automatically to slow or stop the vehicle. Unlike DBS systems, which provide additional braking to supplement the driver's brake input, CIB systems activate when the driver has not applied the brake pedal.

NCAP's Current Crash Imminent Braking (CIB) Test Procedure

The Agency's current CIB test procedure

is comprised of the same four test scenarios (LVS, LVD, LVM, and the STP false positive suppression test) and test speeds specified in the DBS test procedure. However, the performance criteria vary slightly. Whereas collision avoidance is the performance requirement stipulated for all DBS test scenarios except the false positive scenario, only the LVM 40.2 kph/16.1 kph (25 mph/10 mph) CIB test condition requires that the SV not contact the POV. The LVS, LVD, and the LVM 72.4 kph/32.2 kph (45 mph/20 mph) test conditions permit SV-to-POV contact but require minimum SV speed reductions prior to the contact being made. For the CIB false positive tests, the performance criterion is little-to-no activation, like the comparable DBS tests. Also, like NCAP's DBS tests, the SSV is the POV presently used in the program's CIB testing. A short description of each test scenario and the requirements for a passing result are provided below:

National Highway Traffic Safety Administration. (2015, October).

Crash imminent brake system performance evaluation for the New Car Assessment Program. http://www.regulations.gov.

Docket No. NHTSA-2015-0006-0025.

• LVS—The SV encounters a stopped POV directly in front of it on a straight road. The SV is moving at 40.2 kph (25 mph) and the POV (

i.e.,

the SSV) is stationary. The SV throttle is released within 500 ms after the SV issues an FCW. To pass this test, the SV speed reduction attributable to CIB intervention must be ≥ 15.8 kph (9.8 mph). See Figure 1 for a scenario diagram.

• LVD—The SV encounters a POV slowing with constant deceleration directly in front of it on a straight road. The SV and POV are both driven at 56.3 kph (35 mph) with an initial headway of 13.8 m (45.3 ft.). The POV then decelerates, braking at a constant deceleration of 0.3g in front of the SV, after which the SV throttle is released within 500 ms after the SV issues an FCW. To pass this test, the SV speed reduction attributable to CIB intervention must be ≥ 16.9 kph (10.5 mph). See Figure 2 for a scenario diagram.

• LVM—The SV encounters a slower-moving POV directly in front of it on a straight road. In the first test, the SV and POV are driven on a straight road at a constant speed of 40.2 kph (25 mph) and 16.1 kph (10 mph), respectively. In the second test, the SV and POV are driven at a constant speed of 72.4 kph (45 mph) and 32.2 kph (20 mph), respectively. In both tests, the SV throttle is released within 500 ms after the SV issues an FCW. To pass the first test, the SV must not contact the POV. To pass the second test, the SV speed reduction attributable to CIB intervention must be ≥ 15.8 kph (9.8 mph). See Figure 3 for a scenario diagram.

• STP test (to assess false positive suppression)—The SV is driven towards a steel trench plate at 40.2 kph (25 mph) in one test and 72.4 kph (45 mph) in the other test. If an FCW is issued, the SV throttle is released within 500 ms of the alert. If no FCW is issued, the throttle is not released until the test's validity period (the time when all test specifications and tolerances must be satisfied) has passed. To pass these tests, the SV must not achieve a peak deceleration equal to or greater than 0.5g at any time during its approach to the steel trench plate. See Figure 4 for a scenario diagram.

To pass NCAP's CIB system performance criteria, the SV must pass at least five out of seven trials for each of the six test conditions.

C. Linking Current FCW and AEB Test Scenarios With Real-World Crashes

NCAP's FCW and AEB test scenarios are directly related to real-world crash data. From its analysis of 2011 to 2015 Fatality Analysis Reporting System (FARS) and National Automotive Sampling System General Estimate System (GES) data, the Agency found that crashes analogous to the LVS test scenario, where a struck vehicle was stopped at the time of impact, occurred in 65 percent of the rear-end crashes studied.

54 55

The LVD scenario, in which the struck vehicle was decelerating at the time of impact, occurred in 22 percent of the rear-end crashes, and the LVM scenario, in which the struck vehicle was moving at a constant, but slower, speed compared to the striking vehicle at impact, occurred in 10 percent of the rear-end crashes. Collectively, these test scenarios represented 97 percent of rear-end crashes.

Wang, J.-S. (2019, March),

Target crash population for crash avoidance technologies in passenger vehicles

(Report No. DOT HS 812 653), Washington, DC: National Highway Traffic Safety Administration.

NHTSA notes that the target crash populations reported for the LVS, LVD, and LVM scenarios encompass all related real-world rear-end crashes (where the light vehicle is making a critical action) that could potentially be addressed by a DBS system. As such, the target crash populations for each crash scenario reflect crashes exhibiting variations in vehicle overlap, roadway curvature, environmental conditions, etc.; target crash populations were not reduced to align exactly with those represented by NCAP's LVS, LVD, and LVM test scenarios.

With respect to test speed, in its independent review of the 2011-2015 FARS and GES data sets, the John A. Volpe National Transportation Systems Center (Volpe) concluded that, when posted speed limit was known, 2 percent of fatal rear-end crashes and 6 percent of all rear-end crashes occurred on roadways with posted speed limits of 40.2 kph (25 mph) or less.

56 57 58

Eleven percent of fatal rear-end crashes and 33 percent of all rear-end crashes where posted speed limit was known occurred on roads with posted speeds of 56.3 kph (35 mph) or less. For posted speeds of 72.4 kph (45 mph) or less, Volpe found the comparable statistics to be 29 percent and 70 percent, respectively.

Swanson, E., Foderaro, F., Yanagisawa, M., Najm, W.G., & Azeredo, P. (2019, August),

Statistics of light-vehicle pre-crash scenarios based on 2011-2015 national crash data

(Report No. DOT HS 812 745), Washington, DC: National Highway Traffic Safety Administration.

NHTSA notes that throughout this notice, all crash statistics cited from Report No. DOT HS 812 745 encompass those where the light vehicle made the critical action (

e.g.,

losing control, departing road, changing lanes, striking, maneuvering, etc.).

For rear-end crashes, posted speed limit was unknown or not reported for 2 percent of fatal crashes and 11 percent of injurious crashes.

Roadway alignment and grade for the current FCW and AEB test scenarios are also comparable to those found where real-world rear-end crashes occur. NHTSA's LVS, LVD, and LVM procedures are to be performed on straight, level roads. In its review of 2011-2015 FARS and GES data sets, for rear-end crashes where roadway alignment was known, Volpe found that 95 percent of both fatal and injurious crashes occurred on a straight roadway.

For rear-end crashes where roadway grade was known, 77 percent of fatal crashes and 84 percent of crashes with injuries occurred on level roads.

For rear-end crashes, roadway alignment was unknown or not reported for 1 percent of fatal crashes and 3 percent of injurious crashes.

For rear-end crashes, roadway grade was unknown or not reported for 4 percent of fatal crashes and 18 percent of injurious crashes.

AEB Installation Rates, Effectiveness, and Research Tests

a. AEB Installation Rates

When NHTSA's CIB test scenarios were developed, relatively few vehicles were equipped with this technology; those that were equipped had systems with limited capabilities. Since then, fitment rates for CIB systems have increased significantly due in part to a voluntary industry commitment made in March 2016.

Per this commitment, 20 vehicle manufacturers, representing more than 99 percent of light motor vehicle sales in the U.S., voluntarily committed to make FCW and CIB standard on virtually all light-duty vehicles with a gross vehicle weight rating (GVWR) of 3,855.5 kg (8,500 pounds) or less beginning no later than September 1, 2022, and all trucks with a GVWR between 3,856.0 and 4,535.9 kg (8,501 and 10,000 pounds) beginning no later than September 1, 2025.

Conforming vehicles were required to be equipped with (1) an AEB system that earned at least an “advanced” rating from IIHS in its then-current front crash prevention track LVS tests and (2) an FCW system that met the performance requirements specified in two of NCAP's current three FCW test scenarios, LVD and LVM.

By 2019, participating manufacturers had equipped 75 percent of their new vehicle fleet with AEB,

and for model year 2023 vehicles, approximately 91 percent of the fleet was equipped with FCW and AEB systems as standard equipment.

The Agency also asserts that its recommendation of AEB systems (

i.e.,

CIB and DBS) that meet NCAP performance criteria on its website since the 2018 model year has further encouraged adoption of these technologies.

Insurance Institute for Highway Safety (2016, March 17),

U.S. DOT and IIHS announce historic commitment of 20 automakers to make automatic emergency braking standard on new vehicles, https://www.iihs.org/news/detail/u-s-dot-and-iihs-announce-historic-commitment-of-20-automakers-to-make-automatic-emergency-braking-standard-on-new-vehicles.

To achieve an advanced rating in IIHS' front crash prevention track tests, a vehicle's AEB system must show a speed reduction of at least 16.1 kph (10 mph) in either IIHS's 19.3 or 40.2 kph (12 or 25 mph) tests, or a speed reduction of 8.0 kph (5 mph) in both tests.

https://www.iihs.org/news/detail/u-s-dot-and-iihs-announce-historic-commitment-of-20-automakers-to-make-automatic-emergency-braking-standard-on-new-vehicles.

National Highway Traffic Safety Administration (2019, December 17),

NHTSA announces update to historic AEB commitment by 20 automakers, https://www.nhtsa.gov/press-releases/nhtsa-announces-update-historic-aeb-commitment-20-automakers.

As fitment increased, the sensor technology for CIB systems also advanced significantly. In 2017, many systems were not designed to meet the voluntary commitment thresholds, whereas by 2021, most vehicles with FCW and CIB systems could pass all relevant NCAP test scenarios, most of which are more stringent than those included in the voluntary agreement.

In its RFC, NHTSA noted that the original equipment manufacturer (OEM)-reported pass rate for NCAP's FCW and CIB tests for model year 2021 vehicles

equipped with these technologies, and for which manufacturers submitted data, was 89 percent and 70 percent, respectively.

Furthermore, NHTSA mentioned that only 63 percent of model year 2017 vehicles avoided contacting the POV for at least five out of seven of the required runs in the LVS CIB scenario during the Agency's testing, whereas 100 percent of model year 2021 vehicles were able to repeatedly avoid contact when tested.

It should be noted that a speed reduction of 15.8 kph (9.8 mph) for at least five out of seven trial runs is currently required to pass NCAP's CIB LVS test, not complete crash avoidance. For the model year 2023 vehicle fleet, the OEM-reported pass rate for the Agency's FCW test was 98 percent of equipped vehicles, and 86 percent for

the CIB test. In the Agency's model year 2023 CIB testing, all vehicles avoided contacting the POV test device for at least five out of seven runs, and thus received credit for passing performance. For the FCW assessments, only one vehicle model failed to provide a passing performance for the LVS and LVD scenarios.

NCAP's CIB test protocol requires a speed reduction of at least 15.8 kph (9.8 mph) in the program's 40 kph (24.9 mph) LVS test. However, the voluntary commitment allows a vehicle to comply with the memorandum for a speed reduction of 8.0 kph (5 mph) in IIHS's 19.3 and 40.2 kph (12 and 25 mph) LVS tests.

In this instance, “vehicles” refers to the total number of vehicles in the 2021 fleet, and not the total number of vehicle models for that year.

These values assume a 50 percent take rate for vehicles having optional equipment.

No contact was assumed if the test vehicle did not contact the POV in five or more of the seven required trial runs.

b. Model Year 2019 and 2020 Research Testing

As NHTSA noted in its March 2022 RFC, research testing conducted for a sample of model year 2019 and 2020 vehicles from various manufacturers also confirmed advancement of CIB system capabilities in recent years. The goal of this testing was to characterize the performance of then-current CIB systems and evaluate the technology's future potential for the new model years' vehicle fleet. For this purpose, the Agency chose to focus testing on NCAP's LVS and LVD test scenarios, as its review of the 2011-2015 FARS and GES rear-end crash data sets showed that LVS and LVD rear-end scenarios resulted in the highest number of crashes and MAIS 1-5 injuries.

NHTSA conducted testing for each scenario in accordance with NCAP's current CIB test procedure. These tests were then repeated using an ABD GVT as the surrogate vehicle in lieu of the SSV to verify that little to no change in performance would result.

The Agency also performed additional tests for each scenario using the GVT to assess how specific procedural changes (

i.e.,

increases in test speed and POV deceleration magnitude) affected CIB system performance.

Per Wang, J.-S. (2019, March),

Target crash population for crash avoidance technologies in passenger vehicles

(Report No. DOT HS 812 653), Washington, DC: National Highway Traffic Safety Administration, there were 1,099,868 LVS, 374,624 LVD, and 174,217 LVM crashes annually. Furthermore, there were 561,842 MAIS 1-5 injuries resulting from the LVS crash scenario, 196,731 for LVD, and 97,402 for LVM. The LVS scenario also had the second highest number of fatalities.

The Agency desired to use the GVT in lieu of the SSV for its higher speed testing because, given its material properties, the GVT significantly reduced the potential for damage to the testing equipment and test vehicles.

Test reports related to NHTSA's CIB characterization testing can be found in the docket for the March 2022 RFC notice.

For the additional LVS tests, the Agency incrementally increased the vehicle speed from 40.2 to 72.4 kph (25 to 45 mph) in 8.0 kph (5 mph) increments to identify when or if the vehicle reached its operational limits and/or did not react to the POV ahead. When the vehicle's intervention was insufficient (

i.e.,

the SV's maximum (peak) deceleration was less than 0.5g), the Agency repeated the test scenario at a test speed that was 4.0 kph (2.5 mph) lower. This reduced speed was used to define the system's upper capabilities for the LVS scenario.

For the additional LVD tests, the Agency evaluated how changes made to either the SV and POV speed (72.4 kph versus 56.3 kph (45 mph versus 35 mph)) or POV deceleration magnitude (0.5g versus 0.3g) affected CIB performance. No changes were made to the SV-to-POV headway; it was retained at 13.8 m (45.3 ft.).

The Agency chose to increase the test speeds for the scenarios included in its CIB characterization study because, in its independent analysis of the 2011-2015 FARS data set, Volpe found that, when the posted speed limit was known, approximately 29 percent of fatalities and 70 percent of injuries in rear-end crashes occurred when the posted speed on roadways was 72.4 kph (45 mph) or less.

The additional change to increase the POV deceleration in the LVD scenario was intended to create a more stringent test to address situations where the driver of a lead vehicle brakes aggressively, causing the driver of the following vehicle to have even less time to avoid or mitigate the crash than had the lead vehicle braking been at the 0.3g level presently specified in the Agency's test procedure. Based on previous Agency research, when drivers need to apply the brakes in a non-emergency situation, they do so by decelerating up to approximately 0.306g, while drivers encountering an unexpected obstacle apply the brakes at 0.48g.

Further, NHTSA noted that a deceleration of 0.5g falls within the range of deceleration magnitudes prescribed by Euro NCAP in its AEB Car-to-Car systems test protocol, Version 3.0.3, dated April 2021 for the Car-to-Car rear braking CCRb scenario. In its CCRb test, Euro NCAP specifies POV deceleration magnitudes of 2 m/s

and 6 m/s

(approximately 0.2 to 0.6 g) for an SV-to-POV headway of 12 m (39.4 ft.) and SV test speed of 50 kph (31.1 mph).

Swanson, E., Foderaro, F., Yanagisawa, M., Najm, W.G., & Azeredo, P. (2019, August),

Statistics of light-vehicle pre-crash scenarios based on 2011-2015 national crash data

(Report No. DOT HS 812 745), Washington, DC: National Highway Traffic Safety Administration.

Gregory M. Fitch, Myra Blanco, Justin F. Morgan, Jeanne C. Rice, Amy Wharton, Walter W. Wierwille, and Richard J. Hanowski (2010, April) Human Performance Evaluation of Light Vehicle Brake Assist Systems: Final Report (Report No. DOT HS 811 251) Washington, DC: National Highway Traffic Safety Administration, p. 13 and p. 101.

The Agency's characterization testing showed that many vehicles were able to repeatedly provide complete crash avoidance at higher test speeds and generally more aggressive conditions than those specified in NCAP's current CIB test procedure. For the 56.3 kph (35.0 mph) LVS tests conducted with a POV deceleration of 0.3g, seven out of the eleven vehicles avoided contact with the lead vehicle in every test trial. One of the remaining vehicles avoided contact in six out of seven test trials and the other three vehicles demonstrated an average speed reduction that exceeded 30.6 kph (19 mph). For the 72.4 kph (45.0 mph) LVS tests conducted with a POV deceleration of 0.3g, four out of the eleven vehicles avoided contact in every test trial and two other vehicles avoided contact in all but one test trial. Three of the remaining vehicles avoided contact in one or two test trials, while the two other vehicles could not avoid contact but both demonstrated an average 21 kph (13 mph) speed reduction. For the 56.3 kph (35.0 mph) LVD tests conducted at 0.5g rather than 0.3g, as specified in NCAP's current CIB test procedure, eight vehicles demonstrated the ability to avoid contact with the lead vehicle in at least one trial and three vehicles avoided contact in all trials, despite having less time to avoid the crash.

Similarly, when the speed of the SV and lead vehicle was increased to 72.4 kph (45 mph), nine vehicles demonstrated the ability to avoid contact with the lead vehicle in at least one test while four vehicles avoided contact in all tests. One vehicle avoided contact in all lead vehicle decelerating trials, including both increased speeds and increased lead vehicle deceleration.

Two vehicles avoided contact with the POV in four out of five trials.

Given these findings, the Agency concluded that current CIB systems are capable of significantly exceeding NCAP's current testing requirements. Thus, it is feasible to update the program's CIB test conditions to further safety improvements and address a greater number of rear-end crashes, particularly those which cause a greater number of injuries and fatalities in the real world.

c. AEB Effectiveness

In its March 2022 RFC notice, NHTSA discussed findings from several studies suggesting that AEB systems (

i.e.,

CIB and DBS) have collectively been effective in reducing rear-end crashes. As noted in the introductory section, UMTRI

found that AEB systems

produced an estimated 46 percent reduction in applicable rear-end crashes when combined with a forward collision alert, which alone showed only a 21 percent reduction.

Similarly, in a 2017 study, IIHS found that rear-end collisions decreased by 50 percent for vehicles equipped with AEB and FCW.

Furthermore, a 2019 study conducted by IIHS

suggested that the increasing effectiveness of AEB technology in certain crash situations, particularly those evaluated by NCAP and other consumer information programs, is changing the rear-end crash problem.

Leslie, A.J., Kiefer, R.J., Meitzner, M.R., & Flannagan, C.A. (2019, September),

Analysis of the field effectiveness of General Motors production active safety and advanced headlighting systems,

The University of Michigan Transportation

Research Institute and General Motors LLC, UMTRI-2019-6.

The AEB systems studied by UMTRI consisted of camera-only, radar-only, and fused camera-radar AEB systems, the latter two systems of which also included adaptive cruise control functionality.

Cicchino, J.B. (2017, February), Effectiveness of forward collision warning and autonomous emergency braking systems in reducing front-to-rear crash rates,

Accident Analysis and Prevention,

2017 Feb;99(Pt A):142-152,

https://doi.org/10.1016/j.aap.2016.11.009.

Cicchino, J.B. & Zuby, D.S. (2019, August), Characteristics of rear-end crashes involving passenger vehicles with automatic emergency braking,

Traffic Injury Prevention,

2019, VOL. 20, NO. S1, S112-S118

https://doi.org/10.1080/15389588.2019.1576172

While these studies suggest that AEB systems (

i.e.,

CIB and DBS) have collectively been effective in reducing rear-end crashes, NHTSA stated in its March 2022 notice that it was not clear how effective each of these systems is independently, or whether their individual effectiveness may change for certain crash scenarios, environmental conditions, or driver factors (

e.g.,

poor judgement, distraction, etc.). The Agency also stated it is not aware of any studies of current-generation AEB systems that have determined the extent to which CIB and DBS individually contribute to crash reduction. Since NHTSA could not differentiate between the individual effectiveness of CIB and DBS systems, it tentatively concluded that NCAP should continue to assess CIB and DBS system performance individually and therefore retain DBS assessments. NHTSA explained that this approach would ensure vehicles would not suppress AEB operation in situations where the driver applies the vehicle's foundation brakes.

However, as discussed later, the Agency also sought comment on removing the DBS test conditions from NCAP entirely in an effort to reduce test burden.

Foundation brake system means all components of the service braking system of a motor vehicle intended for the transfer of braking application force from the operator to the wheels of a vehicle.

See

49 CFR 579.4.

The Agency did not perform DBS testing as part of its characterization study to evaluate system performance capabilities beyond what is currently required in NCAP's respective test procedure. However, DBS systems have historically been shown to impart additional braking beyond that afforded by CIB systems. NHTSA has observed complete crash avoidance in DBS tests but only speed reduction in the equivalent CIB tests conducted for the same vehicle models. Therefore, it was expected that DBS performance should typically be as good as, if not better than, CIB performance. NHTSA believed that it was fitting to align the proposed CIB and DBS evaluations for the assessed situations, since doing so would allow the Agency to evaluate whether a vehicle's DBS system would provide sufficient supplemental braking if the driver brakes but additional braking is warranted. To verify that equivalent performance requirements and criteria proposed for CIB would also be appropriate for DBS, NHTSA planned research tests for model year 2021 and 2022 vehicles.

d. Model Year 2021 and 2022 Research Testing

In accordance with its plans expressed in the March 2022 RFC, NHTSA conducted a series of AEB research tests in 2022 to further analyze current fleet performance.

This testing, which involved 12 model year 2021 and 2022 light vehicles, included CIB and DBS testing in a variety of CIB and DBS test conditions. The goal of this research was to evaluate NHTSA's AEB proposals (found in subsequent sections) and to gain further knowledge regarding the capabilities of the current vehicle fleet.

https://www.regulations.gov/document/NHTSA-2023-0021-0005

Both CIB and DBS tests were conducted in the LVS, LVM, LVD, and STP false positive scenarios. Additionally, NHTSA conducted two other false positive test scenarios as part of this research: a “Pass Through” test, in which the SV approaches two stationary lead vehicles located to the left and right of the SV forward path, and “Pass Through + STP” test, which is a combination of the STP and Pass Through scenarios.

See Table 11 for the nominal test parameters used in this series of research tests. The ABD GVT Revision G, secured to a GST robotic platform (or carrier), was used as the POV for the model year 2021 and 2022 research testing.

In the Pass Through + STP test, the SV approaches a large steel plate positioned longitudinally on the test surface in the forward path of the SV. Two stationary lead vehicles are located to the left and right of the STP in the SV forward path. The SV is driven over the STP, between the two lead vehicles.

Table 11—Nominal Test Parameters for Model Year 2021 and 2022 Research Testing

Test scenario

Test speeds (kph (mph))

POV

Headway

(m (ft.))

POV

decel. (g)

CIB

DBS

Lead Vehicle Stopped (LVS)

10, 40, 50, 60, 70, 80 (6.2, 24.9, 31.1, 37.3, 43.5, 49.7)

✓

70, 80, 90, 100 (43.5, 49.7, 55.9, 62.1)

✓

Lead Vehicle Moving (LVM)

40, 50, 60, 70, 80 (24.9, 31.1, 37.3, 43.5, 49.7)

20 (12.4)

✓

70, 80, 90, 100 (43.5, 49.7, 55.9, 62.1)

20 (12.4)

✓

Lead Vehicle Decelerating (LVD)

50 (31.1)

12, 40 (39.4, 131.2)

0.4, 0.5

✓

80 (49.7)

12, 40 (39.4, 131.2)

0.4, 0.5

✓

Steel Trench Plate (STP)

80 (49.7)

✓

Pass Through

80 (49.7)

✓

Pass Through + Steel Trench Plate

80 (49.7)

✓

For the LVS and LVM test series, SV speed was increased from lowest to highest. One initial trial was conducted per test speed. If no SV-to-POV contact was observed, the next highest SV speed was run. However, if SV-to-POV contact

occurred and the SV speed at the time of impact was less than or equal to 50 percent of the initial SV speed, up to four additional (repeated) trials were performed at the same SV speed. If two additional SV-to-POV impacts were observed during the repeat sequence, the test series was terminated. Furthermore, if the SV speed at the time of impact was greater than 50 percent of the initial SV speed during the initial trial, no repeat runs were performed; testing for that scenario was terminated. For the LVD test series, testing proceeded in a similar manner. The SV speed was increased from lowest to highest, and POV deceleration was iteratively increased from 0.4g to 0.5g for a given speed combination and headway. Relevant outcomes of this research are detailed throughout the applicable sections of this notice.

D. NHTSA's Proposals, Summary of Comments, Response to Comments, and Agency Decisions

1. AEB

a. Forward Collision Prevention Technologies Inclusion in General

Many commenters, including the NTSB, Bosch, HMNA, and NADA, expressed support for the Agency's proposed updates for NCAP's AEB testing. Additional proponents, such as the Advocates and QuantivRisk, Inc., cited the need for increased test stringency to realize additional safety benefits. In this vein, NTSB specifically asked NHTSA to “strive for the performance we want the systems to be able to reach, not merely evaluate the current capabilities of the systems.” Auto Innovators expressed a need for consistency between changes to NCAP and those planned for AEB standards. Other respondents, such as MEMA, appreciated the Agency's attempts to focus resources on emerging trends and harmonize its AEB test procedures with those used by European New Car Assessment Programme (Euro NCAP) and other consumer information programs. Toyota also supported the Agency's attempts at shared global assessments but recommended that NHTSA select (1) tests that can adequately ensure performance across a range of conditions to improve overall test efficiency and (2) performance criteria that reflect real-world benefits. Citing rising fatalities, several commenters requested that the Agency consider AEB test additions for cyclists and motorcyclists, while others mentioned current AEB system limitations, such as systems' inability to detect cyclists and other vulnerable road users (VRUs) at higher speeds and in low light and inclement weather.

b. AEB Test Procedure Changes, Including Higher Speeds and POV Deceleration Magnitude for LVD Test; and Removal of DBS Tests and Only High-Speed DBS Assessments

NHTSA proposed harmonizing many aspects of changes to NCAP's CIB and DBS procedures with Euro NCAP's AEB Car-to-Car systems test protocol. The Agency reasoned this approach was most appropriate based on requirements of the BIL. The Agency also argued that it would be beneficial to standardize the current AEB test specifications with other consumer information programs, as doing so would allow the Agency and vehicle manufacturers to focus resources on emerging trends for rear-end crashes as AEB-equipped vehicles become more abundant in the fleet.

NHTSA also noted it would consider making additional updates to its AEB test evaluation as the rear-end crash problem evolves.

Cicchino, J.B., & Zuby, D.S. (2019, August), Characteristics of rear-end crashes involving passenger vehicles with automatic emergency braking,

Traffic Injury Prevention,

2019, VOL. 20, NO. S1, S112-S118,

https://doi.org/10.1080/15389588.2019.1576172

The updated CIB and DBS tests proposed in the 2022 RFC are detailed below for each test scenario. Tables 12 and 13 summarize the proposed test scenarios and conditions.

Table 12—CIB Test Scenarios and Conditions Proposed in the 2022 RFC

Test

scenario

SV speed

(kph (mph))

POV speed

(kph (mph))

POV

headway

(m (ft.))

POV

deceleration

(g)

Requirement to pass

LVS

40 (24.9)

50 (31.1)

60 (37.3)

70 (43.5)

80 (49.7)

n/a

(1) No SV-to-POV contact on first trial;

(2) Any SV-to-POV contact where the relative velocity between the SV and POV is ≤ 50% of initial SV speed

AND

No SV-to-POV contact in 3 out of 5 total trials.

LVM

40 (24.9)

20 (12.4)

n/a

50 (31.1)

20 (12.4)

n/a

60 (37.3)

20 (12.4)

n/a

70 (43.5)

20 (12.4)

n/a

80 (49.7)

20 (12.4)

n/a

LVD *

50 (31.1)

12 (39.4)

0.5

60 (37.3)

12 (39.4)

0.5

70 (43.5)

12 (39.4)

0.5

80 (49.7)

12 (39.4)

0.5

* For LVD, NHTSA requested comment on whether at least five of seven trials should be required for vehicles whose contact velocity is ≤50 percent of the initial velocity, whether a 40 m headway should be included, and whether NHTSA should employ a 0.6g POV deceleration in lieu of 0.5g.

Table 13—DBS Test Scenarios and Conditions* Proposed in the 2022 RFC

Test

scenario

SV speed

(kph (mph))

POV speed

(kph (mph))

POV

headway

(m (ft.))

POV

deceleration

(g)

Requirement to pass

LVS **

70 (43.5)

80 (49.7)

n/a

(1) No SV-to-POV contact on first trial;

(2) Any SV-to-POV contact where the relative velocity between the SV and POV is ≤50% of initial SV speed

AND

No SV-to-POV contact in 3 out of 5 total trials.

LVM **

70 (43.5)

20 (12.4)

n/a

80 (49.7)

20 (12.4)

n/a

LVD ***

70 (43.5)

12 (39.4)

0.5

80 (49.7)

12 (39.4)

0.5

* For all DBS conditions, NHTSA requested comment on removal of all DBS test conditions.

** For LVS and LVM, NHTSA requested comment on the additional inclusion of 40, 50, and 60 kph (24.9, 31.1, and 37.3 mph).

*** For LVD, NHTSA requested comment on the additional inclusion of 50 and/or 60 kph (31.1 and/or 37.3 mph) SV/POV speeds or only 70 and 80 kph (43.5 and 49.7 mph) (if they were adopted for CIB as well). NHTSA also requested comment on whether at least five of seven trials should be required to satisfy the performance requirement for vehicles whose relative velocity at contact is ≤50 percent of the initial SV speed, whether 40 m headway should be included, and whether NHTSA should employ 0.6g POV deceleration in lieu of 0.5g.

Crash Imminent Braking (CIB)

Lead Vehicle Stopped (LVS)

Currently, NCAP's CIB LVS test is conducted at a speed of 40.2 kph (25 mph). In its upgrade proposal, the Agency recommended assessing CIB system performance over a range of test speeds for this test scenario. Specifically, NHTSA proposed a minimum SV test speed of 40 kph (24.9 mph) (similar to that currently specified in NHTSA's CIB test procedure) and a maximum SV test speed of 80 kph (49.7 mph). NHTSA also proposed increasing the SV test speed in 10 kph (6.2 mph) increments from the minimum test speed to the maximum test speed for the LVS assessment, performing one trial per speed. To achieve a passing result for each speed, NHTSA proposed that the test trial must be valid (all test specifications and tolerances satisfied), and the SV must not contact the POV. Further, the Agency proposed that it would conduct four additional trials for any specific test speed that resulted in a test failure (

i.e.,

contact) as long as the SV relative velocity at impact was less than or equal to 50 percent of the initial SV speed. For these five trials (

i.e.,

one failed trial and four additional trials), NHTSA proposed that the SV must avoid contact with the POV for at least three trials to pass the test condition (

i.e.,

combination of test scenario and test speed).

In justifying its recommendation to incorporate higher test speeds for the LVS scenario, in addition to Volpe's real-world data analysis, which illustrated the safety need, the Agency indicated its CIB characterization testing demonstrated that several vehicles repeatedly afforded full crash avoidance (

i.e.,

no contact) at speeds up to 72.4 kph (45 mph) when subjected to this test. Further, NHTSA recognized that Euro NCAP's Car-to-Car Rear stationary (CCRs) scenario, which is comparable to the Agency's LVS test, is conducted at speeds as high as 80 kph (49.7 mph) in the “AEB only” test condition. NHTSA reasoned that Euro NCAP's use of higher test speeds suggests higher test speeds are, from the perspective of test conduct, practicable for NCAP's LVS test as well.

The Agency believed it was appropriate to harmonize with Euro NCAP on the maximum LVS test speed of 80 kph (49.7 mph), as this should better address the higher severity, high-speed crash problem and, in turn, further reduce fatalities and serious injuries. However, NHTSA did not propose to harmonize with Euro NCAP's protocol on the minimum SV test speed. Euro NCAP's CCRs scenario specifies a minimum SV speed of 10 kph (6.2 mph) for AEB systems, but the Agency stated it did not see the need to conduct its updated LVS testing at a speed less than that which is specified in its existing test procedure (40.2 kph (25 mph)). As such, a minimum test speed of 40 kph (24.9 mph) was proposed instead.

European New Car Assessment Programme (Euro NCAP) (April 2021),

Test Protocol—AEB Car-to-Car systems, Version 3.0.3.

See section 8.2.3.

The Agency sought comment on whether the proposed speeds and overall assessment approach were appropriate for LVS or whether alternatives should be considered.

Lead Vehicle Moving (LVM)

As mentioned previously, NCAP's CIB test procedure currently includes two LVM test conditions: a lower speed assessment that specifies an SV speed of 40.2 kph (25 mph) and POV speed of 16.1 kph (10 mph), and a higher speed assessment that prescribes an SV speed of 72.4 kph (45 mph) and POV speed of 32.2 kph (20 mph). For this NCAP update, NHTSA proposed to assess CIB system performance over a range of SV test speeds for the LVM scenario. Similar to its proposal for the LVS scenario, NHTSA proposed to implement a “no contact” performance criterion for the LVM scenario and to increase the SV test speed for the LVM assessment in 10 kph (6.2 mph) increments from a minimum speed of 40 kph (24.9 mph) to a maximum speed of 80 kph (49.7 mph), with a POV speed of 20 kph (12.4 mph) for every SV test speed. The Agency also proposed to perform one trial run per speed and four additional trials for any specific test speed that resulted in a test failure for initial runs where the SV had a relative velocity at impact less than or equal to 50 percent of the initial SV speed. Similar to its proposal for LVS, the Agency proposed that the SV must not contact the POV for at least three out of the five test trials performed at that same speed to pass the LVM test condition.

The Agency noted that the proposed minimum SV test speed of 40 kph (24.9 mph) is nearly equivalent to the speed currently specified for its lower speed LVM assessment, 40.2 kph (25 mph), and the proposed maximum SV test speed of 80 kph (49.7 mph) is only slightly higher than the speed specified for its higher speed LVM assessment, 72.4 kph (45 mph). Since NCAP's higher speed CIB LVM assessment (conducted at an SV speed of 72.4 kph (45 mph) and POV speed of 32.2 kph (20 mph)) showed that many vehicles were able to stop without contacting the POV test device for each of the required test trials, NHTSA believed it was reasonable to raise the SV speed in NCAP's LVM test even though it had not performed additional LVM testing as part of its characterization study. The Agency also noted that Euro NCAP performs its Car-to-Car Rear moving (CCRm) scenario (which is comparable to NCAP's LVM tests) at speeds as high as 80 kph (49.7 mph), further suggesting that higher SV test speeds are practicable.

Given this, NHTSA believed it was appropriate to harmonize with Euro NCAP on the maximum SV test speed of 80 kph (49.7 mph) for the Agency's LVM test. NHTSA reasoned that adopting a higher maximum SV test speed than that which is currently required in the Agency's CIB procedure should encourage improved CIB system performance at higher speeds, and thus further reduce fatalities and serious injuries.

European New Car Assessment Programme (Euro NCAP) (April 2021),

Test Protocol—AEB Car-to-Car systems, Version 3.0.3.

See section 8.2.4.

Although it proposed to harmonize with Euro NCAP's protocol with respect to the maximum SV speed adopted for

its LVM test, the Agency did not suggest harmonizing with Euro NCAP with respect to the minimum required SV test speed. Euro NCAP's CCRm scenario specifies a minimum SV test speed of 30 kph (18.6 mph) for AEB-equipped vehicles; however, the Agency did not believe there was not a compelling reason to perform its updated LVM test at a speed that is less than the current required test speed (

i.e.,

40.2 kph (25 mph)) since most vehicles have been able to meet NCAP's current LVM test requirements at 40.2 kph (25 mph) to date with a similar POV test speed. Accordingly, NHTSA proposed a minimum SV test speed of 40 kph (24.9 mph).

NHTSA proposed to adopt a POV test speed of 20 kph (12.4 mph). The Agency noted this POV speed is specified in Euro NCAP's CCRm protocol, and therefore adopting this speed for NHTSA's LVM testing seemed appropriate since it would further support harmonization efforts.

Comments were requested on whether the SV/POV speeds and assessment approach proposed for NCAP's CIB LVM tests were appropriate or whether alternative speeds or approaches should be considered instead.

Lead Vehicle Decelerating (LVD)

For the LVD scenario, NHTSA proposed to reduce the minimum nominal SV and POV test speeds from 56 kph (34.8 mph), as specified in NCAP's test procedure, to 50 kph (31.1 mph), as stated in Euro NCAP's AEB Car-to-Car systems test protocol, Version 3.0.3, dated April 2021 for the Car-to-Car rear braking (CCRb) scenario.

The Agency stated that, given additional changes proposed for the SV-to-POV headway and deceleration magnitude for the LVD scenario, the proposed reduction in test speed would not lead to an overall reduction in test stringency or loss of safety benefits. NHTSA requested comment on whether this proposed test speed change was appropriate for NCAP's LVD testing.

European New Car Assessment Programme (Euro NCAP) (April 2021),

Test Protocol—AEB Car-to-Car systems, Version 3.0.3.

See section 8.2.5.

The Agency also sought comment on whether it would be appropriate to incorporate additional SV and POV test speeds for the LVD test scenario: 60, 70, and 80 kph (37.3, 43.5, and 49.7 mph, respectively). Similar to the proposed CIB LVS and LVM test scenarios, NHTSA proposed to concurrently increase the SV and POV test speeds in 10 kph (6.2 mph) increments from the minimum test speed to the maximum test speed for NCAP's LVD assessment if multiple speeds were adopted. The Agency also proposed, as discussed in a later section, to perform one trial run per speed and four additional trials for any specific test speed that resulted in a test failure (

i.e.,

SV-to-POV contact) where the SV had a relative velocity at impact less than or equal to 50 percent of the initial SV speed. Like the other two CIB test scenarios, the Agency proposed the SV must not contact the POV for at least three out of the five test trials performed at that same speed to pass the test condition. Alternatively, the Agency sought comment on whether testing at only 50 kph (31.1 mph) and 80 kph (49.7 mph) would be acceptable.

NHTSA acknowledged in its proposal that it had not yet performed LVD testing at 80 kph (49.7 mph), mainly due to equipment and test track length limitations, as this test scenario requires that both the SV and POV be travelling at the same speed at the onset of the test validity period. However, the Agency recognized that higher speed tests may be warranted. For instance, Volpe's analysis of the 2011-2015 FARS data set showed that, when posted speed limit was known, the majority of fatal rear-end crashes (71 percent) occurred on roads with posted speeds exceeding 72.4 kph (45 mph). Considering the braking performance observed during the high-speed LVS tests conducted as part of its characterization study, the Agency noted current vehicles may perform well in LVD tests conducted at even higher speeds. Additionally, NHTSA believed that CIB systems may be able to classify POVs more confidently in the LVD test compared to the LVS test due to the POV's detected motion (

i.e.,

path history). Accordingly, NHTSA conducted research to assess vehicles' CIB system performance in the LVD test at SV and POV speeds ranging from 50 kph (31.1 mph) to 80 kph (49.7 mph) to determine the feasibility of adopting one or more of these speeds for this test scenario.

The Agency proposed these speeds would each be assessed for both 12 and 40 m (39.4 and 131.2 ft.) headways and POV deceleration magnitudes of 0.4g and 0.5g.

In its March 2022 RFC notice, NHTSA also proposed to reduce the minimum nominal SV-to-POV headway of 13.8 m (45.3 ft.), currently specified for the LVD scenario, to 12 m (39.4 ft.) for the proposed test speed of 50 kph (31.1 mph). Although not assessed as part of its CIB characterization testing, the Agency asserted this change would not only harmonize with Euro NCAP's CCRb scenario with respect to test conduct, but also maintain similar stringency to NCAP's current LVD test scenario, given the proposed test speed reduction from 56 kph (34.8 mph) to 50 kph (31.1 mph). Euro NCAP also specifies an additional SV-to-POV headway of 40 m (131.2 ft.); however, the Agency did not propose to conduct this assessment as part of the RFC, as NHTSA suggested there would not be a safety benefit in adopting 40 m (131.2 ft.) as an additional, and presumably less stringent, headway. Therefore, the Agency did not want to increase the test burden unnecessarily. However, the Agency indicated that it would assess vehicle performance at both 12 and 40 m (39.4 and 131.2 ft.) headways as part of its future research for each of the test speeds to be evaluated. The Agency also sought public comment on which SV-to-POV headway(s) may be appropriate for adoption not only for the proposed test speed (

i.e.,

50 kph (31.1 mph)), but also for each of the additional test speeds (ranging from 60 kph (37.3 mph) to 80 kph (49.7 mph)) it planned to evaluate and possibly incorporate.

The last change the Agency proposed for the LVD test scenario was increasing the POV deceleration magnitude currently specified in its CIB test procedure from 0.3g to 0.5g. In the Agency's CIB characterization study, three vehicles repeatedly afforded full crash avoidance (

i.e.,

no contact) for all trials when the POV executed a 0.5g braking maneuver in the LVD condition with an SV test speed of 56.3 kph (35 mph) and SV-to-POV headway of 13.8 m (45.3 ft.), demonstrating that the change to POV deceleration for the revised LVD test conditions (which also includes a slightly lower test speed and slightly shorter SV-to-POV headway) is likely feasible. The Agency also noted that, in Euro NCAP's AEB Car-to-Car systems test protocol, the organization specifies POV deceleration magnitudes of 2 m/s

and 6 m/s

(approximately 0.2 and 0.6g) for its CCRb scenario.

As such, NHTSA reasoned that adopting a 0.5g POV deceleration magnitude would be practicable. To verify this assumption, as part of its research study, NHTSA committed to evaluating POV deceleration magnitudes of both 0.4 and 0.5g for the range of test speeds considered (

i.e.,

60, 70, and/or 80 kph (37.3, 43.5, and/or 49.7 mph)) for future LVD testing.

The Agency also sought comment on what deceleration magnitude(s) would be appropriate for the proposed test speed (

i.e.,

50 kph

(31.1 mph)), as well as each of these additional test speeds.

European New Car Assessment Programme (Euro NCAP) (),

Test Protocol—AEB Car-to-Car systems, Version 3.0.3.

See section 8.2.5.

NHTSA notes that the LVD research tests were conducted only for 50 and 80 kph (31.1 and 49.7 mph) test speeds.

NHTSA did not propose a 0.6g POV deceleration magnitude for use in its LVD test, even though Euro NCAP specifies 0.6g as the maximum POV deceleration for its CCRb scenario. In proposing 0.5g as the maximum POV deceleration in lieu of 0.6g, the Agency stated a lower POV deceleration may reduce equipment wear, particularly for the tires and braking components of the POV propulsion system, thus improving test efficiency. Specifically, NHTSA explained it has observed instances where the tires of the low-profile robotic vehicle (LPRV) platform

used to move the GVT developed flat spots while performing a braking maneuver similar to that specified in the Agency's CIB LVD test

but with higher POV decelerations. During this testing, NHTSA also found it was more difficult to achieve and accurately control POV deceleration within prescribed tolerances when braking maneuvers higher than 0.5g were used, even with extensive LPRV tuning efforts.

The Agency noted that a deceleration of 0.6g is not only very close to the maximum braking capability of the LPRV, but also very close to the default magnitude used by the LPRV during an emergency stop (maximum deceleration). However, NHTSA acknowledged that newer robotic platforms (

i.e.,

robotic carriers) offering greater capabilities are now becoming available, and they may resolve the issues observed in the Agency's testing. Accordingly, NHTSA requested comment on whether it may now be feasible to adopt a POV deceleration magnitude of 0.6g in lieu of 0.5g, as proposed.

The GVT is secured to the top of the LPRV. The LPRV is responsible for any movement of the GVT during test conduct.

Fogle, E. E., Arquette, T. E. (TRC), and Forkenbrock, G. J. (NHTSA), (2021, May),

Traffic Jam Assist Draft Test Procedure Performability Validation

(Report No. DOT HS 812 987), Washington, DC: National Highway Traffic Safety Administration.

From Section 4.1 of DOT HS 812 987: “POV deceleration validity check failures occurred during six trials of the eight LVDAD trials performed. Four of the seven 0.6 g failures were because the POV was unable to achieve the minimum deceleration threshold of 0.55 g. The remaining three 0.6 g failures were because the POV was unable to maintain a minimum average deceleration of at least 0.55 g.” Here, LVDAD refers to “Lead Vehicle Accelerates, Decelerates, then Decelerates.” The LVDAD test is a more complex variant of the LVD test and was used by NHTSA to perform traffic jam assist research.

Dynamic Brake Support (DBS)

With respect to DBS, the Agency proposed to align all test conditions (

e.g.,

SV and POV test speed(s), headway(s), POV deceleration magnitude(s), etc.) for the comparable LVD, LVM, and LVS test scenarios with those proposed for CIB. Likewise, NHTSA proposed a similar performance criterion (

i.e.,

no contact) and assessment approach as well; when applicable, speeds would be increased in 10 kph (6.2 mph) increments from the minimum test speed to the maximum test speed, with one trial performed per speed, and four additional trials conducted for any specific test speed that resulted in a test failure (

i.e.,

contact) as long as the SV had a relative velocity at impact less than or equal to 50 percent of the initial SV speed. Similar to CIB, the Agency proposed that the SV must avoid contact with the POV for at least three out of the five test trials performed at that same speed to pass the test condition (if the vehicle fails the initial trial at a given test speed).

Although the Agency had not conducted DBS testing as part of its characterization study to evaluate system performance capabilities beyond what is currently required in NCAP's DBS test procedure, NHTSA believed it was nonetheless fitting to align the proposed CIB and DBS evaluations, as it would allow NHTSA to assess whether a vehicle's DBS system will provide supplemental braking if the driver manually applies a brake pedal input but additional braking is warranted to afford crash avoidance. Further, the Agency noted its CIB and DBS test procedures are currently aligned with respect to test scenarios, test speeds, headways, etc. Differences exist only with respect to the use of manual brake application (

i.e.,

for the SV in DBS testing) and (most) performance criteria.

Therefore, the Agency reasoned it would be appropriate to adopt the CIB test conditions (

i.e.,

test speeds, headways, etc.) for the comparable DBS test conditions for future testing as well. NHTSA requested comments on whether this proposal for future DBS testing, including the assessment method, was appropriate.

NHTSA's DBS test procedure currently specifies “no contact” as the performance criterion for all DBS test conditions, whereas the Agency's CIB test procedure currently requires a specified speed reduction for each of the CIB test conditions (with the exception of the lower speed LVM condition where the POV speed is 16.1 kph (10 mph) and the SV speed is 40.2 kph (25 mph), which requires “no contact.”

The Agency also sought comment on removing the LVD, LVM, and LVS DBS test conditions from NCAP entirely (in addition to the false positive test conditions, discussed later) to reduce test burden and associated costs given findings from Volpe's analysis of the 2011-2015 FARS and GES data sets and other changes NHTSA proposed for its CIB assessments.

Specifically, Volpe found that the driver braked in just 8 percent of rear-end crashes involving fatalities and in 20 percent of those crashes involving injuries. The study also showed that the driver made no attempt to avoid the crash (

e.g.,

no braking, steering, accelerating) for 56 percent of crashes involving fatalities and for 21 percent of those involving injuries.

These findings were contrary to those documented by NHTSA during a review of 2003-2009 National Automotive Sampling System Crashworthiness Data System data to define the target population for rear-end crashes.

For that analysis, the Agency concluded that the driver braked in approximately half of the crashes and did not brake in the other half, which lends merit to performing both CIB and DBS tests. The Agency believed it was possible the brake application rates differed in the two studies because of (1) target crash population refinements made for NHTSA's original analysis and (2) differences in data collection methods between the crash databases. For instance, high-speed crashes were excluded from NHTSA's target crash population review because the AEB systems tested at the time had limited speed reduction capabilities.

The Agency notes that for the rear-end pre-crash scenario group, the driver avoidance maneuver was unknown in 25 percent and 54 percent of the FARS and GES crashes, respectively. When excluding cases where a driver avoidance maneuver was unknown, the driver made no attempt to avoid the crash in 75 percent and 48 percent of the FARS and GES crashes, respectively. Likewise, when a driver's avoidance maneuver was known, the driver braked in 11 percent of FARS crashes and 45 percent of GES crashes.

National Highway Traffic Safety Administration (2012, June),

Forward-looking advanced braking technologies research report, https://www.regulations.gov/document?D=NHTSA-2012-0057-0001

As previously mentioned, NHTSA proposed to adopt a more stringent “no contact” performance criterion for each of NCAP's CIB test conditions. The Agency's existing CIB test procedure requires a specified speed reduction for each of the CIB test conditions (with the exception of the lower speed LVM condition, which requires “no contact”), whereas the DBS test procedure currently specifies “no contact” as the performance criterion for all DBS test conditions. The proposed change for CIB would effectively align the CIB performance requirements with those currently specified for DBS, and NHTSA questioned whether it was necessary to continue performing DBS tests in NCAP given public comments previously

received. For example, in its comments to NCAP's December 2015 notice, the Alliance

stated that since crash avoidance (

i.e.,

no SV-to-POV contact) is the desired outcome for all imminent rear-end crash events, if an SV avoids contact with the POV in all CIB tests, DBS testing should not be necessary. The Agency agreed with the Alliance's rationale in principle but questioned whether there would be merit to ensuring both AEB systems perform as designed and help the driver to mitigate or prevent the crash. NHTSA hypothesized that it may be possible for the driver to apply the brakes but with a magnitude that does not result in achieving the vehicle's maximum crash avoidance potential (

i.e.,

deceleration). Further, the Agency explained that, in the past, some manufacturers had assumed the driver was in control when the brake pedal was depressed, and designed CIB systems such that automatic braking was overridden by the driver's input, even when the driver's braking was insufficient to avoid a crash. Based on this reasoning, NHTSA explained it was hesitant to assume that if a vehicle's CIB system works effectively during testing, its DBS system would automatically do so as well.

Alliance of Automobile Manufacturers (The Alliance) merged with Global Automakers in January 2020 to create the Alliance for Automotive Innovation (Auto Innovators). Both automotive industry groups separately submitted comments to the December 2015 notice.

Thus, as an alternative to removing the DBS performance evaluations from NCAP entirely (or retaining the LVD, LVM, and LVS DBS tests in NCAP, as proposed), the Agency concluded that it might be more reasonable to conduct only LVS and LVM DBS tests in NCAP at the highest two test speeds proposed for CIB—70 and 80 kph (43.5 and 49.7 mph, respectively)—to ensure (1) the DBS system functions properly at these speeds and (2) the SV will not suppress AEB operation when the driver applies the vehicle's foundation brakes. The Agency further noted that it would also consider conducting the DBS LVD test at only 70 and 80 kph (43.5 and 49.7 mph, respectively) if it decided to adopt those same higher test speeds for the CIB LVD test. Comments were requested on this alternative proposal and whether an alternative assessment method would be more appropriate if any or all of the DBS test scenarios were conducted only at the two highest test speeds.

Summary of Comments

Regarding NHTSA's AEB Proposal, In General

Several commenters, including BMW, FCA, and Honda, supported the Agency's proposal for CIB and DBS with respect to SV and POV speeds, headway distances, and POV deceleration magnitudes. FCA stated that the proposal was appropriate because it reflects current system capabilities and real-world crashes. With the exception of suggested changes for the LVD scenario, discussed later, Tesla also generally agreed with the Agency's proposal with respect to test speeds, headway, and deceleration magnitudes for CIB testing and the general intent to harmonize with Euro NCAP test protocols. Specifically, Tesla supported conducting LVS and LVM scenarios at test speeds ranging from 40 to 80 kph (24.9 and 49.7 mph) with 10 kph (6.2 mph) increments, as proposed. Auto Innovators also generally agreed with the proposed test requirements but suggested the Agency harmonize with Euro NCAP and conduct CIB testing up to 50 kph (31.1 mph) and DBS testing at speeds over 50 kph (31.1 mph).

Like Tesla, Advocates and Bosch also supported generally harmonizing the Agency's CIB testing with that performed by Euro NCAP, but with small variations. Advocates supported aligning the LVM and LVS POV and SV speeds with those used by Euro NCAP but did not support the Agency's justification for not aligning minimum test speeds for these two scenarios with those prescribed by Euro NCAP (10 kph, or 6.2 mph) as being sufficient. Bosch also mentioned harmonization with respect to test speeds but suggested the Agency should further investigate whether there is merit to increasing test speeds to assess AEB systems.

Conversely, GM opposed any change to the test conditions prescribed in the Agency's current CIB test procedure. The automaker stated that the current AEB test speeds show significant real-world safety benefits

and, as documented in DOT HS 811 521A, “Objective Tests for Automatic Crash Imminent Braking (CIB) Systems,” the current test parameters are “well-supported by field crash scenarios most relevant to these features and associated with the highest societal harm, as measured by Functional Years Lost.”

See GM Appendix 1.

Other commenters suggested that the Agency remove certain test conditions. Auto Innovators suggested that the Agency remove one of the two original LVM scenarios, preferably the lower speed condition (

i.e.,

SV and POV speeds of 40 and 16 kph (25 and 10 mph), respectively), since real-world data shows only 2 percent of fatalities and 6 percent of injuries occur on roads having posted speed limits of 40 kph (25 mph) or less. Similarly, Toyota stated that the Agency should adopt only the number of test conditions sufficient to communicate accurate performance information to consumers. The automaker suggested that, if testing only at a certain speed would ensure performance for a large speed range, then that approach was acceptable for testing.

With respect to other procedural considerations, Subaru recommended that NHTSA adopt a speed increment of 20 kph (12.4 mph) in lieu of 10 kph (6.2 mph) for LVM testing. A few other respondents also generally supported the test parameters, but suggested slight modifications, which are addressed later in this section.

Adopt Higher AEB Test Speeds Than Those Proposed

State Farm, IIHS, and Uhnder supported CIB and DBS testing at higher speeds, stating that such speeds better reflect real-world driving conditions. Uhnder supported adoption of test speeds that exceed 88.5 kph (55 mph), citing a May 2022 study from IIHS finding nearly 70 percent of fatal rear-end crashes occurred when the speed limit was 88.5 kph (55 mph) or higher.

Similarly, IIHS noted that nearly 80 percent of police-reported rear-end crashes occurred on roads having speed limits ranging from 48.3 to 104.6 kph (30 to 65 mph),

and the speed of the striking vehicle was more than 40 kph (24.9 mph), even on roads with speed limits of 40.2 kph (25 mph).

See footnote 11 of Uhnder response.

Kidd, 2022a.

Kidd, 2022b.

Adasky, NTSB, and CAS also favored higher speed AEB assessments than those proposed. CAS asserted that NHTSA should conduct CIB tests at the highest speeds possible to still afford safe testing so that consumers may identify vehicles offering superior CIB performance at each test speed. Similarly, NTSB encouraged NHTSA to consider more challenging test speeds to drive desired (

i.e.,

ideal) system performance instead of testing to current system capabilities. Finally, Rivian also suggested adopting higher speeds for DBS tests than those proposed if the Agency continued DBS testing in the future.

Test Speeds and Headway for the LVD Test Scenario Specifically

Several commenters favored adopting AEB test speeds up to 80 kph (49.7 mph) for the LVD test scenario, with BMW and Honda stating that these test speeds were appropriate since they were supported by crash data.

Tesla, along with Subaru, recommended conducting LVD scenarios at 50 kph (31.1 mph), as proposed (similar to Euro NCAP), and also at 80 kph (49.7 mph), as suggested by NHTSA. Subaru added that, if a test failure occurs at 80 kph (49.7 mph), the test speed should then be reduced by 10 kph (6.2 mph). Advocates favored harmonization with Euro NCAP with respect to test speed, headway, and deceleration for the LVD scenario, but preferred NHTSA's alternative proposal of adopting multiple higher test speeds (above 50 kph (31.1 mph)), suggesting NHTSA should also include a “range of test speeds” based on crash data and the Agency's testing.

Toyota encouraged NHTSA to conduct additional feasibility studies and research, particularly for the LVD test scenario, to: (1) resolve possible GVT stability issues; (2) study the possible conflict with human driver steering avoidance maneuver timing; and (3) research the effectiveness of FCW and DBS based on the physical limitations imposed by the proposed LVD DBS test condition, and, considering driver reaction times, determine whether such higher-speed testing is feasible and appropriate before modifying the AEB test conditions. With respect to the first request, Toyota noted the Agency's statement that it has not conducted CIB/DBS LVD testing at 80 kph (49.7 mph) because of equipment and test track length limitations. Regarding its second point, the automaker asserted that the time required to steer to avoid a collision at higher speeds is less than the time required to brake, and that by imposing the suggested high speed CIB test conditions, the Agency may create a challenging `braking' situation that could interfere with a driver's ability to avoid the crash by steering instead. Lastly, Toyota voiced concern that the SV and POV dynamics for DBS LVD testing at higher speeds may pose physical limitations.

100

More specifically, for speeds of 50 kph (31.1 mph) and greater, the manufacturer asserted that, given the (constant) headway prescribed, the time (

i.e.,

TTC) required to activate the brake to avoid impact, and the proposed brake application timing of 1.0 s after issuance of the FCW, it is possible the FCW would have to be issued before the POV test device begins to decelerate in the DBS test for the SV to avoid contact with the POV. Toyota was supportive of DBS testing at higher speeds for the LVS test scenario.

100

See case study included in Toyota's comments.

Intel shared Toyota's concerns about the LVD DBS tests, explaining the proposed headway (12 m (39.4 ft.)) may be too small given a POV deceleration of 0.5 to 0.6g, such that it may not be possible to issue the FCW early enough to achieve brake activation one second after the issuance of the FCW as NHTSA proposed for the test procedure. Intel also cautioned the Agency that it should ensure it is feasible for test labs to conduct LVD tests at the proposed higher speeds considering the GVT platform experiences performance degradation at high speeds. Intel further noted many automakers limit speed reductions to approximately 60 kph (37.3 mph) per Automotive Safety Integrity Level (ASIL) considerations.

Auto Innovators did not support the Agency's adoption of test speeds exceeding the capabilities afforded by current systems for the LVD test condition because it may induce false positives under real-world driving conditions, which may in turn discourage AEB use. The group, like Toyota, also cautioned that the proposed high-speed testing may cause unexpected interactions between the SV and POV during testing. Auto Innovators recommended that the Agency consider the proposed changes for CIB and DBS for future program updates to (1) allow additional time to investigate the field relevance of the proposed changes for both technologies and (2) provide sufficient time for system capabilities to improve.

To better align with Euro NCAP testing, Intel suggested that LVD testing should be limited to 50 kph (31.1 mph) and should be performed only for vehicles equipped with both AEB and FCW. HATCI recommended that the Agency harmonize with the test speeds prescribed in Euro NCAP's protocol for the LVD test scenario if it ultimately adopts higher test speeds, and asked that the SV-to-POV headway be increased for each higher speed test condition based on field-representative distances or TTCs. Subaru recommended that the Agency maintain vehicle headway in LVD testing at a spacing equivalent to 1.0 second (instead of 12 m (39.4 ft.), as proposed) regardless of test speed. FCA also favored higher speed assessments for the LVD test scenario. However, FCA stated that the SV-to-POV headway should be adjusted for each speed to reflect a 0.9 second following distance, asserting this following distance is typical of real-world driving.

POV Deceleration Magnitude for LVD Test Scenario

Most commenters addressing this issue favored a 0.5g POV deceleration for the LVD CIB test instead of 0.6g, with TRC citing issues with repeatability when attempting to tune the GVT braking system to operate at a deceleration greater than 0.5g. Although it acknowledged that new robotic platforms make tuning easier, TRC also suggested that they are expensive and require modification of existing equipment before they can be utilized. BMW also cited robotic platform operational limits and tire wear as reasons not to adopt a 0.6g POV deceleration requirement. HATCI favored a 0.5g deceleration magnitude because of proven repeatability and minimal equipment damage. The commenter recommended that the Agency increase the vehicle headway if it chooses to adopt a 0.6g POV deceleration instead.

GM and Auto Innovators supported a 0.5g deceleration, asserting that this is a “common” deceleration level (based on crash data reviewed by NHTSA) and therefore “realistic.” Both commenters mentioned that a 0.6g deceleration has not been shown to induce differences in vehicle performance, but can cause problems with test equipment (based on experience in conducting Euro NCAP tests at 6 m/s

). In a similar vein to the repeatability concerns mentioned by others, GM also noted that China NCAP no longer generally performs LVD tests (and other consumer groups are expected to follow suit) because they are difficult to conduct and test results for a given vehicle model are often widely variable.

A few commenters expressed conditional support for the higher POV deceleration. Specifically, Honda and Auto Innovators offered support for adoption of a 0.6g deceleration if crash data indicates such a limit is more representative of driver braking in real-world crashes. Auto Innovators added that the Agency must also ensure testing tolerances. FCA suggested that a 0.6g deceleration may be acceptable if NHTSA wanted to “reduce validation effort.”

Intel expressed support for adopting a 0.6g deceleration criterion for the LVD CIB test to harmonize with other entities and regulations and thus reduce test burden. The company stated that, considering the tolerance currently prescribed for the POV deceleration, the

difference between 0.5g and 0.6g is small. Bosch also supported adoption of a 0.6g deceleration magnitude, as did Tesla. However, Tesla suggested that in lieu of a single POV deceleration of 0.5 or 0.6g, as proposed, the Agency should adopt deceleration magnitudes of -2 m/s

and -6 m/s

(approximately 0.2 and 0.6g), respectively, for each test speed to harmonize with Euro NCAP. Advocates shared this opinion.

Agree With Removal of DBS Tests

MEMA, Subaru, and HATCI agreed with the Agency's proposal to remove the DBS test scenarios from NCAP's AEB test matrix, with the latter commenter suggesting that CIB and DBS functionality may overlap at certain speeds such that DBS functionality would be redundant when CIB is activated. HATCI therefore suggested that, if the Agency decided to continue conducting separate DBS assessments in NCAP, such tests should only be performed when a vehicle exhibits a test failure during CIB testing for that condition, as this would reduce test burden. Rivian remarked that the DBS testing was unnecessary because a vehicle's CIB system will activate and slow the vehicle when the braking imparted by DBS is insufficient. Subaru recommended removal or replacement of any ADAS test that currently has a high rate of passing results if adoption rates for the related ADAS technology are also high.

Retain Some or All DBS Tests

Several commenters expressed that the Agency should continue to conduct DBS assessments in NCAP because DBS affords additional safety benefits compared to CIB. ZF Group favored retaining the DBS tests to ensure that vehicles continue to be equipped with DBS, noting that DBS systems “can react earlier in critical situations.” Similarly, CAS asserted that DBS “can provide additional safety margin.”

Other commenters recommended that NHTSA continue DBS testing to ensure system functionality. Advocates, GM, and Auto Innovators suggested the Agency should (Advocates and GM), or could (Auto Innovators), continue to conduct DBS tests to ensure that brake pedal application does not override AEB system functionality in general. NTSB also agreed that DBS functionality should be verified and supported NHTSA's alternative proposal to retain DBS testing in NCAP and conduct LVM and LVS testing at higher speeds (70 and 80 kph (43.5 and 49.7 mph)).

GM and Auto Innovators also recommended other options centered on performing DBS tests only at certain speeds that NHTSA could adopt to reduce the burden of DBS testing. GM noted that China NCAP performs CIB tests at lower speeds and DBS tests at higher speeds. The automaker suggested that for speeds higher than 40 kph (24.9 mph) the Agency could alternate between CIB and DBS testing. Auto Innovators stated that DBS testing would be unnecessary in situations where the CIB system provides complete avoidance in all tests, noting that the Agency could simply assume DBS performance and apply points for both systems equally. Auto Innovators also stated that each assessed test speed should afford equal weighting for both CIB and DBS, noting it should not be the case that one test speed carries twice the weight simply because both systems are assessed at that speed, whereas another test speed carries less weight because only one of the two systems is assessed at that speed. Both GM and Auto Innovators noted, similar to HATCI, that NHTSA could conduct CIB tests until the system can no longer provide full avoidance and then begin DBS testing for the next subsequent higher test speed. If the CIB system was able to provide complete avoidance at all test speeds, then the commenters suggested that DBS testing could be repeated for only the maximum test speed to ensure system functionality. GM also noted that for 2023 Euro NCAP removed the DBS tests for LVM and LVD from their assessment and going forward it will only perform the DBS test for the LVS scenario. Finally, Auto Innovators encouraged the Agency to reduce the number of test scenarios and evaluate FCW during DBS testing (

i.e.,

record the time of the FCW) to further reduce test burden.

Like other commenters' suggestions, FCA and Intel recommended that the Agency continue to perform DBS tests in NCAP for higher test speeds where the CIB system does not afford full crash avoidance, with Intel suggesting that it may be appropriate to start the DBS tests at 60 kph (37.3 mph) to harmonize with Euro NCAP. IDIADA also suggested that the Agency only perform higher speed DBS tests. Similarly, Toyota, like the NTSB, was supportive of the Agency conducting LVS and LVM DBS tests at only the highest test speeds proposed for CIB—70 and 80 kph (43.5 and 49.7 mph), respectfully. However, Toyota did not support conducting the LVD DBS test at 80 kph (49.7 mph) since NHTSA stated that it had not conducted testing at this speed due to equipment and test track limitations. Intel expressed similar concerns, stating that manufacturers may not be able to issue the FCW early enough to achieve brake activation one second after the issuance of the FCW as NHTSA proposed for the LVD tests.

Unlike those commenters who expressed that it was sufficient for the Agency to only conduct high-speed DBS assessments or alternate CIB and DBS assessments for incremental speeds, Honda stated that it is most appropriate to conduct CIB and DBS tests at the same test speeds. Honda asserted that evaluating DBS performance only at higher test speeds may skew performance ratings (similar to what Auto Innovators stated) and not accurately convey the real-world safety benefits DBS provides at lower test speeds. Since CIB and DBS address different safety needs (

i.e.,

the driver is either not responsive, or responsive, respectively, to an imminent collision), the automaker indicated that it is imperative to ensure ratings reflect the benefits afforded by both technologies. Accordingly, Honda, like Auto Innovators, suggested that if the Agency moves forward with such an approach, vehicles should be awarded credit for lower speed DBS tests as well (even though they would only be tested for CIB and not DBS) if the vehicle received passing results for the CIB system at the lower test speeds. Honda asserted this credit would be appropriate, noting that DBS systems should afford equivalent or higher performance than CIB systems when tested at the same speeds. Bosch similarly responded that it would be appropriate to cover the entire speed range by performing one test per scenario and incrementing speeds for each separate scenario by 10 kph (6.2 mph) if NHTSA decided to continue to perform separate DBS assessments and if there are benefits to increasing both CIB and DBS test speeds. CAS also noted that if the Agency imposed higher test speeds for LVD CIB assessments, those same test speeds should be used to assess DBS. The group stated that DBS activation was highly likely for the LVD scenario and all technologies that may contribute to a given scenario/crash outcome should be assessed. Likewise, Tesla asserted that DBS testing should not be reserved only for higher-speed assessments and should be conducted using the same test specifications (speed, headway, and POV deceleration) as the corresponding CIB tests.

Advocates encouraged NHTSA to select the appropriate number of tests and test speeds to ensure acceptable performance across a range of conditions, including those that would be expected during real-world driving.

Response to Comments and Agency Decisions

NHTSA's decision regarding CIB and DBS testing specifics can be found in the following sections as well as in Tables 12 and 13.

CIB Test Speeds for the LVS Test Scenario

NHTSA will proceed with assessing CIB performance in NCAP's LVS scenario using the proposed SV test speeds and increments. The Agency will initiate the LVS test series at the lowest vehicle test speed, 40 kph (24.9 mph), and test speeds will increase in increments of 10 kph (6.2 mph) as each test condition's criteria are met (

i.e.,

no SV-to-POV contact is observed), up to and including the 80 kph (49.7 mph) test condition.

Although several commenters, including Advocates, recommended that the Agency set the minimum LVS test speed to 10 kph (6.2 mph) to harmonize with Euro NCAP, the Agency asserts a 40 kph (24.9 mph) minimum LVS test speed is appropriate for NCAP testing. As noted in Auto Innovators' comments to the March 2022 RFC notice, Volpe's review of 2011-2015 crash data sets showed that, for rear-end crashes, only 2 percent of fatalities and 6 percent of injuries occurred on roadways

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.