DEPARTMENT OF TRANSPORTATION National Highway Traffic Safety

This document is scheduled to be published in the Federal Register on 06/22/2022 and available online at federalregister.gov/d/2022-12860, and on govinfo.gov
DEPARTMENT OF TRANSPORTATION National Highway Traffic Safety Administration
49 CFR Part 563 [Docket No. NHTSA-2022-0021]
RIN 2127-AM12 Event Data Recorders
AGENCY: National Highway Traffic Safety Administration (NHTSA), Department of Transportation (DOT). ACTION: Notice of proposed rulemaking (NPRM). SUMMARY: NHTSA is proposing to amend its regulations regarding Event Data Recorders (EDRs) to extend the EDR recording period for timed data metrics from 5 seconds of pre-crash data at a frequency of 2 Hz to 20 seconds of pre-crash data at a frequency of 10 Hz (i.e., increase from 2 samples per second to 10 samples per second). This NPRM begins the process of fulfilling the mandate of the Fixing America’s Surface Transportation Act (FAST Act) to establish the appropriate recording period in NHTSA’s EDR regulation. DATES: You should submit your comments early enough to be received not later than [INSERT DATE 60 DAYS AFTER DATE OF PUBLICATION IN THE FEDERAL REGISTER]. We are proposing an effective date of the first September 1st one year from the publication of the final rule. ADDRESSES: You may submit comments to the docket number identified in the heading of this document by any of the following methods:  Federal eRulemaking Portal: Go to https://www.regulations.gov. Follow the online
instructions for submitting comments.  Mail: Docket Management Facility: U.S. Department of Transportation, 1200 New Jersey
Avenue S.E., West Building Ground Floor, Room W12-140, Washington, D.C. 20590-0001.

 Hand Delivery or Courier: 1200 New Jersey Avenue S.E., West Building Ground Floor, Room W12-140, between 9 a.m. and 5 p.m. ET, Monday through Friday, except Federal holidays.
 Fax: 202-493-2251. Instructions: All submissions must include the agency name and docket number. Note
that all comments received will be posted without change to http://www.regulations.gov, including any personal information provided. Please see the Privacy Act discussion below. We will consider all comments received before the close of business on the comment closing date indicated above. To the extent possible, we will also consider comments filed after the closing date.
Docket: For access to the docket to read background documents or comments received, go to https://www.regulations.gov at any time or to 1200 New Jersey Avenue, S.E., West Building Ground Floor, Room W12-140, Washington, D.C. 20590, between 9 a.m. and 5 p.m., Monday through Friday, except Federal Holidays. Telephone: 202-366-9826.
Privacy Act: In accordance with 5 U.S.C. 553(c), DOT solicits comments from the public to better inform its decision-making process. DOT posts these comments, without edit, including any personal information the commenter provides, to www.regulations.gov, as described in the system of records notice (DOT/ALL-14 FDMS), which can be reviewed at www.transportation.gov/privacy. In order to facilitate comment tracking and response, we encourage commenters to provide their name, or the name of their organization; however, submission of names is completely optional. Whether or not commenters identify themselves, all timely comments will be fully considered.
Confidential Business Information: If you wish to submit any information under a claim of confidentiality, you should submit three copies of your complete submission, including the information you claim to be confidential business information, to the Chief Counsel, NHTSA, at the address given under FOR FURTHER INFORMATION CONTACT. In addition, you

should submit two copies, from which you have deleted the claimed confidential business information, to the Docket at the address given above. When you send a comment containing information claimed to be confidential business information, you should include a cover letter setting forth the information specified in our confidential business information regulation (49 CFR part 512). FOR FURTHER INFORMATION CONTACT: For technical questions, please contact Ms. Carla Rush, Office of Crashworthiness Standards, National Highway Traffic Safety Administration, 1200 New Jersey Avenue S.E., Washington, D.C., 20590 (telephone: 202-3661740, fax: 202-493-2739). For legal questions, please contact Ms. Sara Bennett, Office of Chief Counsel, National Highway Traffic Safety Administration, 1200 New Jersey Avenue S.E., Washington, D.C., 20590 (telephone: 202-366-2992, fax: 202-366-3820). SUPPLEMENTARY INFORMATION: Table of Contents
I. Background A. Overview of Event Data Recorder Technology and Regulatory History B. The Fixing America’s Surface Transportation Act C. Event Data Recorders Duration Study
II. Proposal A. Pre-crash EDR Recording Duration B. Pre-crash EDR Recording Frequency C. Benefits D. Costs E. Lead Time
III. Rulemaking Analyses and Notices I. Background A. Overview of Event Data Recorder Technology and Regulatory History
Event data recorders (EDRs) are devices that are used to record safety information about motor vehicle crashes immediately before and during a crash. The recorded information can aid crash investigators to assess the performance of specific safety equipment before and during a

crash. This information can assist the agency and others with identifying potential opportunities
for safety improvement in vehicles already on the road, as well as contributing to improve future
vehicle designs and more effective safety regulations. This information could also aid first
responders in assessing the severity of a crash and estimating the probability of serious injury in
vehicles equipped with Advanced Automatic Crash Notification (AACN) systems and can
improve defect investigations and crash data collection quality. (See the 2006 final rule
establishing the EDR regulation (discussed below) for further details. (71 FR 50998.)
In August 2006, NHTSA established 49 CFR part 563 (part 563), which sets forth
requirements for data elements, data capture and format, data retrieval, and data crash
survivability for EDRs. (71 FR 50998.) Part 563 does not mandate that vehicles have EDRs, but
is instead an “if equipped” standard that applies only to light vehicles required to have frontal air bags that a manufacturer chooses to voluntarily equip with EDRs.1 Part 563 ensures that all
EDRs subject to the regulation capture the same core set of data elements in a crash, standardizes
the parameters (format, duration, etc.) of captured data elements, and sets minimum requirements for data survivability.2 Part 563 further requires that manufacturers of vehicles with EDRs that are subject to part 563 make commercially available a tool for the purpose of imaging3 the data
collected by the EDR.
Tables I and II of part 563 list the various data elements that are covered under the
standard. Table I lists data elements that all EDRs subject to part 563 are required to record,
along with the recording interval (duration) and sampling frequency. Table II lists data elements
that EDRs subject to part 563 are not required to record, but that are subject to part 563 if they
1 In 2012, NHTSA proposed to convert part 563’s “if equipped” requirements for EDRs into a new Federal Motor Vehicle Safety Standard (FMVSS) mandating the installation of EDRs in most light vehicles. The NPRM did not propose making any changes to the current EDR regulation’s performance requirements, including those for the required data elements (77 FR 74145). In 2019, NHTSA withdrew this proposal due to the near universal installation of EDRs on light vehicles (84 FR 2804). 2 Part 563 requires EDR data to survive the crash tests in FMVSS Nos. 208, “Occupant crash protection,” and 214, “Side impact protection.” 3 For the purposes herein, we are using the term “imaging” to refer to the process by which data are retrieved from an EDR. When imaging the data on an EDR, the original data set remains intact and unchanged in the memory banks of the EDR.

are recorded. Table II also provides the recording interval (duration) and sampling frequency for each listed data element. In addition, all data elements in Tables I and II must be reported according to the range, accuracy, and resolution in Table III. As is relevant to this rulemaking, several data elements in both Table I and Table II must be captured for a duration of 5 seconds prior to the crash (speed, engine throttle, service brake, engine RPM, ABS activity, stability control, steering input). NHTSA established this 5-second duration because the agency concluded that it would be long enough to ensure the usefulness of the data in crash reconstruction while also minimizing the risk that the data capture process would over-tax the EDR’s microprocessor, which could cause a malfunction that could lead to a loss of data.4
Part 563 became fully effective on September 1, 2012. The agency estimates that 99.5 percent of model year 2021 passenger cars and other vehicles with a gross vehicle weight rating (GVWR) of 3,855 kilograms (kg) (8,500 pounds) or less have part 563 compliant EDRs.5 B. The Fixing America’s Surface Transportation Act
Section 24303 of the Fixing America’s Surface Transportation Act (FAST Act), Pub. L. No. 119-14 (Dec. 4, 2015), requires NHTSA to conduct a study “to determine the amount of time event data recorders installed in passenger motor vehicles should capture and record for retrieval [of] vehicle-related data in conjunction with an event in order to provide sufficient information to investigate the cause of motor vehicle crashes,” and to submit a report containing the findings of this study to Congress. Further, within two years of submitting this report to Congress, NHTSA “shall promulgate regulations to establish the appropriate period during which event data recorders installed in passenger motor vehicles may capture and record for retrieval vehiclerelated data to the time necessary to provide accident investigators with vehicle-related information pertinent to crashes involving such motor vehicles.”
4 NHTSA had originally proposed an 8-second duration in the NPRM. 69 FR 32942 (June 14, 2004). However, NHTSA decided to reduce the duration in response to public comments. 71 FR 51020 (Aug. 28, 2006). 5 In the 2012 NPRM it was estimated that about 92 percent of model year 2010 light vehicles had some EDR capability.

As discussed in detail in section C below, NHTSA completed the Event Data Recorders Duration Study required by Section 24303. On September 28, 2018, NHTSA submitted a Report to Congress summarizing the results of the study to the House Committee on Energy and Commerce and the Senate Committee on Commerce, Science, and Transportation.6 This NPRM begins the process of promulgating regulations to establish appropriate EDR data recording durations as mandated under the FAST Act. C. Event Data Recorders Duration Study
To meet the agency’s obligations under Section 24303 of the FAST Act, NHTSA contracted with researchers at Virginia Polytechnic Institute and State University (Virginia Tech) to conduct a study to determine the recording duration that would be necessary for EDRs to provide sufficient vehicle-related data to investigate the cause of motor vehicle crashes (the “EDR Duration Study”).7 Because crash investigators must understand the events leading up to a crash to determine crash causation, the EDR Duration Study sought to determine the necessary recording duration to encompass a vehicle’s relevant maneuvers for three crash types that could benefit from more than 5 seconds of pre-crash recording time: rear-end, intersection, and road departure crashes.8 For all three of these crash types, the study hypothesizes that it is necessary to capture the initiation of crash avoidance maneuvers by the driver, if any, to better determine causation. The specific crash avoidance maneuvers examined in the study were the driver’s release of the accelerator, and the initiation of pre-crash braking and evasive steering. In addition, for intersection crashes, it is also necessary to capture vehicle data for the duration that the vehicle is approaching and traversing an intersection, since intersection crashes often have complex causes that extend back further than when the driver begins making crash avoidance
6 National Highway Traffic Safety Administration. (2022, March) Results of event data recorders pre-crash duration study: A report to Congress (Report No. DOT HS 813 082A). 7 Chen, R. J., Tatem, W. M., & Gabler, H. C. (2022, March) Event data recorder duration study (Appendix to a Report to Congress. Report No. DOT HS 813 082B). National Highway Traffic Safety Administration. 8 Ibid. Phase I did not analyze lane departure behavior prior to a road departure crash.

maneuvers (e.g., a rolling stop at the stop sign or any indication of erratic driving during the
approach).
The EDR Duration Study was conducted in two phases. Phase I provided an estimate of
how often EDRs fail to record a sufficient duration of pre-crash data; however, this analysis did
not provide insight into what duration beyond 5 seconds of pre-crash data is needed to capture
crash causation. The emphasis in Phase II was on using driver actions in normal driving to
determine the complete duration of driver pre-crash actions.
Phase I used cumulative distributions of the EDR data pulled from NHTSA’s National
Automotive Sampling System Crashworthiness Data System (NASS-CDS) database9,10 to
estimate how frequently the current 5-second EDR duration requirement failed to capture the
initiation of pre-crash driver maneuvers in rear-end, intersection, and road departure crashes.
The Phase I study also estimated how frequently the 5-second duration did not capture the
vehicle’s approach and traversal phase of an intersection or road departure.11 The results of
Phase I helped establish the need for an increase in the EDR recording duration by proving the
inadequacy of the 5-second recording duration.
For Phase II of the EDR Duration Study, researchers used data from two previously
conducted naturalistic driving studies (NDS) to understand the complete duration (5 seconds or
longer) of driver pre-crash actions and estimate the recording duration that would be necessary to
capture the initiation of these actions in the same three types of crash scenarios examined in
Phase I.12
9 NASS-CDS was utilized because it contains over 9,000 EDR downloads. NASS-CDS sampling weights were used in the calculations unless otherwise specified. 10 The National Motor Vehicle Crash Causation Study (NMVCCS) was also analyzed, but due to the small sample size distributions of pre-crash maneuvers were not conducted. However, the NMVCCS dataset was analyzed to determine the frequency of vehicle malfunctions in crashes, and none of the 50 vehicles in the final dataset were reported as having a vehicle malfunction by the on-site investigator. 11 Intersection traversal time is not directly measured by a vehicle’s EDR; researchers calculated traversal time for this study by reconstructing crash events. 12 The two studies used were a 100-Car NDS conducted by Virginia Tech Transportation Institute [Neale, V.L., Klauer, S.G., Knipling, R.R., Dingus, T.A., Holbrook, G.T., and Petersen, A. (2002) The 100-Car Naturalistic Driving Study, Phase 1 – Experimental Design. (DOT Report HS 809 536) Washington, DC: National Highway Traffic Safety Administration], and the Second Strategic Highway Research Program (SHRP-2) NDS conducted by

1. Phase I Study The purpose of the Phase I study was to determine the frequency with which EDRs with a
5-second recording duration fail to record a sufficient duration of pre-crash data to determine crash causation for rear-end,13 intersection, and roadway departure crashes. Using EDR data pulled from NHTSA’s NASS-CDS database from 2000-2015,14 Phase I researchers examined 1,583 raw cases. Of these cases, 329 were rear-end crashes, 839 were intersection crashes, and 415 were road departure crashes. Based on these cases, researchers found that the current 5second recording duration required under part 563 failed to capture the initiation of driver crash avoidance maneuvers for a certain percentage of all three selected crash types. These findings are good indications that a 5-second pre-crash recording duration is inadequate if the goal is to capture the complete pre-crash time history – principally the driver’s pre-crash behavior – so that NHTSA, crash investigators, and manufacturers can better understand the crash causation.
To determine whether the EDR had captured an entire crash event, Phase I researchers examined the status of the available EDR pre-crash data elements – vehicle’s accelerator pedal, service brakes, and steering angle – over the course of the 5 seconds of data. The initiation of the crash event would be indicated by the release of the accelerator pedal, the initiation of braking, or a change in the steering angle from zero degrees. Again, cumulative distributions of the data were used to determine the percentage of crashes where the initiation of the driver’s precrash maneuver falls outside the 5-second pre-crash recording duration.
For rear-end crashes, the Phase I researchers found that the current 5-second EDR recording duration failed to capture 9% of accelerator pedal releases, 35% of pre-crash braking initiations, and 80% of evasive steering initiations. For intersection crashes, the 5-second recording duration failed to capture 4% of accelerator pedal release instances, 35% of pre-crash
the Transportation Research Board of The National Academies, [Hankey, J. M., M.A. Perez, and J.A. McClafferty. Description of the SHRP 2 naturalistic database and the crash, near-crash, and baseline data sets, Task Report, Virginia Tech Transportation Institute, Blacksburg, VA, 2016]. 13 For rear-end crashes the striking vehicle was examined. 14 Up until 2015, NASS was comprised of two probability sampling systems: the General Estimates System (GES) and CDS. Then in 2016, the Crash Investigation Sampling System (CISS) replaced the CDS.

braking initiations, and 64% of evasive steering initiations. In addition, it did not capture 13% of initial intersection boundary crossings.15 Finally, for road departure crashes, the 5-second recording duration failed to capture 8% of accelerator pedal releases, 35% of pre-crash braking initiations, and 88% of evasive steering initiations. However, the analysis of road departure traversal time shows that, in nearly all road departure events, the time period between initial road departure to final rest was less than 5 seconds, which indicates that the pre-crash maneuvers that were not recorded by the 5-second duration likely took place before the vehicle went off-road. Table 1 below summarizes the Phase I findings.

Table 1. Percentage of Events for which 5 Seconds of EDR Recording Duration was Insufficient from NASS-CDS

Driver Pre-crash Maneuver

Rear-End
Intersection
Road Departure

Braking Input Steering Input Accelerator Release Braking Input Steering Input Accelerator Release Braking Input Steering Input Accelerator Release

EDR Failed to Record Maneuver Initiation
35% 80% 9% 35% 64% 5% 35% 88% 8%

Based on these findings, the EDR Duration Study concluded that in many cases, the 5second recording duration may not be sufficient to determine the factors that led to the crash or the pre-crash actions taken by the driver to avoid the collision, meaning that EDRs currently would not always provide investigators crash-related information that could assist in the determination of crash causation.
2. Phase II Study

15 Intersection boundaries were used as a reference point to divide the approach and traversal phase of an intersection (e.g., the edge of the stop bar or cross walk marking closest to the center of the intersection was used as the boundary).

The purpose of the Phase II study was to determine an appropriate EDR recording duration to provide crash investigators with sufficient data to determine crash causation. NDS data were analyzed to understand the complete duration (5 seconds or longer) of driver pre-crash actions in car following, intersection traversal, and lane departure crashes. The Phase II study used data from two previously conducted naturalistic driving studies: a 2002 100-Car study conducted by Virginia Tech, and the 2016 Second Strategic Highway Research Program (SHRP2) NDS conducted by the Transportation Research Board of The National Academies.16 To estimate the recording duration needed to capture the initiation of a crash event, the Phase II researchers analyzed near-miss driving events as proxies for actual crash avoidance driving maneuvers that were analyzed in the Phase I study.17 The main finding in Phase II of the study was that 20 seconds of pre-crash data would encompass the 90th percentile recording duration required for the three crash modes and the crash avoidance maneuvers analyzed. A “90th percentile recording duration” means that, based on the cumulative distributions for all three crash modes and crash avoidance maneuvers analyzed, a minimum of 20 seconds of pre-crash data recording is necessary to investigate crash causation, as this period captures the driver precrash actions in 90% of the dataset.18
To determine the recording duration needed to capture rear-end crashes, the Phase II researchers examined the duration of “car following” braking events from the 100-car NDS. By looking at the time duration between the start of the braking event (i.e., when the driver applies the brake) and the vehicle’s closest approach to the lead vehicle, the Phase II researchers were able to approximate the duration of a rear-end crash event. The results were different depending on whether the lead vehicle was stopped or travelling (e.g., events with stopped lead vehicle are
16 A naturalistic driving study is a research method that involves equipping vehicles with unobtrusive cameras and instrumentation to record real-world driver behavior and performance. 17 Phase II of the study assumed that the driver’s behavior in near-miss driving events would correlate to actual crash avoidance driving maneuvers. 18 This duration is influenced heavily by the inclusion of intersection crashes. Without the inclusion of intersection crashes 12.3 seconds of data would encompass the 90th percentile recording duration for rear-end and road departure crashes.