[Federal Register Volume 89, Number 248 (Friday, December 27, 2024)]
[Notices]
[Pages 105675-105680]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2024-30860]
=======================================================================
-----------------------------------------------------------------------
DEPARTMENT OF TRANSPORTATION
Federal Motor Carrier Safety Administration
[Docket No. FMCSA-2023-0071]
Parts and Accessories Necessary for Safe Operation; Application
for an Exemption From Waymo LLC (USDOT #3000336) and Aurora Operations,
Inc. (USDOT #3441156)
AGENCY: Federal Motor Carrier Safety Administration (FMCSA), Department
of Transportation (DOT).
ACTION: Notice of final disposition; denial of exemption.
-----------------------------------------------------------------------
SUMMARY: The Federal Motor Carrier Safety Administration (FMCSA or the
Agency) publishes its decision to deny an application from Waymo LLC
(Waymo, USDOT #3000336) and Aurora Operations, Inc. (Aurora, USDOT
#3441156) (together, Applicants), requesting an exemption from certain
Federal Motor Carrier Safety Regulations (FMCSRs) related to
requirements to place specific types of warning devices at prescribed
locations around commercial motor vehicles (CMVs) stopped on the
traveled portion or shoulder of a highway for any cause other than
necessary traffic stops and requirements that lamps on CMVs be steady
burning. Applicants requested that they and ``other similarly situated
companies'' be permitted to instead use ``Cab-Mounted Warning Beacons''
(including ``variants'' and ``any configuration of similar
effectiveness'') when the CMV is operated by a Level 4 Automated
Driving System (ADS) either without a human on board or with a human on
board when testing the warning beacons. Safety is FMCSA's highest
priority. The Agency embraces USDOT's Innovation Principles, including
adapting as technology changes and supporting technologies that further
our policy goals. While the application and the public comments show
promise for alternative warning devices to provide safety benefits for
warning motorists of a stopped CMV under certain conditions, the
present application does not demonstrate how Applicants or other
proposed exempted parties would ensure an equivalent or greater level
of safety than would be achieved absent the exemption. The application
does not provide sufficient details about proposed alternative devices,
and the limited data presented does not support a likely equivalent
level of safety for a national, industry-wide exemption for all
companies operating autonomous CMVs. If Applicants can reasonably
address the reasons for the denial, Applicants may resubmit an
exemption application.
FOR FURTHER INFORMATION CONTACT: Mr. David Sutula, Vehicle and Roadside
Operations Division, Office of Carrier, Driver, and Vehicle Safety,
FMCSA; 1200 New Jersey Avenue SE, Washington, DC 20590-0001; (202) 366-
9209; [email protected].
SUPPLEMENTARY INFORMATION:
I. Viewing Comments and Documents
To view comments, go to www.regulations.gov, insert the docket
number ``FMCSA-2023-0071'' in the keyword box, and click ``Search.''
Next, sort the results by ``Posted (Newer-Older),'' choose the first
notice listed, click ``Browse Comments.''
To view documents mentioned in this notice as being available in
the docket, go to www.regulations.gov, insert the docket number
``FMCSA-2023-0071'' in the keyword box, click ``Search,'' and choose
the document to review.
If you do not have access to the internet, you may view the docket
online by visiting Dockets Operations on the ground floor of the DOT
West Building, 1200 New Jersey Avenue SE, Washington, DC 20590, between
9 a.m. and 5 p.m. ET, Monday through Friday, except Federal holidays.
To be sure someone is there to help you, please call (202) 366-9317 or
(202) 366-9826 before visiting Dockets Operations.
II. Legal Basis
FMCSA has authority to grant exemptions from FMCSRs where the
Agency finds that ``such exemption would likely achieve a level of
safety that is equivalent to, or greater than, the level that would be
achieved absent such exemption.'' 49 U.S.C. 31136(e); 49 U.S.C.
31315(b)(1).
An exemption application must provide specific information relevant
to the scope of the exemption sought including, but not limited to, the
provisions from which the person requests exemption, the reason the
exemption is needed, the time period during which the requested
exemption would apply, an analysis of the safety impacts the requested
exemption may cause, and the specific countermeasures the person would
undertake to ensure an equivalent or greater measure of safety than
would be achieved absent the requested exemption. 49 U.S.C.
31315(b)(5); 49 CFR 381.310.
Upon receipt of an exemption request, FMCSA must publish a notice
of the request in the Federal Register and provide the public an
opportunity to inspect the applicant's safety analysis and any other
relevant information known to the Agency, and to comment on the
request. See 49 U.S.C. 31315(b)(6)(A); 49 CFR 381.315(a). The Agency
reviews the application and any public comments submitted and
determines whether to grant or deny the request. 49 CFR 381.315(b). The
Agency must publish its decision in the Federal Register. Id. If the
request is granted, the decision will identify the name of the person
or class of persons granted the exemption, the provisions from which
the person is exempt, the effective period, and the terms and
conditions of the exemption. 49 U.S.C. 31315(b)(6)(B); 49 CFR
381.315(c)(1). If the request is denied, the Agency will publish the
name of the person denied the exemption and the reasons for such
denial. 49 U.S.C. 31315(b)(6)(C); 49 CFR 381.315(c)(2). A granted
exemption may be renewed, upon request, for subsequent 5-year periods.
49 U.S.C. 31315(b)(2); 49 CFR 381.300(b). If an exemption request is
denied, and the applicant can reasonably address the reason for the
denial, the Agency may allow the applicant to resubmit the application.
49 U.S.C. 31315(b)(3); 49 CFR 381.317.
[[Page 105676]]
III. Background
A. Current Regulatory Requirements Under the FMCSRs
Applicants request an exemption from certain FMCSRs related to
requirements for placing warning devices around a stopped CMV,
requirements that exterior lamps be steady burning, and requirements
that specific types of warning devices be used.
The applicable FMCSRs require the driver of a CMV stopped on the
traveled portion or the shoulder of a road for any cause other than a
necessary traffic stop to activate hazard warning signal flashers and
place required warning devices as soon as possible but within ten
minutes, at specified locations behind and in front of the stopped CMV.
49 CFR 392.22(b). They also specify placement of warning devices in
certain circumstances, such as during daylight hours, or where devices
may be obstructed from view (e.g., when stopped within 500 feet of a
curve or the crest of a hill). Id.
The FMCSRs also require that all exterior lamps be steady burning,
with exceptions not relevant here. 49 CFR 393.25(e).
Finally, the FMCSRs specify the types and number of warning devices
to be used for stopped vehicles, namely 3 bidirectional emergency
reflective triangles, or at least 6 fusees or 3 liquid-burning flares.
Other warning devices may be used in addition to required devices, as
long as they do not reduce the effectiveness of required devices. 49
CFR 393.95(f).
B. Applicants' Requested Exemption
Applicants request an exemption on behalf of themselves and
``similarly situated companies'' from requirements of 49 CFR 392.22(b),
393.25(e), and 393.95(f). Application (App.) at 1. They request that
CMVs operated by a Level 4 ADS be permitted to use ``Cab-Mounted
Warning Beacons'' instead of FMCSR-specified warning devices when the
vehicles are operating either without a human on board, or with a human
on board when testing the warning beacons. Id.
Applicants define Cab-Mounted Warning Beacons as ``cab-mounted
warning devices, consisting of both forward- and rearward-facing amber
flashing lights mounted at a height above the upper edge of the
sideview mirrors.'' App. at 2. The proposed warning beacons are amber
colored and meet certain Society of Automotive Engineers (SAE)
requirements. Id. at 7-8. These warning beacons would ``flash at a
different rate than the required hazard warning signal flashers to
increase conspicuity of the stopped CMV,'' which Applicants contend
would not interfere with or impair the effectiveness of hazard warning
signal flashers. Id. at 11. Applicants propose that at least one
rearward-facing light be mounted on each side of the cab ``at some
point on or above the upper edge of the sideview mirrors and below the
top edge of the cab,'' and that at least one forward-facing light be
mounted ``similarly high'' on the front of the cab. Id. at 9-10.
Applicants provide a diagram illustrating areas where the beacons
``could be mounted,'' noting that the description ``purposely allows
for flexibility with regard to the mounting location of the beacons so
long as the beacons are positioned at a sufficiently high point on the
cab.'' Id. at 10 n.19. Applicants propose that ``[t]he exemption, if
granted, should permit any configuration of similar effectiveness.''
Id. at 10.
Applicants state that the exemption is needed because compliance
with 49 CFR 392.22(b) is ``not feasible for autonomous CMVs without a
human on board,'' and because proposed Cab-Mounted Warning Beacons are
not steady burning (as required under 49 CFR 393.25(e)) and are not
among the approved warning devices under 49 CFR 393.95(f). Applicants
posit that if the exemption is not granted, the deployment of
autonomous CMVs without a human driver on board will be ``sufficiently
impeded,'' and that alternatives to the exemption would be
unnecessarily costly. App. at 14-15.
C. Research Submitted by Applicants in Support of Proposed Beacons
To support the asserted equivalent level of safety of Cab-Mounted
Warning Beacons, Applicants submitted two reports on testing certain
beacon ``variants.'' App. at 9. Waymo sponsored a study by the Virginia
Tech Transportation Institute (VTTI) (``Stopped Automated Commercial
Motor Vehicle Warning Device Surrogates,'' prepared for Waymo, LLC, by
Erin Mabry, Susan Soccolich, Kary Meissner, Josh Radlbeck, and Andy
Schaudt, VTTI, August 1, 2022 (the Waymo study)). Aurora conducted what
it calls a ``naturalistic study'' (``Naturalistic Study for Warning
Device Equivalency,'' prepared by Aurora Operations, Inc., October 2022
(the Aurora study)). Id. at 11, apps. B-C.
The Waymo study evaluated whether drivers detected, recognized, and
responded appropriately to certain beacons while traveling on a closed-
circuit test track in daylight and nighttime conditions and
encountering a stopped CMV at three different locations (on the
shoulder, in the lane ahead, and on the right shoulder after a curve).
App. at 12. The study compared driver reactions to Cab-Mounted Warning
Beacons with driver reactions to warning triangles and concluded that
study participants subjectively preferred beacons over triangles for
communicating the presence of a stopped CMV. Waymo study at 20. The
study was conducted in Virginia with 48 participants (24 daylight, 24
nighttime). Id. at 7-8. The report does not describe the weather
conditions.
The Aurora study observed the reactions of over 7,000 passing
vehicles to the presence of certain beacons versus warning triangles.
App. at 13. The study was conducted on public roads with passing
motorists who were unaware of the study and traveling at highway speeds
when they encountered a stopped CMV on the right shoulder. Id., Aurora
study at 3. Aurora used sensors mounted on the stopped CMV to collect
data about motorist responses. Aurora study at 8. Aurora reports that
similar proportions of motorists responded to the presence of beacons
as they did to warning triangles, and that among those responding,
responses occurred at similar distances and motorists increased lateral
separation in similar proportions. App. at 13-14. The Aurora study was
conducted in Texas on Interstate 45 between Exits 258 and 249. Aurora
study at 5. The report does not describe the weather conditions.
Applicants posit that the two reports confirm that the proposed
beacons achieve a level of safety equivalent to or greater than the
level achieved with FMCSR-specified warning triangles placed at
prescribed distances. App. at 14.
IV. Overview of Public Notice and Comments
FMCSA published in the Federal Register on March 9, 2023, notice of
the exemption application and the opportunity for public comment. 88 FR
14665. In response, the Agency received 51 public comments.\1\
---------------------------------------------------------------------------
\1\ FMCSA considered all comments in the docket received through
October 31, 2024. FMCSA notes that it received two comments from the
Association for Uncrewed Vehicle Systems International but is
treating those submissions as one comment because they are
substantively identical. See https://www.regulations.gov/comment/FMCSA-2023-0071-0020 and https://www.regulations.gov/comment/FMCSA-2023-0071-0037.
---------------------------------------------------------------------------
Twenty-five comments generally supported granting the application,
including comments submitted by the Association for Uncrewed Vehicle
Systems International (AUVSI); the American Trucking Associations
(ATA);
[[Page 105677]]
the Autonomous Vehicle Industry Association (AVIA); the Consumer
Technology Association (CTA); Continental Automotive Systems, Inc.;
Daimler Trucks of North America (Daimler); Hirschbach Motor Lines, Inc.
(Hirschbach); Kodiak Robotics (Kodiak); the Motor and Equipment
Manufacturers Association (MEMA); PACCAR Inc.; Stack AV Co. (Stack);
TechNet; Uber Freight US, LLC (Uber Freight); the U.S. Chamber of
Commerce; Volvo Autonomous Solutions NA, Inc. (VAS); Waabi Innovation
US, Inc. (Waabi); Werner Enterprises (Werner); CWK Consulting, LLC; the
Texas Trucking Association (TXTA); the Institute for Safer Trucking
(IST); Isuzu Motors Limited, Isuzu Technical Center of America, Inc.
(Isuzu); the Montana Trucking Association (MTA); the Allegheny
Conference on Community Development (Allegheny Conference); the
National Fraternal Order of Police (NFOP); and an individual commenter.
One commenter, HAAS Alert, was neutral on the proposal, although it
supported updating current standards for hazard warnings and
notifications.
Twenty-five comments generally expressed concerns with granting the
exemption, including comments submitted by the Amalgamated Transit
Union, AFL-CIO (ATU); AWM Associates, LLC; the Motor Vehicle Lighting
Supplier Safety Institute (MVLSSI); the Owner Operator Independent
Drivers Association (OOIDA); the Towing and Recovery Association of
America, Inc. (TRA); the Transportation Trades Department, AFL-CIO
(TTD); the Transport Workers Union of America, AFL-CIO (TWU); the Truck
Safety Coalition (TSC)/Citizens for Reliable and Safe Highways (CRASH)/
Parents Against Tired Truckers (PATT) (joint comment); eleven
individual commenters; and six anonymous commenters.
Generally, commenters supporting the exemption cited the need for a
warning system that does not require human intervention to place
warning devices for ADS-operated CMVs and the need for continued
development of automated technologies. AVIA, for example, argued that
the exemption application proposes a new safety solution that would
avoid the need for human intervention when a Level 4 or 5 ADS-equipped
CMV is stopped on the highway. This position was echoed by comments
from ATA; AUVSI; CTA; CWK Consulting, LLC; Daimler; Hirschbach; Kodiak;
PACCAR Inc.; Stack; TechNet; Uber Freight; the U.S. Chamber of
Commerce; VAS; Waabi; and Werner. These commenters stated that the
exemption would help address regulatory burdens that they believe are
hindering the deployment of ADS-equipped CMVs. AVIA also commented that
it believes the proposed beacons not only address needs of ADS-equipped
vehicles but also could enhance safety if applied to traditional CMVs
by reducing risks to drivers who would otherwise be required to exit
their vehicles to place warning devices. IST expressed concern over the
``significant risk'' to truck drivers placing warning triangles
(particularly in conditions of poor visibility or adverse weather).
Waabi also emphasized that it is ``extremely dangerous'' to require
human drivers to ``get out of a CMV, walk into or next to oncoming
traffic (including at night and during severe weather events), and
proceed 100 feet beyond the vehicle, and then physically deploy a
warning triangle or flare.'' ATA and Kodiak commented that the
exemption would be a positive step toward improving overall roadway
safety. Similar comments from TXTA, Isuzu, MTA, IST, and Allegheny
Conference noted that this would be a step toward enhancing or
improving ``roadway safety'' for ``autonomous trucks and other road
users.'' NFOP stated that granting the exemption would be ``a step
toward collecting data that can support a broader rulemaking effort to
update FMCSA's current regulation on warning devices for all CMVs.''
Commenters generally opposed to granting the exemption cited
concerns about the effectiveness of the proposed beacons and the lack
of evidence in Applicants' studies to address certain safety concerns.
ATU, for example, commented that there was no data demonstrating how
the beacons would perform where there are blind corners blocking the
sight lines of oncoming vehicles, where a vehicle is turned on its
side, or if there is a fire or electrical problems. Other efficacy
concerns included visibility of the beacons due to a single location,
or based on roadway topography or alignment of the truck tractor and
trailer; electrical failure causing failure of the beacons; and
desensitization of the public to flashing beacons or confusion about
what beacons mean (ATU, OOIDA, four individual commenters, one
anonymous commenter, MVLSSI, and TRA). MVLSSI, OOIDA, TSC/CRASH/PATT,
and trade unions (TWU, ATU, and TTD) also commented that Applicants
presented insufficient evidence to demonstrate an equivalent level of
safety, noting that more testing would be needed, with TSC/CRASH/PATT
emphasizing the lack of performance standards for ADS-equipped CMVs.\2\
---------------------------------------------------------------------------
\2\ Among other FMCSA efforts to promote the safe testing and
deployment of ADS-equipped CMVs, a proposed rulemaking, ``Motor
Carrier Operation of Automated Driving Systems (ADS)-Equipped
Commercial Motor Vehicles'' (RIN 2126-AC17), is currently underway.
---------------------------------------------------------------------------
Two individuals commented that the exemption, if granted, should
apply to all CMVs, not just ADS-equipped CMVs. TWU and others called
for FMCSA or DOT to consider larger studies on alternative warning
devices to potentially update existing standards rather than granting
an exemption from the current safety standards, and TTD similarly noted
that the Applicants' request would have been more appropriate as a
petition for rulemaking. Some commenters believed that the exemption
request was too broad, seeking an exemption that would apply to an
entire class of CMVs, rather than limited to those CMVs under the
control of the applicants (see, e.g., ATU, TWU, and TSC/CRASH/PATT),
with some noting that many of these operators currently use different
systems and processes. Commenters also noted that the cited testing
does not account for the safety case for a broader spectrum of
autonomous and non-autonomous CMVs operated by different entities.
Applicants submitted a 2.5-page response to public comments, dated
April 19, 2023 (Applicant's Response). They responded to certain
commenter concerns regarding curves and hills by stating that the Waymo
and Aurora studies evaluated beacons on or after curves, and that
Aurora's study assessed beacons beyond the crest of a hill. They argue
that warning beacons' positioning high on the cab made them effective.
Regarding commenters' concerns about electrical failure, Applicants
responded that beacons ``can and should be designed and installed to
avoid a single-point power failure'' (such as by establishing two power
domains to draw from, although they encouraged FMCSA to remain
technology neutral as to how they are designed). They also noted that
existing warning devices are not immune from failure. They further
contended that the proposed beacons would activate immediately and
thereby provide an immediate warning, unlike specified warning devices,
which may take up to ten minutes to be placed. Applicants did not
submit additional evidence to support these positions.
V. Exemption Decision
Safety is the Agency's highest priority. The Agency embraces the
United States Department of Transportation's Innovation Principles,
[[Page 105678]]
including adapting as technology changes and supporting technologies
that further our policy goals. Having carefully evaluated Applicants'
exemption application, supporting documentation, the public comments,
and the safety implications of Applicants' request, the Agency denies
the application. While the application and public comments suggest
promise for alternative warning devices in some circumstances--and the
Agency encourages continued innovation and analyses in this area--given
the scope of the exemption sought, including the proposed flexible
standard for Cab-Mounted Warning Beacons and the proposed national and
industry-wide exemption for all companies operating autonomous CMVs,
FMCSA finds that the application does not demonstrate that granting the
exemption will likely achieve a level of safety that is equivalent to,
or greater than, the level of safety that would be achieved absent the
exemption. As discussed below, the broad exemption is not supported by
the data presented and lacks necessary monitoring controls to ensure
highway safety. FMCSA notes that this decision does not preclude
Applicants or others from seeking an exemption to use better defined
warning beacons for specified companies in particular locations, as one
of the bases of the Agency's decision here is the broad reach of
Applicants' request.
A. Applicants Do Not Provide Sufficient Details About the Proposed
Alternative Devices and Do Not Demonstrate an Equivalent or Greater
Level of Safety for Those Devices, Particularly for a Nationwide and
Industry-Wide Exemption
To obtain an exemption from the FMCSRs, an applicant must provide
evidence allowing the Agency to conclude that the exemption would
``likely achieve a level of safety that is equivalent to, or greater
than, the level that would be achieved absent such exemption.'' 49
U.S.C. 31315(b)(1). An applicant likewise must include sufficient
details about the specific countermeasures they would take to ensure an
equivalent or greater measure of safety. 49 U.S.C. 31315(b)(5); 49 CFR
381.310(c)(5). The present application fails to do so.
1. Applicants Do Not Provide Sufficient Details About the Proposed
Alternative Warning Devices
Applicants propose Cab-Mounted Warning Beacons consisting of ``at
least one'' rearward-facing light mounted ``at some point on or above
the upper edge of the sideview mirrors'' and ``at least one'' forward-
facing light. App. at 9-10. Applicants provide a diagram of where
beacons ``could be mounted,'' proposing that the description
``purposely allows for flexibility with regard to the mounting
location.'' Id. at 10 n.19. Applicants further propose that the
exemption should ``permit any configuration of similar effectiveness.''
Id. at 10. As several commenters noted, the proposed devices are not
specific enough to ensure an equivalent level of safety. The
variability of the description also does not readily enable the Agency
to monitor the exemption terms and conditions. See 49 U.S.C.
31315(b)(8).
2. Applicants Do Not Demonstrate an Equivalent Level of Safety
Supporting the Full Scope of the Exemption Sought
Applicants' submitted studies do not support an equivalent level of
safety for Cab-Mounted Warning Beacons for the full scope of the
exemption sought.
First, the studies at best address beacons mounted at 110 inches
from the ground (Waymo study at 6) and at an undetermined height
(Aurora study),\3\ and with the particular placement and operational
constraints described in the studies. They do not address the
visibility of beacons for the full range of the ``flexible'' standard
that Applicants propose, or ``any configuration of similar
effectiveness.''
---------------------------------------------------------------------------
\3\ The Aurora study indicates that one light is temporarily
mounted on each side of the cab at ``approximately one foot behind
the sideview mirror'' and ``approximately four inches above the top
of the sideview mirror.'' Aurora study at 3 n.3.
---------------------------------------------------------------------------
Second, while Applicants contend that both studies demonstrated
similar overall reactions from highway users driving by a stopped CMV
for the proposed Cab-Mounted Warning Beacons compared to warning
triangles, the studies provided insufficient data to demonstrate an
equivalency in driver responses. For example, as the Waymo study
acknowledged, unfamiliarity of the warning beacons may have contributed
to motorist behavior. Waymo study at 20. And while Applicants rely on
the study to demonstrate that warning beacons were ``easier to see'' or
``preferred'' by drivers, the study also showed that for respondents
who answered which device was better for signaling a stopped truck, 8
out of 10 selected warning triangles over beacons. Waymo study at 55.
Aurora's study notes that some motorist responses may have occurred
wholly outside the sensor range (Aurora study at 8)--an indication that
the study may not have captured enough data points to compare whether
drivers responded earlier to one warning device over the other.
The studies also failed to fully support whether motorist behavior
was causally based on seeing warning beacons rather than a stopped CMV.
One distinction between FMCSR-specified warning triangles and the
proposed beacons is that warning triangles are placed at the rear of a
stopped CMV (in addition to the front), while the proposed beacons are
located only at the front of the cab--raising the possibility that
drivers see the rear of a stopped CMV before they see the beacons.
While Applicants contend that the beacons are visible from behind the
vehicle, the evidence was inconclusive. For example, the studies failed
to demonstrate through photos or videos the point at which a beacon or
warning triangle was visible or recognized by a motorist, or the point
at which a stopped CMV came into view. Changes in driver behavior were
observed in 11 out of 12 observations with warning beacons (as opposed
to 12 out of 12 for triangles) (Waymo study at 16, Table 3); still, the
evidence does not clearly demonstrate whether drivers reacted to the
CMV or the warning device. For example, neither study establishes a
baseline condition that measured driver reaction to a stopped CMV with
no warning device deployed, a concern raised by MVLSSI in its comments.
Applicants' contentions of equal or higher overall responses to
warning beacons, moreover, fail to acknowledge circumstances where
beacons performed worse than triangles. Aurora's study, for example,
concluded that a warning beacon elicited a 2.75% higher percentage of
responses than warning triangles overall. Aurora study at 24, Table
3.2.1-1. The same table, however, shows a lower percentage responded
overall to warning beacons than to warning triangles in five of eight
(62.5%) scenarios in the table, including daytime tests at left curve
and straight locations, and nighttime tests at crest, right curve, and
straight locations. Id. Although the differences in percentage
responding were small, the overall worse performance for beacons in the
majority of categories does not support Applicants' conclusion that the
proposed beacons performed at a level equal to or better than warning
triangles.
Another limitation of the studies was the limited data. The Aurora
study tested for 8 minutes with warning beacons, replaced the warning
beacons with warning triangles, and then tested the next 8 minutes with
warning triangles. Aurora study at 5-6. This leads to 16 minutes of
total testing time for both warning devices for each roadway geometry
(straight, curve left, curve right, and beyond the crest of a
[[Page 105679]]
hill), with the tests repeated in daytime and nighttime lighting
conditions. The total test time is just over 2 hours for all devices
across all scenarios, with only around 64 minutes of total testing time
for the proposed beacons across 8 scenarios. The Waymo study was
limited to 48 participants. Waymo study at 7. This limited amount of
data does not support an equivalent level of safety, particularly
considering the scope of the exemption requested, which is nationwide
for all current and future autonomous CMV operators during the
exemption period.
Third, the studies do not demonstrate an equivalent level of safety
across sufficiently varied circumstances. Many commenters expressed
particular concern with the visibility of proposed warning beacons when
a CMV is stopped along a curve. For example, TTD stated that the
requirement in 49 CFR 392.22(2)(iv) is in place ``to account for the
simple geometry of curves, hills, and other obstructions that can make
cab lights invisible to oncoming traffic.'' \4\ FMCSA agrees that road
curvatures or elevations may affect visibility of a beacon located in a
fixed position on the cab of a CMV, and that placement of warning
devices at FMCSR-prescribed distances away from the CMV allows for
improved visibility in such situations. FMCSA notes that, due to the
different placement of the devices, the range of the warning provided
by a cab-mounted beacon to an approaching motorist would need to be
longer than the range of the warning provided by an appropriately
placed warning triangle because the source of the warning (the location
at which the warning originates) would be farther from the approaching
motorist for a cab-mounted beacon than for a warning triangle placed
behind a vehicle. FMCSA also expects the radius of the curve could
affect the range of the warning beacon in a similar fashion to how
curves affect the range of adaptive driving beams, as described in 49
CFR 571.108 (FMVSS No. 108) under Table XXII--Adaptive Driving Beam
System Test Matrix.
---------------------------------------------------------------------------
\4\ See https://www.regulations.gov/comment/FMCSA-2023-0071-0036. See also https://www.regulations.gov/comment/FMCSA-2023-0071-0038 (comment from HAAS Alert noting that ``all line-of-sight
alerting solutions are inherently limited by their relative
visibility,'' which can be significantly reduced by weather
conditions, road curvature, hills, and other factors).
---------------------------------------------------------------------------
The studies also gave insufficient information about the nature of
the curves in the studies,\5\ making it difficult to understand whether
the curves were sufficiently representative of the types of curves
(e.g., slight vs. truly blind, sharp curves) the beacon-equipped
autonomous CMVs would encounter across the United States, and therefore
difficult to conclude that there is an equivalent level of safety on
curves. Neither study presented photos or videos demonstrating the
point of view of an approaching motorist entering or exiting the curve.
Moreover, the Waymo study supports that the beacons performed
significantly worse than warning triangles at a curve during daytime
runs. Specifically, at truck exposure location III (shoulder after
curve), 9 out of 12 drivers detected and recognized the truck with the
beacon,\6\ while 12 out of 12 did so with the warning triangles. This
finding is salient to examine, considering that locations with curves
would logically present a challenging scenario for fixed beacons versus
triangles whose placement can be adjusted based on curves.
---------------------------------------------------------------------------
\5\ Appendix C of the Waymo study (Waymo study at 25) includes a
map with a star at the curve location that did not appear from the
map to be a very sharp curve. In Aurora's study, it was also
difficult to determine whether the curves were sharp from the maps.
For example, when examining the left curve shown in Appendix A (at
coordinates 32.3089, -96.593) in Google maps, it appears this
location has a posted speed limit of 75 mph. FMCSA notes a sharp
curve would have a lower posted speed limit, depending on the
curve's radius.
\6\ Table 3 (Waymo study at 16), clearly shows that only 9 out
of 12 drivers detected the scenario with the beacon at location III
for daytime runs. The driver detection data for nighttime runs at
that same location is unclear, however, since Table 17 (Id. at 41)
shows non-response for the button press for 2 beacons and 1
triangle, with 2 non-response observations occurring at location III
and 1 occurring at location II, making it difficult to determine
whether the beacons were missed 16.67% of the time (2 out of 12) or
8.33% of the time (1 out of 12) at the curve.
---------------------------------------------------------------------------
Fourth, while Applicants seek an exemption for CMVs operating in a
Level 4 autonomous mode (App. at 9), Applicants did not support that
vehicles operating in an autonomous mode will stop and engage flashers
and beacons similar to the CMVs in the studies. For example, in the
Waymo study, it appears that CMVs began in a stopped position, while in
the Aurora study, a human driver apparently stopped the vehicle, and in
both studies, a human apparently activated both the hazard warning
flashers and the beacon devices. Waymo study at 7, 9; Aurora study at
5-6; App. at 11 n.21. Thus, neither study demonstrated that CMVs
operating in Level 4 autonomous mode will effectively engage hazard
lamps and/or proposed beacons, or the timing with which they would do
so, and how that timing compares to timing required under the FMCSRs.
And while Applicants contend that they will be able to address
electrical failure, Applicants did not provide evidence regarding how
they will do so. Applicants' request to exempt a class of unspecified
carriers using unspecified equipment on unspecified vehicles only
further undermines the claimed likely equivalent level of safety.
Applicants and some commenters noted potential safety benefits to
human drivers. Indeed, in their ``safety impact'' analysis, Applicants
state that the exemption would increase safety ``most notably by
eliminating the need for a human to enter the roadway to place
traditional warning devices'' and that while the application applies to
CMVs without a human driver, the safety rationale could apply to
conventional CMVs. App. at 9. The exemption request, however, is not
directed at such circumstances--Applicants request that the exemption
be permitted where there is no human driver or where drivers are used
only for testing. App. at 4. Indeed, Applicants contend that no human
drivers will be affected. App. at 9. While FMCSA agrees that there is
promise for improving safety for CMV drivers if they do not need to
leave a vehicle to place warning devices, neither Applicants nor the
public comments presented data to allow FMCSA to find an equivalent or
higher level of safety when using the proposed beacons for carriers
operating CMVs with human drivers, and Applicants did not request an
exemption to cover such operations.
Fifth, there are other limitations of the application to support
the scope sought. For example, the studies do not address visibility of
proposed beacons for vehicle orientations with varying trailer heights,
such as (for example) a truck tractor pulling a flatbed trailer versus
a van type trailer that is taller than the truck tractor. As certain
commenters observed, Applicants did not address how proposed beacons
might perform in scenarios such as a CMV on its side. Applicants
likewise seek a nationwide exemption based only on limited data from a
test track in Virginia and a portion of Interstate 45 in Texas between
Exits 258 and 249, and without establishing how warning beacons might
perform in different geographical or weather conditions across the
country.
Sixth, should Applicants (or future applicants) provide sufficient
details about proposed alternative warning devices, FMCSA notes that
industry-wide exemptions are not the norm and FMCSA grants them only on
a very limited basis. Applicants present little data on the effect such
a broad exemption would have on overall safety, particularly
considering the unknown group of autonomous CMV operators at
[[Page 105680]]
issue, the unknown fleet sizes, and potential differences among them in
implementation and operations.\7\ FMCSA also is concerned that
operators will assume that their own similar but varying versions of
Cab-Mounted Warning Beacons are a ``configuration of similar
effectiveness'' (App. at 10) and therefore are exempted, without such
operators applying for and supporting the safety of an exemption. Thus,
at this stage, the record before the Agency does not show that
Applicants' petition for an industry-wide exemption adequately
demonstrates the required threshold, of likely to achieve an equivalent
level of safety.
---------------------------------------------------------------------------
\7\ Applicants do not provide the names of motor carriers that
would be responsible under the exemptions or the estimates--even for
their own operations--of the number of drivers and CMVs that would
be operated under the exemption. See 49 CFR 381.310(b) and (c). Such
information would assist the Agency in assessing the equivalent
level of safety for operations with the proposed alternative
devices.
---------------------------------------------------------------------------
VI. Conclusion
For the reasons given above, the Agency denies the application for
an exemption.
FMCSA notes that the Agency may accept a resubmission of an
exemption application that has been denied, provided that the applicant
can reasonably address the reason for the denial. 49 U.S.C.
31315(b)(3); 49 CFR 381.317.
Vincent G. White,
Deputy Administrator.
[FR Doc. 2024-30860 Filed 12-26-24; 8:45 am]
BILLING CODE 4910-EX-P