[Senate Hearing 115-313]
[From the U.S. Government Publishing Office]
S. Hrg. 115-313
INNOVATION AND AMERICA'S INFRASTRUCTURE: EXAMINING THE EFFECTS OF
EMERGING AUTONOMOUS TECHNOLOGIES ON AMERICA'S ROADS AND BRIDGES
=======================================================================
HEARING
before the
COMMITTEE ON
ENVIRONMENT AND PUBLIC WORKS
UNITED STATES SENATE
ONE HUNDRED FIFTEENTH CONGRESS
SECOND SESSION
__________
JUNE 13, 2018
__________
Printed for the use of the Committee on Environment and Public Works
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://www.govinfo.gov
U.S. GOVERNMENT PUBLISHING OFFICE
31-205 PDF WASHINGTON : 2018
COMMITTEE ON ENVIRONMENT AND PUBLIC WORKS
ONE HUNDRED FIFTEENTH CONGRESS
SECOND SESSION
JOHN BARRASSO, Wyoming, Chairman
JAMES M. INHOFE, Oklahoma THOMAS R. CARPER, Delaware,
SHELLEY MOORE CAPITO, West Virginia Ranking Member
JOHN BOOZMAN, Arkansas BENJAMIN L. CARDIN, Maryland
ROGER WICKER, Mississippi BERNARD SANDERS, Vermont
DEB FISCHER, Nebraska SHELDON WHITEHOUSE, Rhode Island
JERRY MORAN, Kansas JEFF MERKLEY, Oregon
MIKE ROUNDS, South Dakota KIRSTEN GILLIBRAND, New York
JONI ERNST, Iowa CORY A. BOOKER, New Jersey
DAN SULLIVAN, Alaska EDWARD J. MARKEY, Massachusetts
RICHARD SHELBY, Alabama TAMMY DUCKWORTH, Illinois
CHRIS VAN HOLLEN, Maryland
Richard M. Russell, Majority Staff Director
Mary Frances Repko, Minority Staff Director
C O N T E N T S
----------
Page
JUNE 13, 2018
OPENING STATEMENTS
Barrasso, Hon. John, U.S. Senator from the State of Wyoming...... 1
Carper, Hon. Thomas R., U.S. Senator from the State of Delaware.. 151
WITNESSES
Panos, William ``Bill,'' Director, Wyoming Department of
Transportation................................................. 3
Prepared statement........................................... 5
Responses to additional questions from:
Senator Barrasso......................................... 18
Senator Carper........................................... 20
Senator Whitehouse....................................... 26
Bhatt, Shailen, President and CEO, Intelligent Transportation
Society of America............................................. 29
Prepared statement........................................... 31
Response to an additional question from Senator Barrasso..... 52
Responses to additional questions from:
Senator Carper........................................... 54
Senator Whitehouse....................................... 62
Doerzaph, Zachary, Director, Center for Advanced Automotive
Research....................................................... 69
Prepared statement........................................... 71
Responses to additional questions from:
Senator Carper........................................... 74
Senator Whitehouse....................................... 79
Trottenberg, Polly, Commissioner, New York City Department of
Transportation................................................. 82
Prepared statement........................................... 84
Responses to additional questions from:
Senator Carper........................................... 98
Senator Whitehouse....................................... 103
Kildare, Shaun, Director of Research, Advocates for Highway and
Auto Safety.................................................... 109
Prepared statement........................................... 111
Responses to additional questions from:
Senator Carper........................................... 138
Senator Whitehouse....................................... 143
ADDITIONAL MATERIAL
Public to U.S. Senate: Pump the Brakes on Driverless Car Bill.
ORC International CARAVAN Public Opinion Poll. Advocates for
Highway and Auto Safety, July 2018............................. 169
Testimony of the American Traffic Safety Services Association,
June 13, 2018.................................................. 174
Comments from the American Trucking Associations, March 1, 2018.. 177
Letter to Senators Barrasso and Carper from the Center for Auto
Safety, June 13, 2018.......................................... 179
Letter to Senators Barrasso and Carper from CRASH, the National
Consumers League, and Consumers for Auto Reliability and
Safety, June 13, 2018.......................................... 184
Support Proposed Amendments to the ``AV START Act,'' from the
United States Conference of Mayors, adopted June 11, 2018...... 187
INNOVATION AND AMERICA'S INFRASTRUCTURE: EXAMINING THE EFFECTS OF
EMERGING AUTONOMOUS TECHNOLOGIES ON AMERICA'S ROADS AND BRIDGES
----------
WEDNESDAY, JUNE 13, 2018
U.S. Senate,
Committee on Environment and Public Works,
Washington, DC.
The Committee met, pursuant to notice, at 10:06 a.m. in
room 406, Dirksen Senate Office Building, Hon. John Barrasso
(Chairman of the Committee) presiding.
Present: Senators Barrasso, Carper, Capito, Wicker,
Fischer, Rounds, Ernst, Sullivan, Cardin, Merkley, Gillibrand,
Booker, Markey, and Van Hollen.
OPENING STATEMENT OF HON. JOHN BARRASSO,
U.S. SENATOR FROM THE STATE OF WYOMING
Senator Barrasso. Well, good morning. I call this hearing
to order.
Today we are going to examine the implications of emerging
technologies of America's roadway infrastructure.
Last month our Committee unanimously passed bipartisan
legislation to approve America's water infrastructure. We are
now working together to pass America's water infrastructure act
on the Senate floor. I believe this bipartisan success on water
infrastructure is going to lead to bipartisan success on
America's surface transportation infrastructure, namely,
legislation to address our roads and our bridges.
We are planning to build infrastructure that will last for
decades. We need to understand the new challenges that those
decades will bring to all of us. The ongoing development and
implementation of autonomous vehicles and connected vehicles
and other innovations have the potential to fundamentally
change the way our Nation's infrastructure works.
Autonomous vehicles will likely require modification to our
roadways and changes to the practices of Federal, State, and
local transportation agencies. It is critical that State and
Federal transportation agencies are prepared and equipped to
tackle the potential opportunities and challenges they present
for our roads. Those agencies will need to develop, install,
and maintain traffic control devices in such a way that they
are understood and obeyed by motorists, as well as autonomous
vehicles.
As autonomous vehicles become more common on the road, they
could influence regional traffic models and forecasts. They
will also add new factors as agencies make long-term planning
decisions. At the same time, new vehicles technologies offer
many potential benefits and could transform the way that we
view surface transportation altogether.
Soon, elderly and disabled Americans, as well as those
without a car of their own, may be able to travel by vehicle
with greater ease and greater independence. Likewise, these
innovations have great potential to reduce crashes and
fatalities, to improve mobility, and to increase the efficiency
of the roadway system. How their benefits are realized will
depend on industry and agencies working together to make sure
that our roads keep pace with the vehicles that they
accommodate.
An excellent example of infrastructure innovation is
happening in my home State of Wyoming. The Wyoming Department
of Transportation is implementing a Connected Vehicle Pilot
program to improve monitoring and reporting of road conditions
on Interstate 80. Projects like these are vital for the future
of our Nation's roadway infrastructure.
So, I am glad that Bill Panos, the Wyoming Director of the
Department of Transportation, is here to tell us more about
that project and other work being done in his department.
I also want to thank all of the other witnesses for
participating in today's hearing. Your expertise and insight
will help us understand potential high tech challenges and
opportunities for our Nation's roadway infrastructure.
Senator Carper is unavoidably delayed; he will be here
shortly, and he will make his full opening statement at that
time.
So, I want to thank all of you for being here.
I am pleased to welcome Bill Panos, who has served as the
Director of the Wyoming Department of Transportation. He is a
graduate of California State University, where he received a
degree in forensic science and technology. Before moving to
Wyoming, he accrued 37 years of experience leading private and
public service organizations. He has also served as the
Director of the Wyoming School Facilities Department.
Joining Bill today in testifying will be Mr. Shailen Bhatt,
who is the President and CEO of the Intelligent Transportation
Society of America; Mr. Zachary Doerzaph, who is the Director
of the Center for Advanced Automotive Research; Polly
Trottenberg, who is the Commissioner for the New York City
Department of Transportation; and Shaun Kildare, the Director
of Research at Advocates for Highway and Auto Safety.
I want to remind the witnesses that your full written
testimony will be made part of the official hearing record
today, so please keep your statements to 5 minutes so that we
will have time for questions. I look forward to hearing the
testimony from each of you.
With that, we will begin with Mr. Panos. Appreciate your
being here.
STATEMENT OF WILLIAM ``BILL'' PANOS,
DIRECTOR, WYOMING DEPARTMENT OF TRANSPORTATION
Mr. Panos. Chairman Barrasso and members of the Committee,
I am Bill Panos, Director of the Wyoming Department of
Transportation. I am here today on behalf of WYDOT and AASHTO.
The Committee's premise in holding this hearing is correct;
connected and automated vehicles have implications for highway
infrastructure. In addressing those infrastructure issues,
safety is a top priority for State DOTs. That includes
attention to striping and signage, as well as to more complex
issues. Collaboration between technology developers, vehicle
manufacturers, and government agencies is important. This
includes working to ensure interoperability of systems so that
deployment of dedicated short range communications, or DSRC,
along the highway system will be effective. Let me elaborate.
These new technologies have the potential to reduce
crashes, save lives, and provide other benefits. However, there
is still uncertainty surrounding these innovative technologies,
including infrastructure related issues. For example, for there
to be highway infrastructure, work zones on the roads are
inevitable. How will an automated driving system, or ADS, get
by work zones? What are the specific signage and striping
needs? What advisories from DSRC enabled infrastructure would
help connected and autonomous vehicles and improve safety?
State DOTs are preparing for deployment of CAVs by, among
other things, starting to plan and deploy relevant technology
as part of the highway infrastructure, and these investment
needs are near term, in part because a connected vehicle need
not be an automated vehicle. Cars driven by people are
increasingly equipped with electronics that can receive data
from DSRC enabled equipment along the roadway. DSRC signals can
help non-automated, as well as automated, vehicles effectively
``see'' in bad weather, provide other information, such as on
traffic congestion.
State and local agencies are active in deploying and
testing these systems today. Approximately 50 U.S. locations
are deploying connected vehicle technologies. This represents
roughly 72,000 vehicles and 65,000 devices installed on the
Nation's infrastructure. WYDOT is an active participant.
To improve safety along the 402 miles of Interstate 80 in
Wyoming, particularly in our tough winters, Wyoming is
implementing a pilot program using DSRC enabled technology to
connect vehicles to infrastructure and to other vehicles.
During Federal fiscal year 2016 more than 1,600 crashes
occurred on I-80 in Wyoming, resulting in 18 fatalities and 271
injuries. In December 2016 there were only 8 days when I-80 in
Wyoming was fully open.
As part of the effort to improve the situation, the pilot
program will test warnings advising travelers of crashes ahead.
It will also advise about weather, speed restrictions, work
zones, and other matters. Testing with WYDOT vehicles occurred
last winter, and we will be testing commercial freight vehicles
this winter.
Let me turn to a few more points on the infrastructure
implications of CAV deployment.
Currently State DOTs are unsure which roadway elements are
critically important to automated driving system, or ADS,
technology. We know that quality signage and striping are
important, but welcome more details. In most cases, striping is
a maintenance, not a capital, activity. Similarly, sign
maintenance is not a capital activity.
All other things being equal, an increase in maintenance
costs would reduce funding available for capital investments in
transportation, so we want to understand the maintenance
implications of CAV deployment. In addition, if ADS equipped
vehicles have sensors that could be adversely impacted by poor
ride quality, that could place upward pressure on already high
needs for investments to maintain and improve pavement ride
quality.
There are additional issues where State DOTs want to better
understand what type of information would help all CAVs.
Deployment of ADS technology in rural environments also
raises issues. For example, DSRC enabled infrastructure on long
rural highways would be costly. This suggests different
solutions for advisories in very rural areas, such as cellular
technology, at least where there is adequate cell phone
service. In any event, we want vehicles to have access to
weather, crash, and other key advisory information. It seems
that we have to put technology into the infrastructure to do
that.
So, when a State DOT talks about interoperability, it is
not to tell a developer how to equip a vehicle. Simply, if a
State is deciding whether to use scarce infrastructure dollars
to deploy DSRC enabled systems, it wants to know that the
investment in DRC can successfully communicate advisory
information to vehicles.
In conclusion, Mr. Chairman, CAVs are arriving and are in
our future. We want the infrastructure to be safer than it has
ever been when they are deployed. States are eager to work hard
toward those ends.
Thanks for the opportunity to appear before the Committee
today.
[The prepared statement of Mr. Panos follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Barrasso. Thanks for your testimony, Mr. Panos.
Appreciate it.
Mr. Bhatt.
STATEMENT OF SHAILEN BHATT, PRESIDENT AND CEO,
INTELLIGENT TRANSPORTATION SOCIETY OF AMERICA
Mr. Bhatt. Chairman Barrasso, Ranking Member Carper, and
members of the Committee, thank you for the opportunity to be
here today. When last I testified in front of this Committee I
served as the Executive Director of the Colorado Department of
Transportation. At that time I discussed how a mix of road
investment and innovation was needed to solve Colorado's 21st
century transportation challenges.
I am now President and CEO of the Intelligent
Transportation Society of America, which brings together public
sector agencies, private sector companies, and researchers
unified by our vision of a better world transformed by
intelligent mobility--one that is safer, greener, and smarter.
I am honored to be on this panel today with three ITS America
members.
Today's hearing takes place at an important time. Just as
infrastructure was critical to the development of our economy
in the 20th century, maintenance of existing roads, bridges,
and other infrastructure, and deployment of intelligent
infrastructure will be critical for our global competitiveness
in this century.
Advances in robotics, artificial intelligence, and wireless
communications have inspired a race to make the next generation
of mobility a reality. We are on the leading edge of a
technology revolution that will define the way people, goods,
and information move in the 21st century. It is a whirlwind of
innovation that will change entire industries, as well as
transform communities large and small, urban and rural.
That transformation will positively affect both the safety
and operation of our transportation system if we do it well. In
2016, 37,461 people died on U.S. roads. That is more than 100
people per day. Pedestrian deaths in that year rose 9 percent.
More than 90 percent of roadway crashes are caused in some way
by human error. That is a sobering statistic, but we have
technologies that will make a difference.
Technologies such as pedestrian automatic emergency braking
will use automation to detect pedestrians and prevent
tragedies. Through dedicated short range communications and
other technologies, we also have the capability to allow
vehicles to communicate with other vehicles, smart
infrastructure, and other users of the system.
Safety has always been our top priority and is the reason
many of us are passionate about this work.
In addition to injuries and fatalities, however, there are
major mobility and environmental implications. Americans
currently lose roughly 40 hours per person per year sitting in
traffic, which costs each driver almost $1,500 per year.
Collectively, this drains $305 billion from our economy and
wastes 3.1 billion gallons of fuel. Here, technology can play a
key role in putting money back into Americans' pockets and
improving the environment in which we live. Freight that is
stuck in traffic costs Americans more.
This is why, under Colorado's RoadX program in 2016, we
sent a truck on the world's first autonomous freight delivery.
One of our members, Peloton, is developing driver assisted
truck platooning technology that will improve the flow of goods
across the country, while reducing fuel consumption of trucks
by 7 percent.
Another way of reducing fuel consumption is through
electrification, which is an important part of the future of
intelligent mobility. Right now, by reducing crashes, we can
greatly reduce traffic congestion in this country. More than 50
percent of congestion is caused by non-recurring incidents.
Minor fender benders result in hours of frustration and
inefficiencies in our system. The deployment of connected
vehicle technologies will improve traffic flow across the
country.
Connected and automated vehicle technologies and smart
infrastructure have the potential to give us back our most
precious resource--time. By applying intelligent transportation
technologies toward existing infrastructure, we can maximize
the efficiency of our system.
Twenty-six States and 45 cities are currently deploying V2I
communications that use DSRC. Some of these applications will
include bridge and pavement monitoring, curb speed warning,
reduce speed areas such as in construction zone, and spot
weather warnings, all of which will reduce crashes by providing
vehicles and drivers with the most accurate and up to date
information.
The most important connection between vehicles and
infrastructure has historically been the tire. Today there are
many ways for vehicles to interface with infrastructure and
with other vehicles. We need to prepare for a future that
involves a mixed fleet of intelligent and unconnected vehicles.
The best way to do this is to maintain our infrastructure in a
state of good repair, specifically as that relates to pavement
markings and signage.
However, we need to understand signs that work well for
human eyes may need to be adapted for machine reading. We also
need to understand how cities and States will take these waves
of big data that vehicles are producing and turn it into
actionable information.
Before I close, I urge Congress and the Administration to
identify long-term sustainable funding for the Highway Trust
Fund to maintain our infrastructure for all the reasons I have
just outlined. In my nearly 10 years as a State DOT leader, I
have always said that Departments of Transportation exist to
save lives and make people's lives better. I firmly believe
that advances in vehicle technology and in smart infrastructure
are the best tools in our toolbox to achieve those goals.
Thank you again for this opportunity, and I am happy to
answer your questions.
[The prepared statement of Mr. Bhatt follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Barrasso. Thank you very much, Mr. Bhatt. We
appreciate your being here.
Mr. Doerzaph.
STATEMENT OF ZACHARY DOERZAPH,
DIRECTOR, CENTER FOR ADVANCED AUTOMOTIVE RESEARCH
Mr. Doerzaph. Chairman Barrasso and members of the Senate
Committee on Environment and Public Works, at BTTI we conduct
research for many public and private organizations on automated
systems that are infrastructure and the people who own,
operate, and ride within them. It is an honor to be here to
discuss this very important topic with you. I am quite
passionate about it myself.
Automation may indeed substantially one day have a positive
impact on transportation safety and efficiency. However,
exceeding the capability of the human driver is extraordinarily
difficult and is currently underestimated by many. To achieve
the same safety benefit as the best 10 percent of drivers,
automated vehicles, for all practical purposes, can never
virtually crash.
Large scale deployment of automated vehicles will take
decades to achieve, and there will be a significant percentage
of manually driven vehicles for the foreseeable future.
Automation remains costly, does not equally benefit all users,
and does not operate ubiquitously across all environments.
Fortunately, though, automated vehicles are very much being
designed to operate on roadways that were created for human
drivers. As with humans, the reliability of those automated
systems depends on things like roadway design quality, lane
markings, signs, and other traffic control devices.
At the same time, though, specific infrastructure elements
do create particular challenges unique to automated vehicles.
These edge and corner cases, as we call them, pertain to things
like work zones and emergency situations, adverse weather, and
anywhere that humans can exchange a simple nod, glance, or hand
wave in order to communicate with another road user, which is
where connected vehicle technologies come in. These
technologies, which allow vehicles to communicate with other
vehicles, as well as the infrastructure and other road users,
provide an additional mechanism for improving the perception,
recognition, path planning processes for automation, resulting
in safer and more efficient systems overall.
Connectivity also enables this proactive conversation to
take place between vehicles and vehicles in infrastructure,
much like humans do today.
So, in conclusion, I believe there are measured actions
that should be taken by all stakeholders based on careful
planning to exercise that safety is maintained throughout the
evolution and deployment of automated vehicles, and I recommend
doing so through the following approach.
Support partial automation today. These are systems which
are compatible with the infrastructure, and when appropriately
designed, these limited automations, such as emergency
automated braking, lane keep assist, and others, improve safety
and reduce congestion by leveraging the strengths of both the
human and the machine.
The Federal Government can play a role in providing a clear
pathway to increasing levels of automation with appropriate
operational domains based on demonstrated success.
I also believe we need to incentivize precompetitive
collaboration between individual companies, as well as between
those companies and the road operators, to collaboratively
overcome the technology and policy hurdles.
We also need to facilitate the mechanisms for automated
vehicles to report road deficiencies back to the operators so
that we have a closed loop cycle of improving those facilities.
We need to provide the resources and guidance required to
improve our physical and digital infrastructure through applied
research and deployment support.
Connected technologies require a robust, nationally
interoperable back end data system, precise vehicle
localization, and accurate infrastructure information across
city, State, and local borders. It is imperative that security
mechanisms which establish digital trust and identify and
remediate threats are in place.
We also need to facilitate a broad dialogue and correlation
to define the appropriate oversight role for the Federal and
State public agencies to develop mechanisms for monitoring and
updating such oversight in order to balance innovation with
public safety based on objective measures.
And finally, I would be remiss if I did not mention that I
believe we need to maintain this pace of innovation by
facilitating next generation transportation work force through
technology focused multidisciplinary education and by
supporting a variety of programs for students at all levels.
[The prepared statement of Mr. Doerzaph follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Barrasso. Well, thanks so much for your testimony.
It will be interesting about this next generation work force. I
look forward to getting into that with the questioning.
Commissioner Trottenberg, thank you very much for joining
us. Please proceed.
STATEMENT OF POLLY TROTTENBERG, COMMISSIONER,
NEW YORK CITY DEPARTMENT OF TRANSPORTATION
Ms. Trottenberg. Thank you.
Good morning, Chairman Barrasso and members of the
Committee. On behalf of Mayor Bill de Blasio, I thank you for
inviting me here today to share New York City's perspective on
the deployment of highly automated vehicles in major urban
areas.
New York, like our sister cities, shares a common interest
in ensuring HAV technology is deployed in a way that enhances
urban mobility, safety, and environmental sustainability. We
are grateful to have this opportunity to discuss areas of
concern and see where there are areas of partnership.
As the Nation's largest and densest city, with a population
of 8.6 million and growing, New York City is responsible for
the operation and maintenance of a highly complex surface
transportation network, including 6,000 miles of heavily
traveled urban roadways, 12,000 miles of sidewalks, over 13,000
signals, and nearly 800 bridges and tunnels, many of them well
over 100 years old; and we work closely to operate an
integrated and efficient transportation system with the MTA,
which runs our subway and bus system with over 8 million
transit trips per day.
I hope my perspective as a city DOT commissioner and former
undersecretary at USDOT will prove useful as the Senate
deliberates on the opportunities and challenges we face with
HAVs and the implication of this technology for our roads and
bridges.
Comments from industry suggest that cities need to get
their infrastructure ready for the deployment of HAVs. I would
argue just the opposite.
New HAV technology should, instead, be prepared to operate
safely and effectively in complex urban environments; in snow,
with traffic control officers managing an intersection when
signals are out and judgment is needed or where pavement
conditions or lane markings are deteriorated. It is simply not
realistic or feasible to expect cities to overhaul their
existing roadway infrastructure to accommodate a still somewhat
unproven technology.
New York is proud to be the first U.S. city to embrace the
concept of Vision Zero, which declares that all traffic deaths
and serious injuries are preventable. In the last 4 years New
York City has achieved remarkable results. Traffic deaths have
declined by 27 percent, and pedestrian fatalities have declined
by 44 percent. And New York City is bucking the national trend,
where, tragically, roadway fatalities have increased by 15
percent.
We have achieved these results through a strong partnership
between New York City DOT and the NYPD, as well as robust
investment in a comprehensive data driven roadway safety
program relying on engineering, education, and enforcement.
HAVs hold the promise of dramatically reducing traffic deaths
and serious injuries, but to achieve this promise, the U.S.
should first establish rigorous foundational safety standards
across the board.
For example, the European Commission recently proposed
that, starting in 2020, all new vehicles sold in Europe must be
equipped with intelligent speed assistance, pedestrian and
cycling recognition systems, and automated braking.
In the U.S., we should be advancing similar standards and
NHTSA should build on and integrate the best elements of the
approaches being used by California and Boston, adopting an
approach of incremental testing for HAVs with data sharing
requirements.
Cities are where the bulk of Americans live and travel now,
and for many, including New York, congestion has become a
critical challenge. HAVs hold the promise of reducing
congestion or profoundly exacerbating it. Unfortunately, to
date, the Federal Government has not meaningfully involved
cities in its development of HAV policy. Moving forward, we
request that USDOT and NHTSA engage with cities more directly.
We will all be most successful as partners, cities, States,
USDOT, and the industry.
It is critical to establish protocols that allow HAV safety
data to be shared with States and cities. Some data, when
appropriate, should also be shared publicly. Providing for a
robust level of transparency for HAV safety data will be
essential to create a safety culture akin to that of the U.S.
aviation sector.
Throughout U.S. history traffic safety has always been a
shared responsibility of the Federal, State, and local
governments. This authority must be unambiguously preserved in
the AV START Act, and HAVs must be programmed to follow all
State and local laws, including speed limits.
The legislation also does not require standards based
verifiable testing of HAV systems. We urge the Senate to revise
the legislation before it advances.
Of all the disruptive challenges HAVs are poised to bring,
none may be as consequential as the impact on our Nation's work
force. According to recent census data, more than 4.4 million
Americans, including approximately 250,000 New Yorkers, make
their living driving.
All of our communities, urban and rural alike, will need to
confront the potential human toll that this disruptive
technology could take. The Federal Government needs to help
ensure that innovation and opportunity for some does not mean
we are leaving others without a livelihood.
As Congress considers its approach to fast developing HAV
technology, I urge you to enlist cities as partners. New York
City stands ready to work with you.
I thank the Committee and look forward to your questions.
[The prepared statement of Ms. Trottenberg follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Barrasso. Thank you so much for your testimony,
Commissioner Trottenberg.
Mr. Kildare.
STATEMENT OF SHAUN KILDARE, DIRECTOR OF RESEARCH,
ADVOCATES FOR HIGHWAY AND AUTO SAFETY
Mr. Kildare. Good morning, Chairman Barrasso, Ranking
Member Carper, and members of the Committee. I am Shaun
Kildare, Director of Research for Advocates for Highway and
Auto Safety, a coalition of public health, safety, consumer
organizations, and property/casualty insurance companies
dedicated to advancing safer vehicles, safer drivers, and safer
roads. Thank you for the opportunity to testify.
Each year motor vehicle crashes kill tens of thousands of
people and injure millions more, at a cost of over $800
billion. Moreover, at a time when deaths on our Nation's roads
are remaining unacceptably high, America's infrastructure is in
deep disrepair. The American Society of Civil Engineering gives
our roads a grade of D and reports that 1 in 11 of our nearly
615,000 bridges are structurally deficient.
In addition to committing resources to fix our roads and
bridges, substantial investments will be required to ensure
that autonomous vehicles, or AVs, can operate safely. Federal
leadership is needed to achieve infrastructure improvements and
to create a regulatory framework for vehicle design and
performance.
Advocates has always been a strong champion for vehicle
safety technology and infrastructure improvements. So, too, do
we believe that in the long-term AVs may once and for all bring
about meaningful reductions in the death and injury toll on our
Nation's roads. The industry touts that AVs will improve
safety, reduce congestion, and benefit the environment.
Instead, what we have seen is a number of crashes resulting in
at least three deaths.
At the time of the fatal pedestrian crash in Arizona, Uber
had logged approximately 3 million autonomous miles. While that
may sound like a large number, consider that every year
Americans drive 3 trillion miles. During those 3 trillion
miles, on average, a person was killed in a traffic collision
every 85 million miles in 2016. In comparison, Uber's AV
fatality rate is 28 times that of human drivers. This
highlights just how little proof there is that these systems
are safe or certainly not safer than human drivers presently.
Statements regarding reductions in congestion and
improvements in the environment are similarly dubious. There is
a wide variation regarding estimates of changes in vehicle
miles traveled. Often absent from these urban planning utopias
is the reality that AVs may bring the possibility of
hypercommuters living several hours outside of cities. Also
frequently missing is the likelihood of empty vehicles circling
aimlessly between rides.
What we do know is that after the March 23rd Tesla crash in
California, U.S. Highway 101 was closed for nearly 6 hours. The
battery was on fire, emitting dangerous chemicals, and the
battery reignited 5 days afterwards. That paints a very grim
picture.
The fact is that rushing to deploy AVs provides no guaranty
of the benefits claimed and may come with significant costs.
Despite these uncertainties, the USDOT has chosen to take a
hands off approach by issuing only voluntary guidelines.
Therefore, Advocates has put forth several reasonable proposals
which are outlined in my written testimony.
Regarding infrastructure improvements, we offer the
following three recommendations:
First, for road design. The lynchpin for much of the
guidance in numerous infrastructure manuals is a human behind
the world. From sight distances for signs, to lettering, to the
curvature and super elevations of roads, infrastructure design
criteria has been developed to enable a safe operation of
vehicles by human drivers. AVs may require that these basic
premises be modified in order to serve a dual purpose for both
human and computer driven vehicles.
Second, roadway deterioration. We have all experienced road
signs or markings that have been damaged intentionally or
altered or blocked by objects. Research shows that a stop sign
can easily be manipulated with a few pieces of tape, which is
then read by the AV to be a 45 mile an hour speed limit sign,
resulting in the AV speeding up instead of stopping. This is
why Advocates is recommending AVs be subject to a vision test.
When a person goes to a DMV to get a license, he or she has to
take a vision test. With the AV now being the entity that is
seeing the road, AVs should have to demonstrate that they can
see and respond to the roadway challenges.
Third, connected vehicles. These technologies allow a
vehicle to send and receive communications with other vehicles,
known as V2V, or the infrastructure, known as V2I. They will
likely help fill gaps in AV performance. For example, V2V
communication can provide safety applications for Forward
Collision Warnings which alert drivers to stopped or slowed
vehicles ahead. Advocates has filed comments in support of
mandating V2V; however, the rule continues to languish at
USDOT.
To conclude, this hearing is very well timed, considering
that last week the NTSB released their preliminary report on
the fatal Tesla crash in California this March. It appears
likely that infrastructure components may have been a factor in
that crash. As such, we urge the Senate to allow time for the
NTSB to finish its pending investigations on AV systems. There
is a great deal to be learned from our Nation's leading
investigators, and there is no reason to rush through
legislation, especially by tacking it on to an unrelated bill.
I look forward to your questions.
[The prepared statement of Mr. Kildare follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Senator Barrasso. Thanks so much to all of you for being
here. I look forward to starting the questions in a second.
I do, first, ask unanimous consent to enter into the record
a letter from the Alliance of Automobile Manufacturers, which
explains the benefit of a development in eventual deployment of
autonomous vehicles.
Without objection, that will be submitted.
[The referenced information was not received at time of
print.]
Senator Barrasso. Mr. Panos, as you discussed in your
testimony, the Wyoming Department of Transportation is
participating in the U.S. Department of Transportation study to
assess the effectiveness of connected vehicle technologies
really under real world conditions, because that is what this
is all about. How has this effort helped WYDOT anticipate the
changes that are going to be needed so that we can get the
greatest benefit from connected vehicles and AV implementation?
Mr. Panos. For us, the participation with USDOT and with
our partners on the project is really a benefit to us because
our focus is on safety, and the idea that we have the ability
to deploy technology in a variety of different environments. As
you know, the pilot study is not just about Wyoming, but it is
also inclusive of New York City, the New York area, and also
Florida; and working in these kinds of deployments and these
kinds of environments is extremely beneficial to studying the
effectiveness of the technology and the various conditions
within which it is going to have to operate, so that is very,
very important.
As I described both in my written and oral testimony, we
have already started to deploy not only the technology within
our own fleet, but with freight vehicles as we move forward, so
I think that the ongoing funding by USDOT, ongoing funding by
the Federal Highway Administration toward going to Phase 2,
Phase 3, toward further deployment of these technologies is
critical for us to improve safety.
In our particular case, we have a great deal of freight
that moves through the southern part of our State, and our
focus is on not only maintaining a safe environment for that
freight to move, fewer fatalities, fewer crashes, et cetera,
but fewer closures of that road system so that freight can move
from the western side of the country to the eastern side of the
country and vice versa.
So, for us, these connected vehicle programs are an initial
step toward connected braking and then automated vehicle
programs which are necessary to save lives.
Senator Barrasso. In addition to the study that you are
actually involved in, are there any further efforts or follow
up activities that we ought to be thinking about or pursuing to
get more information?
Mr. Panos. Yes. I think that FHWA recently announced a
series of national meetings to discuss automated vehicles and
infrastructure. Congress can encourage the FHWA to move forward
promptly on those meetings.
AASHTO has been working hard to bring people together for
years and has updated its coalition of public and private
sector entities to form the Cooperative Automated
Transportation Coalition, or CAT Coalition. Those types of
efforts should be encouraged. They will help bridge some of the
current gaps in the collaboration among all of the various
stakeholders.
Senator Barrasso. Mr. Bhatt, everybody has testified to
this. In order to successfully prepare our roadways for
autonomous vehicles, State and Federal agencies are going to
need to engage in proactive preparation, rather than just
simply reacting. What do you perceive as kind of the primary
regulatory challenges or opportunities that are going to
determine the infrastructure and agency readiness as the use is
adopted and expanded for autonomous vehicles?
Mr. Bhatt. Thank you, Chairman, for that question. It is a
great question to ask right now because I think that what you
hear in the testimony across this panel is there is both great
opportunity and great challenges that we need to understand,
and I think it is a great time for this Committee, Congress as
a whole, and the Administration to be looking at these issues.
From a regulatory standpoint, I think what we need right
now is a Federal framework so that we don't have 50 different
States and then along with other cities and jurisdictions sort
of developing their own standards when it comes to the
operation of these vehicles. I think that that is one of the
things that we are looking for.
I also think it would be important for us to recognize that
this is not just about the United States; that there is a
global competition around both vehicle manufacturing and
communication technologies where other countries are investing
and actually cite the fact that there is no consistency across
the United States as a competitive advantage for companies to
come and manufacture and test in those places.
So, I think that what we want to make sure is that we
preserve the historic, that we have had for several decades,
relationship of manufacturers make the vehicles and then local
jurisdictions decide how they are operated, but we need a
Federal framework to make that happen.
Senator Barrasso. Thank you very much.
I have a lot more questions, but Senator Rounds, let me go
to you first.
Senator Rounds. Thank you, Mr. Chairman.
As I listened to the discussion here, I noted that we have
folks from New York, Wyoming, Colorado, and when I think about
the differences between the way each of your different
transportation commissions would respond to what your needs
are, and I am not certain that, at the Federal level, we can
determine for all of them what they need to do and the priority
in which they need to do it.
At the same time, I guess I am going to ask the loaded
question, and that is with regard to the Highway Trust Fund. Is
there anybody that thinks that we should have a subsection
dedicated to the advancements of these particular capabilities
within the Highway Trust Fund? Remembering that you are
probably taking money away from bridges and road repair, road
construction, and so forth. Where do you put this at?
Yes, sir.
Mr. Bhatt. I am happy to take the loaded question, and I
appreciate your perspective that it is a very different
country, whether it is New York City or Wyoming, Colorado, or
Washington, DC. And I appreciate the idea that we recognize
that.
I think that from the ITS America perspective, which is a
coalition of State DOTs, cities, private sector companies, and
research institutions, I think, with regard to the Highway
Trust Fund, recognizing the challenges that are there, it is
hard to say, well, how do you fund important technology
investments, and at the same time, many States are struggling
with this idea of we can't maintain the existing roads and
bridges that we have.
So, I would say that our association would say, along with,
I think, a broad bipartisan chorus, that there does need to be
more funding available for transportation, but I think that
what I would focus on is the ability of technology to leverage
existing investments.
We all talked about pavement markings. Striping is a big
deal obviously for the performance of autonomous vehicles. In
Colorado we had the Eisenhower-Johnson Tunnel, the highest
interstate tunnel in the country, and snow can fall there
virtually any time of the year. Those plows go out, and a
stripe is just really paint on asphalt. You plow it enough
times, it comes off. What they have deployed in Colorado now is
almost like a recess within an LED light inside it, so when
that plow goes over it, it doesn't scrape away paint, it
doesn't scrape away anything, and it is a much safer piece of
infrastructure.
And the next step of technology is--I have talked to
companies that want to put a little RFID chip in there so that
that RFID chip can broadcast to a plow that I am here, so if
you can't see the roadway, that you at least understand where
the lane is.
So, whether it is signals through the SPaT Challenge that
can be upgraded, I think that there are many ways that we can
see that technology can leverage the existing investment. We
currently use 2,000 vehicles per lane per hour for what a lane
of interstate will move. Once those vehicles are talking to
each other, you can cut down on stopping distance. I have seen
estimates where that can go to 4,000 or even higher. So, I
think we want to begin the conversation around how do we use
investment to leverage both the existing infrastructure and the
technology investment.
Senator Rounds. Interesting. I am just curious. I look at
the different weather conditions that are out there right now,
and my first thought was is perhaps the first step would be to
provide more information to both drivers and driverless
vehicles, and a lot of that data could be used by both, as
opposed to simply focusing on the autonomous vehicles.
You mentioned snow. I am thinking of the percentage of the
time in which our rural highways and places like in South
Dakota, where I am from, where we are not going to have the
visibility of stripes, we are not necessarily going to have the
visibility of a yellow line down the road, and those are the
times in which these accidents, in many cases, occur, adverse
weather conditions.
Then I look at New York, and I recognize the expertise of
your taxi crew there to actually work its way down through a
5th Avenue challenge is something else, and it is something
that very few of us in South Dakota get an opportunity to
experience, nor want to experience.
So, it would appear to me that what we can do to provide
additional data is something that would help everyone,
including, and I suspect as Mr. Kildare had indicated, the
ability here to provide additional information that is actually
beneficial to both the driver and a non-driven vehicle. I am
wondering if the focus perhaps shouldn't be more, to start out
with, making sure that we have the ability to deliver ongoing
data, such as with a stop sign that could be modified. But if
you have a GPS that indicates clearly that there is a stop sign
expected at that location, to at least show if there is a
difference between what is perceived with the visible
perception versus what the GPS should be there would throw out
the proverbial red flag.
And then I will shut up.
Would you agree with that, Mr. Kildare?
Ms. Trottenberg. I would certainly jump in. From the New
York perspective, that data is key. And as you heard from my
colleague in Wyoming, New York is also participating in the
Connected Vehicle Pilot, and the information that we are
gathering is going to be very, very crucial.
That is why we think it is important, as automated vehicle
testing goes forward, that jurisdictions, cities and States,
that we work out some kind of data sharing arrangement, because
that data will be crucial to us in making some of these
infrastructure decisions and seeing where there are safety
challenges, places we need to improve the efficiency of our
roadways.
You know, you are talking about, in a rural context, snow
covering up your markings, in an urban context, particularly in
a city like New York, we have a ton of infrastructure
underneath our streets. On any given day in New York City there
are a thousand holes being cut into our streets, so the notion
that our striping will always be pristine and visible, I think
it is not likely.
Senator Rounds. Or parked on.
Ms. Trottenberg. Exactly.
Senator Rounds. Thank you.
Thank you, Mr. Chairman.
Senator Barrasso. Well, thank you very much, Senator
Rounds.
Senator Carper.
OPENING STATEMENT OF HON. THOMAS R. CARPER,
U.S. SENATOR FROM THE STATE OF DELAWARE
Senator Carper. Thanks much, Mr. Chairman.
Welcome to all of our witnesses, especially Shailen Bhatt,
who used to be our secretary.
Shailen, it is great to see you, and thanks for bring your
friends with you here today.
I apologize for being late. We all serve on a number of
committees, and sometimes we just bounce back and forth easily,
but we are doing a markup, a business meeting in the Homeland
Security Committee today with a big agenda, and we had exactly
a quorum; and if I had left, we would have lost the quorum and
not been able to proceed, so I appreciate your indulgence here.
I have a short statement I would like to share with all of
you.
This important hearing, we are grateful for it, Mr.
Chairman. I thank our staffs for the work that has gone into
preparing for it.
Harry Truman used to say the only thing new in the world is
the history forgot or never learned. Think about that. The only
thing new in the world is the history we forgot or never
learned.
Today we are here to discuss autonomous vehicles, which do
seem like a very new thing, but over 100 years ago, before the
advent of driverless cars, the new thing was the horseless
carriage, or as we now call them, cars.
Reflecting on the early history of cars and automobiles,
trucks, reflecting on that early history may provide some
lessons as we plan for the deployment of today's new
technology. I doubt there is anyone here who would deny the
tremendous benefits of the development of cars, trucks, vans
has had on our society and our economy.
These vehicles connect urban and rural communities; they
provide new access to schools, to jobs and hospitals. Cars and
trucks have allowed us to travel farther, and to ship and
receive goods more quickly and more cheaply.
It is also fair to acknowledge that these mobility
improvements have come with some costs. We had to make space in
urban areas, often at the expense of existing housing, for
better infrastructure in the form of roads and highways. Think
I-95 going through Wilmington, for example. Motor vehicles
quickly became a major source of emissions and smog,
contributing to the threat of climate change, as well as public
health crises, such as asthma.
The advent of early automobiles also posed a major safety
challenge, and infrastructure was required to ensure that they
operated more safely. I am told in the first decade of the 20th
century there were no stop signs; there were no warning signs;
there were no traffic lights; there were no lane lines; there
were no streetlights; there were no brake lights; there were no
driver's licenses; no seatbelts; no posted speed limits. None
of these traffic controls and safety devices had been
developed, so communities were unprepared for this new
technology, new vehicles that came along about 100 years ago.
As a result, passengers were at risk. I am told, in 1910--
more than 100 years ago--there were 45 deaths for every 100
million miles traveled. Forty-five deaths for every 100 million
miles traveled. We have been able to bring that number down to
a number just about 1 death per 100 million miles traveled
today thanks to a variety of things, but thanks in part to
Federal motor vehicle safety standards and investments in safer
roadways and safer vehicles.
I think most of us would agree that the number even one, if
you happen to be that one, that is one too many. The number is
still too high, but in 2016 I am told almost 40,000 people were
killed in crashes on our roads across this country.
I hope that autonomous vehicles will help us reduce
fatality rates even further. Over 90 percent of traffic
fatalities are the result of driver related errors, including
from drunk, drowsy, or distracted driving. They may be reduced
with driverless cars. That would be a good thing.
Indeed, there is no doubt that this emerging technology has
the potential to enhance safety, to enhance mobility, reduce
congestion, and improve excess. But realizing these benefits
will depend on two important things: No. 1, how the technology
is deployed, and also how much we invest to ensure that our
streets are ready for this new technology.
One of our major goals should be to avoid the mistakes of
the last century, when cars were deployed into our communities
without any of the infrastructure standards, the traffic
devices, the safety protocols, the environmental protections
that we only later realized were essential.
Our hearing today will help us better understand how we can
prepare for this transformative technology so that we can
realize its many potential benefits, but also minimize the
costs associated with cutting corners in our zeal to see this
exciting technology deployed.
We need to better understand the readiness of our
infrastructure and our traffic controls. For instance, we know
that autonomous vehicles can have difficulty navigating certain
road conditions, such as poor lighting, such as bad weather,
such as work zones. So how do we mitigate with these
challenges?
Connected autonomous vehicles may travel more closely
together, which could reduce congestion. That would be a good
thing. But how will a connected series of heavy trucks affect
the weight limits of highway bridges? That could be a dangerous
thing.
Road designs and traffic signs have been optimized for
human comprehension; however, we look at the very simple
changes that can be made to a stop sign that could interfere
with an autonomous vehicle's ability to accurately understand
that same sign. I think we have a poster. If you look at the
stop sign, with just a couple markings that could be made to
the stop sign, it turns into not a stop sign, but a speed limit
that says you can go 45 miles per hour through this
intersection. That would be scary.
So, we need to ensure that vehicle computers will read
signs like that, particularly when graffiti or other
modifications can fool an autonomous vehicle into thinking that
a stop sign is actually a speed limit sign. We all know, as
much as we might hope that something like that is not going to
happen, we know that it very well could.
We may need to digitally connect our vehicles to our
infrastructure. How much will that cost? How do we ensure that
it is compatible with all autonomous vehicle technologies?
Finally, technology is changing at a rapid pace, we know
that, but State and local agencies must plan now for
transportation investments that won't be made until much
further down the road.
How do we align those timeframes and integrate assumptions
about autonomous vehicles into long range transportation plans?
Those are just a few of the many questions that I believe
we need to be examining closely as we prepare our
infrastructure for more widespread use of autonomous vehicles.
I look forward to your testimony. Apologize again for being
late, but better late than not at all. Now I am here, and I am
in the game.
Thanks so much, Mr. Chairman.
Senator Barrasso. Would you like to proceed with questions,
or should I go to Senator Capito?
Senator Carper. I would go to Senator Capito.
Senator Barrasso. Senator Capito.
Senator Capito. Thank you. Thank you, Mr. Chairman.
And I thank all of you.
Mr. Panos, like West Virginia, Wyoming is a largely rural
State with difficult and mountainous terrain. Whenever I hear
and think about autonomous vehicles, I think about going up
Bridge Road, which is where I live, to my house, and I am like,
I am not getting in an autonomous vehicle and doing those
curves.
We also have difficulty with in and out of our service,
satellite service or Internet service. It is very spotty and
can be unreliable. In the best case unreliable, and in the
worst case, non-existent.
I understand that in Wyoming your testing is usually in
ideal weathers and more flat circumstances, but you are doing
some testing in the mountainous regions, so I am interested in
knowing how that is going and what type of challenges that
presents in terms of autonomous vehicles.
Mr. Panos. Thank you for the question. In Wyoming we are
testing not only in open areas, but also in mountainous areas.
We have a very diverse terrain. We also have very diverse
weather conditions. So, we are testing the infrastructure
equipment not only for its ability to speak with vehicles and
for vehicles to speak with it, if you will, but also to test it
in terms of its nature as it relates to the various geologic
and weather sort of conditions that we have there.
What we have found is, for us, we use DRC-enabled equipment
that helps in our communicating with the vehicles. Our focus is
on connected vehicles at this point, and connected freight
vehicles, specifically.
Senator Capito. So, when you are saying connected, are you
saying connected to the device that is on the infrastructure,
or are you talking about connected in a broadband, wireless
way?
Mr. Panos. Well, to use your examples, both.
Senator Capito. Both.
Mr. Panos. They are connected vehicle to vehicle and then
vehicle to infrastructure. And then the infrastructure is
connected back to a central location where we actually can send
messages to those vehicles----
Senator Capito. So you have to have connectivity.
Mr. Panos [continuing]. And transmit them to others. Excuse
me; I apologize. And DSRC helps us to do that, but I think that
what we believe will be enhanced is not just the type of
technology that we use, because we could use a variety of
technology, but really the type of messages and advisories that
we are pushing. Advisories about crashes, advisories about
weather that are up in front are the most effective that we can
see.
But I must say that this idea of developing a collaboration
amongst all of the stakeholders associated with the type and
use of technology, whether it be connected vehicle or
autonomous vehicle, is the key, and these collaborations, and
having Congress support those collaborations, is essential for
us as we go forward not just as Wyoming, but as all State DOTs.
Senator Capito. I am also serving on the Commerce
Committee. We had a very vigorous debate on AV technology and
cars and trucks, and should trucks be included in the first
sort of strike that we went in terms of trying to figure out
the best regulatory environment to move forward. Trucks were
not made a part of that.
I am wondering if any of you all, in your testing or
exposure, has been working with large truck vehicles and what
you are finding there.
Mr. Panos. I will just mention, if I may, Senator, that
that is the focus of our connected vehicle program, is freight
specifically, and freight connected not only to the
infrastructure and to one another, but freight connected to our
emergency response vehicles so that we can reduce the time that
an emergency responder can arrive to the scene of an accident
or some other type of incident.
Senator Capito. OK.
Mr. Bhatt. Senator, obviously, the freight component is
critical.
Senator Capito. Right.
Mr. Bhatt. One of our members is Peloton that is testing
truck platooning, where you use connection between vehicles to
reduce the space that is needed, reduce the fuel economy. In
places like Wyoming and in Colorado, where you have long
distances, it can provide a lot of benefit, safety benefits.
Another issue in Colorado, over Red Mountain Pass, we
sometimes have truckers coming into the State that have never
been there. There is an issue on I-25. Now they are using their
GPS to get around it; they end up on Red Mountain Pass, and Red
Mountain Pass there are a couple of curves where, if you don't
negotiate them properly, it turns into a pretty significant
situation. We are looking at deploying infrastructure to
vehicle communication so that even if the truck driver isn't
aware that it is a safety issue, the truck can be told to slow
down to no more than 10 or 15 miles an hour for some of these
curves.
Senator Capito. Interesting. Thank you very much.
Senator Barrasso. Thank you, Senator Capito.
Senator Carper.
Senator Carper. Did you go to Ben?
Senator Barrasso. Well, it is your choice. You are first to
go with questions, but Senator Cardin would be next in line.
Senator Cardin. Well, thank you. Thank you, Mr. Chairman,
and thank you, Senator Carper.
This hearing is extremely important, and I thank all of our
witnesses.
I have the opportunity to be the Ranking Democrat on the
Infrastructure Subcommittee with Senator Inhofe, and as we are
looking at the infrastructure authorization for America on
transportation, clearly the technology issues need to be part
of those conversations, so I think this panel is particularly
important. We all support the enhanced use of technology to
make transportation more efficient, to make it more friendly,
to make it safer.
Certain challenges are brought out by this, and you have
already mentioned some. How this comes out of the maintenance
budget I thought was an interesting concept, as compared to
construction. We don't have enough resources right now into the
transportation, so unless we have an adequate funding source,
it seems to me that technology is not going to get the
attention it needs in the planning of infrastructure.
Second, I would point out that technology will help make
transportation more efficient, which is what we want it to be,
more efficient. But as we make it more efficient, the revenues
that are coming in to support the infrastructure gets lower,
gets less. So, once again, it is so important as we look at
developing an infrastructure program for this country, that we
take technology into consideration so that we have adequate
resources in order to take advantage of the benefits of
technology.
I want to talk about a couple specific issues. I was
listening to the messaging boards that you talk about. Very
important. If there is a serious issue, motorists need to be
advised so they can plan alternative routes, they can plan
safety decisionmaking.
But I find that many of these messaging systems are putting
up messages that are not terribly relevant to the driver, but
do cause the drivers to slow down, causing congestion and
sometimes a safety problem in and of itself.
Is there a protocol as to how these messaging boards are
located and whether they should be used indiscriminately or
when it is not involving an important message for the driving
public?
Mr. Panos. Senator, could I respond to that? Thank you for
the question. There is a protocol, actually, that we have
developed with the Federal Highway Administration for the
siting, construction, and operation of DMS signs, digital
messaging signs, throughout our State, and we have internal
groups which actually look at the messaging that is going up,
plus we have a 24/7 online traffic management center that
actually executes the messages to the signs and makes them as
real time as possible. So, for instance, if we were to have a
hail incident that would move through or actually blow through
the State from, let's say, west to east, the message signs
would change and follow the hail incident as it moves across
the interstate highway warning----
Senator Cardin. And that is beneficial. I am for that. But
my concern, I will look at my app as to traffic problems on I-
95 as I am driving, and I will see either yellow or red where
the messaging signs are located, and I know the people are
slowing down in order to read the messaging signs. Fine if it
is important, but if it is not--some of the signs will say have
a happy holiday weekend, drive safely, or something like that,
which I don't think is worthy of creating a particular problem
of a slowdown because people are slowing to read the message
signs.
Mr. Panos. Senator, if I may, again, we have a protocol so
that we generally would not be putting up non-relevant
information on the signs relative to that particular area of
the State. Remember, I-80 in our State, where there are a lot
of these DMSes, is 400 miles long, so we would put different
messages, but also safety messages; and those are the only two
things we are allowed to put on the signs based on the
protocols that at least we use in Wyoming.
Senator Cardin. Appreciate it.
I want to cover one other subject, if I might, and that is
vulnerable populations, bikers, pedestrians. As we develop
these new technologies, what protections are there for what
this Committee has brought forward under the TAP program to
promote pedestrian and bicycle opportunities? If you are now
going to have automatic technologies, how do we protect the
bikers and walkers?
Ms. Trottenberg. I will take a crack at that, Senator,
because I think from the New York City context that is an
extremely important question. For our Federal pilot, the
Connected Vehicles program, that is one of the issues we are
very much looking at; it is not just vehicles talking to
infrastructure, it is how can we make roadway safety better for
pedestrians, for cyclists. That is part of why we are so
interested in having some robust safety protocols as we start
the testing. Europe is looking at requiring certain pedestrian
and cyclist recognition technologies and why we are interested
in having good data sharing, so we can learn what these
vehicles are seeing. For us in New York, pedestrian and cyclist
safety is a huge, huge priority.
Mr. Kildare. Thank you very much for the opportunity to
respond. I think we saw the dangers writ large by what happened
with Uber and the crash that occurred. There was mention
earlier about edge cases. The problems we are seeing with AVs
at this time are not edge cases. A pedestrian crossing a road
at night is not a surprise. A fire truck being stopped on a
road is not a surprise. A tractor trailer making a left turn
across traffic is not a surprise. These aren't edge cases.
So, we need to collect a lot more of that data. We can do
that during testing in controlled ways, but without allowing
the system to put people in danger. You can run these systems
with a human driver actually doing the driving, collect the
data streams, analyze it, and see what they are seeing. If we
see that the machine would have made the decision to drive
toward a bicyclist, we would know that without actually
endangering the bicyclist and letting the machine drive toward
that bicyclist.
The same thing we saw about the infrastructure problem,
what happened with the latest Tesla crash. We would have the
data stream saying, hey, the vehicle would have liked to have
driven into this roadside hardware, but it didn't because the
human driver was always engaged and always doing the driving.
So that data is absolutely crucial to collect and have.
Senator Cardin. Thank you all very much. Appreciate it.
Senator Barrasso. Thank you, Senator Cardin.
Senator Booker.
Senator Booker. Thank you very much, Mr. Chairman.
I have a lot of concerns about just the rising levels of
traffic fatalities on our roads. It is stunning to me that we
have this kind of carnage. 37,461 people were killed on our
Nation's roads in 2016 alone. And we seem to have--maybe
popularly--seem to have this resignation that this is just a
normal, and it shouldn't be. You know, when you see lives
devastated like we have seen in New Jersey very recently,
especially with the horrible crash involving a school bus, I
think we have to start having more of a conviction to prevent
these tragedies from happening.
So emerging sorts of technology is hopeful to me, that
there might be a lot of possibilities. Autonomous vehicles
present, to me, first and foremost, a chance not to ease
congestion; the biggest thing that excites me is the potential
to save lives. There are other collateral benefits, reducing
emissions. If you live between New York and New Jersey, one
side of the river or the other, you understand how traffic is
eroding the quality of life of people.
So, I just want to make sure that we are making the kind of
investments in our infrastructure that we should be making, and
I want to start with Mr. Kildare, which is the greatest name,
man, you have, by the way. If I had that name, I would have
made it to the NFL in football. The name alone would have
gotten me through the combines.
Senator Barrasso. Or he could have been a doctor on
television.
Senator Booker. Exactly. Exactly.
America's roads continue to receive a D from the ASCE,
which noted that 20 percent of the Nation's highways alone had
poor pavement conditions. The Federal Highway Administration
estimates that $142 billion in capital investment would be
needed on an annual basis over the next 20 years to get our
conditions to where they should be.
What is, again, compelling me is that, tragically,
approximately one-third of road fatalities are caused in part
due to deficient infrastructure.
The first question, Mr. Kildare, is whether it is State
DOTs, local governments, transit agencies, the whole list of
sort of multiple crisscrossing responsibilities, the
transportation sector is going to continue to wrestle about how
to deal with the new technologies that we are seeing and
integrate them in an effective way, and to make this, capital
investments are going to be needed in order to accommodate
connected and autonomous vehicles, cars, buses, trucks, and the
like.
So, I just want to know what would you recommend in terms
of the investment that you think Congress needs not only to
accommodate the issues I am talking about, but even just to
compete globally to other nations that seem to be making
infrastructure investments much more rapidly than we are.
Mr. Kildare. I believe that the Federal coordination is
going to be a critical part to this. A lot of our manuals and a
lot of the instruction that we currently have of how we spend
our funds and what our designs are are left to the local level,
and that is important. It is important because we have the
difference between rural and urban and the different
experiences that we have.
However, things are going to change as we bring in
autonomous vehicles. A tractor trailer trying to find a lane
line in Wyoming is the same as the system that is trying to
find a lane line in New York City; it needs to know that that
is a lane line. You can travel from place to place and see that
lanes here are 12 foot wide; this one is 8 foot wide. Lines are
10 foot; they are 15 foot. Everything starts to change because
we have allowed this engineering judgment. And it has been
critical up to this point, but it is also because they were
critical to have the engineering judgment to change based on
your area.
When we start having autonomous vehicles, it starts to
level that playing field. Finding the right technology and
working together so that we know what is the best way we can
get each car to know that that is a lane line. What is the best
way to get each car to know what stop signs are and not have
the confusion that we saw from the stop signs that Senator
Carper had put up? How do we get everyone organized on that so
we are making the best investment, especially considering that
we need to do both for a long time?
We foresee for a long time coming we are going to have both
human drivers and autonomous vehicles, and the last thing we
want to do is spend our money heading toward fixing one problem
and then creating problems for the other.
Senator Booker. I appreciate that.
I want to turn in the seconds I have left, and I see
Senator Markey has come, to the DOT Commissioner for--I am
sorry, which city is that again?
[Laughter.]
Senator Booker. It sits in the shadow of Newark, New
Jersey; I understand that.
[Laughter.]
Senator Booker. It is tough. It is tough. You and Senator
Schumer have city envy, I understand.
Senator Cardin asked the question about the challenges as
this technology comes in, and the excitement for me about our
metropolitan area is tons of new technology could really ease
what is an infrastructure crisis that we have in our region.
So, I wonder if everything from AVs to, frankly, just drones
alone could take a lot of the traffic off of our streets, and
in places like New York City, drone technologies might be sort
of a possible help.
I said this to the head of the FAA for a while, I said,
look, they were choking sort of the ability for localities to
experiment with new technologies, and I felt, overregulating
them. France, for example, is doing so much more on drones than
we were doing because they had better regulatory structure. And
I said, if you guys were around during the time of Wilbur and
Orville Wright, we would have never gotten off the ground.
So, I am just wondering, as this person that understands
the critical crisis we have. I live 12 miles, 11 miles from
Manhattan, and it can take me upwards of 2 hours to traverse
that. New Jersey transit, and again, all the work I have been
doing on the rail tunnels.
We are just in a crisis proportion. So how are you feeling
in the ability to sort of embrace innovation, to create
sandboxes for different technologies that you think are
critical? And are there things that we can learn from your
experiences in one of the most congest metropolitan areas on
the planet Earth, the greater Newark metropolitan area?
Ms. Trottenberg. It is great to be in the greater Newark
area.
You are right, Senator, congestion for us in the entire
metropolitan region is a huge, huge challenge. The potential of
AVs is that they can reduce congestion, but I can just say
right now, I think the New York City experience--and it is
probably true across the river, we are finding services like
Uber and Lyft, which say they will eventually be a ridesharing
system, right now they are just adding more cars to our
streets.
Senator Booker. Yes.
Ms. Trottenberg. And in midtown Manhattan, travel speeds
are really slowing, and it is a real challenge. I am actually
fortunate right now to be chairing Transcom, which is actually
a coalition of New York, New Jersey, and Connecticut
transportation and enforcement officials. It was actually the
group that got Easy Pass started throughout the region. And we
are actually trying, as a region, to look at some of what are
these next generations of technologies and bridge one of the
challenges we all face, which is how do we bring all our
jurisdictions together, New York, New Jersey, the Port
Authority, the MTA, New Jersey Transit. Because there is both
the technology piece and the jurisdictional and governance
piece in making sure, to the extent that we are all pursuing
these technologies, they can talk to each other and we can work
together.
Drones I think, for New York City, our airspace is pretty
dense. I don't know if we are ready to go there, but we are, as
we have said here today, doing connected vehicles and starting
to look at, again, things we could deploy regionally, because
we are one region.
Senator Booker. Yes.
And just a question for the record: Where do the Jets and
the Giants play?
Ms. Trottenberg. In the region.
[Laughter.]
Senator Barrasso. Case closed.
Senator Markey.
Senator Markey. Thank you, Mr. Chairman, once again.
Once again, this is an area where Senator Booker and I
compete. We compete. I just had him check it out. New York has
the second worse congestion and we are No. 7 in congestion,
Boston. One place where we would like to be behind New York in
this one category, maybe the only category. And Washington, DC,
is No. 6.
So, between New York, Boston, and Washington, we have a lot
of reason to solve this problem. It is consuming a large
percentage of the discretionary time that we have left on the
planet, just sitting in vehicles and waiting to get to places,
and this new revolution is really computers on wheels, just
going down the street. And these computers on wheels could soon
be able to send speed and direction data to other vehicles,
roads, bridges, other transportation infrastructure, to improve
safety, reduce traffic, improve efficiency.
In the very near future we may be spending Federal highway
funds to embed sophisticated sensors into our roads and our
bridges, and like the everyday computer and smartphones, these
transportation technologies could be vulnerable to cyberattacks
if appropriate safeguards are not put in place up front.
Just a few years ago hackers remotely took control of the
brakes, steering, and acceleration of a Jeep Cherokee. Chrysler
had to recall 1.7 million vehicles to fix this cybersecurity
problem. If we are to imagine a world where massive 18-wheelers
carrying hazardous materials and minivans full of children can
drive themselves, it shouldn't be a stretch of the imagination
to envision that these vehicles may be targets of cyberattacks.
And unlike many technologies that are already deployed, we have
the unique opportunity to address cybersecurity threats before
they emerge.
Mr. Kildare, do you believe that we should proactively
ensure that robust cybersecurity protections are built into the
design, construction, and operation of these transportation
technologies?
Mr. Kildare. Absolutely. It is a significant concern. It is
also a big concern that we have about the pending legislation,
the AV START Act, is that there is no consideration for
requirements. I believe the only thing that is in there is a
recommendation that companies have a game plan, but not
necessarily executed or follow any standards that are
available. We have great examples of how this can be done in
lots of other industries. We need to start taking those and
learning those lessons from history about how we make these
things secure.
Senator Markey. Beautiful. I agree with that, and that is
why I have introduced legislation with Senator Blumenthal, the
SPY Car Act, that directs NHTSA to establish Federal standards
to secure our cars.
Mr. Kildare, do you believe that it would be helpful if
NHTSA, the National Highway Traffic Safety Administration,
created a cybersecurity certification program for vehicles,
similar to NHTSA's five-star automobile safety rating program?
The program would reward manufacturers who adopt the strongest
protections, while also helping consumers make more informed
decisions when purchasing or riding in vehicles.
Mr. Kildare. Absolutely. I believe the benefits that we
have seen from the NHTSA five-star program have shown how we
can encourage competition in the goal of getting safety out the
forefront. The same thing can be done with the cyber dashboard
and looking at cybersecurity. We always want to see it move
toward regulation at the end, and we have seen that happen out
of the five-star program as well, so it is an excellent goal.
Senator Markey. Now consumers say, oh, how many miles per
gallon? We can see that. What is the safety rating? They can
see that. And going forward, this cybersecurity protection is
also going to be increasingly important because it is a
computer riding down the street, and there is a sinister side
to cyberspace. It is the best of technologies and the worst of
technologies simultaneously. There is a Dickensian quality to
it; it can enable, it can ennoble, it can degrade, it can
debase. We just saw that in the Facebook hearing. We continue
to see it. If we don't protect against the downside, then bad
things happen, so it is important for us to build in the
cybersecurity protections now.
Finally, the only thing more quintessentially Boston than
cold weather and the Boston Red Sox is traffic. But technology
could help address traffic issues by allowing communities to
more accurately and comprehensively monitor traffic patterns
and then take preventative measures--rather than reactionary
measures--to alleviate congestion.
How can these sophisticated transportation technologies
help us be more efficiently traveling, especially in the
densely populated urban areas of the country?
Ms. Trottenberg. I guess I will speak for densely populated
urban areas. I think, again, sort of speaking of the Dickensian
nature of these potential technologies, as you have heard, they
can potentially enable vehicles to travel together more
closely, to anticipate obstacles in the roadways, reduce
crashes and accidents, which obviously has a big effect on
moving traffic.
I think it is certainly true in the metropolitan region of
New York, and Boston as well. I just have to sort of have to
make a pitch, I guess, for a different committee's
jurisdiction. For us in the end, the biggest thing that is
going to help us probably solve our congestion problems is
going to also be robust investments in our mass transit system.
The efficiency of what a train can carry, even versus a
platooning set of vehicles, there is no comparison; the mass
transit system is always going to be the workhorse for New York
in terms of carrying the population.
Senator Markey. Thank you, Mr. Chairman.
Senator Barrasso. Thank you, Senator Markey.
I was just talking with the good Senator about he was too
young to remember the doctor show that we grew up with, Dr.
Kildare.
Senator Carper. Dr. Kildare, Ben Casey, Dr. Casey. There
were a bunch of them.
Senator Barrasso. There was Ben Casey, there was Dr.
Kildare, there was Marcus Welby, MD. That is why I became a
doctor.
[Laughter.]
Senator Markey. If I may add, that is why my wife became a
doctor, looking at Dr. Kildare. And the beginning of the show
is the front of Massachusetts General Hospital in Boston,
Massachusetts. All politics are local.
[Laughter.]
Senator Markey. The inspiration came out of that location.
Thank you.
Senator Carper. I know his wife. My wife and I are good
friends, and his wife is an inspiration.
The Chairman and I were talking just a little bit. I have
some prepared questions, but one that is just an audible here,
he represents a big State and a lot of roads, and a lot of
those roads in his State are dirt roads or gravel roads.
As Chairman Bhatt may recall as Secretary of Transportation
in Delaware, we used to have a whole lot of dirt roads and
gravel roads. We don't have quite as many anymore, but we have
a lot of roads in the more rural parts of our State that are
not marked. We still have some dirt roads and some gravel
roads, but there are a lot of places across the country that
are more like--not just the town of Wyoming, Delaware, but are
more like the State of Wyoming, where they have a lot of dirt
roads and gravel roads.
How do we handle that? How do we stripe them and get ready
for just minor, inexpensive infrastructure enhancements that
will help make autonomous vehicles work in those areas? How
does that work?
Mr. Bhatt. So, that is a great point to make, Senator
Carper, and a great question. One of the things that I know
Colorado DOT is taking a look at as a research program now,
working with other partners, SAE classifies autonomous vehicles
Level 1 through 5 based on their ability to either be driven or
self-driven.
What Colorado DOT has begun looking at, or the RoadX
program, is a roadway classification system. So, we will have
interstates that have great pavement and markings and signage
and connectivity, all the way down to a rural dirt road that
may not even have cellular coverage. Much like when you used to
buy a cellphone, you were provided a map around where that
cellphone could work, I think there will be part of our network
that will never have those levels of connectivity, and it will
be up to, as Commissioner Trottenberg said, the manufacturers
to make sure that their vehicles are able to drive on a vast
majority of our system.
Senator Carper. But as we all know, there are a lot of
times when we go around States around the country, we try to
use our cellphones, and they don't work. I know a lot of people
with autonomous vehicles say, well, I will take a chance; maybe
it will work today.
All right, thank you. That is interesting.
OK, a question, if I could, for Polly Trottenberg.
How are you today? Very nice to see you.
Have auto manufacturers and software developers shared
information with you about their technologies' reliance on
infrastructure in order to safely navigate public roads? This
information seems like it would be vital to the public agencies
who own and operate our transportation infrastructure when
setting standards and shaping the deployment of autonomous
vehicles to ensure both safety and compatibility.
I guess specifically do you think that access to safety
evaluation reports would better equip our public agencies in
their efforts to set appropriate standards for autonomous
vehicles?
Ms. Trottenberg. Absolutely, Senator. We think that is
crucial. I think Mr. Kildare put it well.
Senator Carper. Dr. Kildare.
Ms. Trottenberg. Dr. Kildare. It is essential for us, as he
was saying, as these vehicles hit the roads and start doing
their testing, to know where they had near misses, where they
might have interacted with a pedestrian or cyclist. It is
important for us to get the sense of their safety operations,
but again, it will help us as well, it will help us look at our
infrastructure. It is really valuable data.
Certainly, I think something to be very thoughtful of as
you are regulating at the Federal level and at the State level,
for a city like New York, which is such a big and unique entity
in and of itself, we really want to make sure that we have a
seat at the table and that we can be part of understanding that
data and very sensitive to protecting it and the proprietary
nature of it. But it is really key for us, in our dense urban
environment, to understand how these safety systems work and
what they are seeing if they hit our roadways.
Mr. Doerzaph. I will add a little to that, too. One of the
neat things about these new vehicles is they come with a
plethora of sensors that provide very direct information about
what it is in the infrastructure that creates a difficulty for
them, which can help an operator really prioritize how they
spend those public dollars by focusing on the elements. It may
be as simple as a reflection off of a particular bridge that
radar over-responds to. So, having not just general feedback to
where incidents are, but details about what it is that is
fouling up a particular sensing or perception system is new and
also potentially very valuable.
Senator Carper. All right, thank you.
Mr. Panos, you have been before us before; so has Shailen,
as have one or two others before us today. Are you a director
of the Department of Transportation? Is it director, secretary,
commissioner? What is it?
Mr. Panos. Director of the Department of Transportation.
Senator Carper. All right. I want to start out with a
question for the whole panel, but I will just start off with
you, if I could.
Recent studies from the University of Michigan and Ford
Motor Company suggest that adding AV equipment to cars adds
weight, aerodynamic drag, and electrical power consumption that
internally increase to fuel consumption. Because AVs could both
reduce the cost and increase the availability of driving that
could induce greater demand and increased vehicle usage, we may
have more vehicles on our roads. We talked about that earlier.
Empty vehicles circling city blocks and riders opting for
longer commutes.
With on road mobile source emissions already the largest
source of greenhouse gas emissions, in our economy, are you
concerned that widespread autonomous vehicle deployment may
actually worsen this problem? Would you agree that we need more
data about how these vehicles will be used in the real world to
better inform our understanding of the role that autonomous
vehicles will play in transportation sector greenhouse gas
emissions?
Mr. Panos. With regards to the idea that we need more
research dollars and more focus on deployment of connected and
autonomous vehicles throughout the United States and the
various environments--some of them represented by some of my
colleagues here today--I would say yes, that we do need to
invest more research dollars, and already have. The
collaborations between the USDOT and AASHTO, the collaborations
between the industry and government have been robust, but need
to continue. And the investments in infrastructure research for
connected and autonomous vehicles need to continue.
Senator Carper. All right.
Shailen.
Mr. Bhatt. Senator, I began my career trying to reduce
single occupant vehicle trips, trying to get people into high
occupancy vehicles. The idea that we would have zero occupancy
vehicle trips with fleets of autonomous vehicles circling
without anybody in them is antithetical to intelligent
mobility, so I think we would want to make sure, in working
with States and local partners, to make sure that that was not
an outcome that we got.
And your point, I think, is well taken on the increased
fuel consumption. We haven't spoken as much about it today, but
I think a great belief, at least globally, is that the future
of mobility also includes a great deal of electrification of
the fleet, and that will be something that will help with some
of the emission issues.
Senator Carper. That is a good point. Good point. Thank
you.
Is it Dr. Doerzaph? Thank you. Same question.
Mr. Doerzaph. I am going to echo what both of them said.
Some of that is also a sign, I think, of the current maturity
of the industry. Sensor racks are hanging off the cars because
we are taking legacy cars and retrofitting them. As we move
toward vehicles which are designed for automation, some of the
aerodynamic drag aspects will fall away. Some of the weight
will probably fall away, as well, as those systems are
optimized.
And then, mostly, I would just echo that, yes, zero
passenger vehicles should be minimized. If we are going to
really have a profound impact on congestion, we need to double
or triple or better on every vehicle.
Senator Carper. All right, thanks.
Commissioner Trottenberg.
Ms. Trottenberg. I think you have----
Senator Carper. Was Iris Schumer commissioner?
Ms. Trottenberg. Yes, it was Iris.
Senator Carper. During the Bloomberg administration?
Ms. Trottenberg. She was Giuliani and Bloomberg.
Senator Carper. OK.
Ms. Trottenberg. You have asked the key question. Today we
are focusing so much, first and foremost, understandably, on is
the technology safe, how do we ensure it is safe, but you are
really getting to what I think, in urban areas, is the key
public policy question. If we deploy this safe technology, and
all it does is further congest our streets, I think this will
not have been a very successful program.
I think it does bring a lot of policy considerations in,
and ones that I will again just make a pitch, localities really
need to be part of that process. In New York, we are now seeing
Ubers and Lyfts are congesting our streets. They are not no
occupant vehicles, but they are one occupant vehicles spending
a lot of time cruising in our central business district.
So, we need to both get the technology safe and then
grapple with the key public policy parts of this. If we are
just adding to the congestion and fuel consumption and
emissions, then this technology won't have realized its
potential.
Senator Carper. All right, thank you.
Dr. Kildare.
Mr. Kildare. Thank you for the question. I believe this
actually highlights what everyone has been saying here, that
there are a lot of questions around these vehicles and what the
implications are going to be, and that is one of Advocates' big
points about the issues with the AV START Act, that there is a
push toward deregulation and toward preempting some of the
States from controlling what is going on with the lack of
Federal leadership, a lack of leadership from NHTSA.
We need that coordination between NHTSA and Federal
Highway, and we need to be moving toward regulation, not away
from it, to making sure that we are going to get the proper
results that we are looking for and not have these unintended
results that could then compound and cost billions of dollars
when it comes to infrastructure improvements that we need to
address the problems that are now created.
Senator Carper. Mr. Chairman, I have two more questions
that are fairly short.
A question, if I could, for Dr. Doerzaph. One of the risks
that connectivity introduces, the ability to spoof--we talked
about this a little bit earlier--spoof, to hack or trick
vehicles into behaving in ways they should not, the stop sign
into 45 mile speed limit sign, such as believing a red light is
green or a stop sign is a speed limit sign. Research news
stories in the California Disengagement Reports have been
useful to identify the trouble that automated vehicles have
with consistently and correctly identifying traffic lights,
bicycles, bridges, pedestrians, animals, and other vehicles.
Given how difficult it is to already maintain our physical
infrastructure free of defects, do you believe that we can
develop a nationwide physical and digital infrastructure that
will be pretty much foolproof, to the point that we can and
will safely rely on it?
Mr. Doerzaph. That is a great question. It gets back to the
security by design question that we were talking about earlier,
so starting that conversation now and baking it into every
aspect of those systems is critical.
I think it is also really important to realize or to think
through that, yes, any one subsystem can have a failure, much
like humans do. We may see an illusion or hear something that
is not actually there, or misrepresent a threat. AV is subject
to very similar set of conditions. They are slightly different
because their sensors are different, but one of the advantages
of AV is that we can have many more sensors.
So, the sensor fusion aspect, where, yes, the vision system
sees a sign that is inappropriate, well, that is OK; the
mapping system knows the correct. And by the way, it is not
just the maps, two maps, it is a local one that is stored and
is known to be highly secured, and it has been confirmed by a
recently updated map as well.
So, really, the security is multifaceted. There is the
aspects of the security which are keeping bad actors from
tampering with the systems, but also the aspect of ensuring
that what is being perceived by a sensor external to the
vehicle is being confirmed by multiple subsystems as well.
So, if that design is appropriately managed, as the
evolution occurs, I think it is reasonable to reach a secure
system. At the same time, we need to be mindful that the
population of hackers and whatnot are bright, and they move
with the times as well, so we need to be able to identify those
bad actors and successfully remove them from the trust network,
which, again, requires a very robust, nationally synchronized,
security mechanism.
Senator Carper. All right, thanks.
Last question I would start with Secretary Bhatt. I don't
know if anyone else would like to comment on this question, but
if you want to, you are welcome to do that.
The solvency of the Nation's Highway Trust Fund is, as you
know, a top concern for our Committee. What are the budgetary
implications of enhanced deployment of autonomous vehicles on
public roadways and the cost to enable infrastructure to be
digitally connected? Are these costs likely to exacerbate the
funding challenges that we face already in public agencies on
this front? On the flip side, is there a new opportunity to
collect mileage based user fees on vehicles that are digitally
connected to the infrastructure that they are using?
Mr. Bhatt. Senator, thank you for that question. I think
that you raise a couple of very important issues. As Director
Panos has mentioned, most of the striping and signage that is
critical for the operation of connected autonomous vehicles as
we move forward comes out of maintenance budgets, and if you
are going to increase your maintenance budget, that is going to
obviously impact your construction budget.
And as Commissioner Trottenberg mentioned earlier, the
manufacturers need to make sure that they are factoring in the
fact that many roads won't be able to be upgraded, given the
budget challenges. So, I think that we need to make sure that
we invest in technology and in infrastructure, because I think
that technology can be used to leverage infrastructure to get
more productivity for our economy. I think of it as a global
economic competitiveness issue.
I also believe that there has been a lot of discussion over
the years; you have led many of the discussions around funding
and gas tax issues. Outside of that, I think that the new
technology is the best way for us to move to whatever will
succeed the gas tax, whether it is a mileage based user fee or
VMT or road usage charge, because I think we now have the
technology piece that can both be secure and delay some privacy
concerns that have been out there, and those are the
discussions we need to begin with now.
Senator Barrasso. Thank you, Senator Carper.
One quick question for Mr. Kildare. Mr. Doerzaph talks
about this plethora of sensors that are available, and I think
you had talked about the idea of individuals have to go and get
an eye exam at the Department of Motor Vehicles, and should
each vehicle would independently have to go and get examined to
make sure all the sensors are working right, or is it something
that would be part of the patenting process when they design
that? I am trying to figure out how that all plays into what we
are looking at.
Mr. Kildare. Using the term loosely for AV vision test is
establishing requirements for what vehicles need to be able to
see and respond to, that we need to design, if we take what we
see at the FAA as an example, and we look at either safety the
intended function, or we look at functional safety, your
product needs to do the things it says it can do, and it needs
to not do the things it says it can't do. And making sure that
vehicles aren't operated outside of those envelopes are very
important.
We had the mention about the sensor fusion, getting sensors
to work together. The first part will be can your sensors see
that stop sign. Does it know that it is a stop sign? Does it
know that it is a stop sign when it has been molested in some
way, either in terms of graffiti or art work or bent or--I have
been through Texas--shot at? They are used for target practice.
There are a lot of changes that can happen out there. We need
to establish what are the requirements for whatever that
operation is going to be.
We know in the Tesla crash down in Florida, that that
vehicle was not supposed to be operating under autopilot on
that road, but it was. So how do we do that? What are the
requirements we need to establish? If you tell us your vehicle
operates only on highways, how do you prove that that vehicle
only operates on highways? The same thing for the vision. What
does it need to see?
The vision test that we have implies some responsibility on
the part of the driver. We ask people to identify a stop sign
because they have to stop the vehicle. Now that we are testing
the machine, it has to see the stop sign at all times, and it
has to decide to stop the vehicle. That is what we have to
require. That is the concern that we have with what we are
seeing in the AV START Act, is that we don't see any of these
requirements coming up, and we need them.
Senator Barrasso. I appreciate it.
I appreciate the panel and all your thoughts.
Senator Carper, if you read broadly on this, the technology
of where this may go to the issue of these driverless cars
circling or do they go someplace else, if they go someplace
else, people write about eliminating many parking lots and
having an opportunity for actually more buildings in
communities.
What does it change in the real estate markets? An article
in The Economist this week, under Free Exchange, called ``Road
Hogs,'' says economies of scale will push the market for
driverless vehicles toward a monopoly.
There are many different components of all of this. They
are talking about car dealers no longer going to sell to an
individual, but to a fleet, and how does this change the number
of vehicles on the market.
I don't know that any of us, if you read enough and
different futurists who look at this, it doesn't seem to be
that there is a uniform agreement as to where the future may be
taking us as this all plays out in time, so I am just very
grateful that all of you would take your time today to come
here to testify from a variety of backgrounds, but to give us
your very best thoughts. I thought it was a very helpful
hearing, and I appreciate all of you.
The hearing record is going to remain open for a couple
more weeks--actually 2 weeks--so some other members who have
been here and haven't had a chance to ask questions may submit
written questions to you, so I want to thank you again for your
testimony today on this very important issue.
With that, the hearing is adjourned.
[Whereupon, at 11:39 a.m. the Committee was adjourned.]
[Additional material submitted for the record follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
[all]