[House Hearing, 115 Congress]
[From the U.S. Government Publishing Office]
SELF-DRIVING CARS: LEVELS OF AUTOMATION
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON DIGITAL COMMERCE AND CONSUMER PROTECTION
OF THE
COMMITTEE ON ENERGY AND COMMERCE
HOUSE OF REPRESENTATIVES
ONE HUNDRED FIFTEENTH CONGRESS
FIRST SESSION
__________
MARCH 28, 2017
__________
Serial No. 115-19
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Printed for the use of the Committee on Energy and Commerce
energycommerce.house.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
25-681 PDF WASHINGTON : 2017
----------------------------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Publishing Office,
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center,
U.S. Government Publishing Office. Phone 202-512-1800, or 866-512-1800 (toll-free).
E-mail, [email protected].
COMMITTEE ON ENERGY AND COMMERCE
GREG WALDEN, Oregon
Chairman
JOE BARTON, Texas FRANK PALLONE, Jr., New Jersey
Vice Chairman Ranking Member
FRED UPTON, Michigan BOBBY L. RUSH, Illinois
JOHN SHIMKUS, Illinois ANNA G. ESHOO, California
TIM MURPHY, Pennsylvania ELIOT L. ENGEL, New York
MICHAEL C. BURGESS, Texas GENE GREEN, Texas
MARSHA BLACKBURN, Tennessee DIANA DeGETTE, Colorado
STEVE SCALISE, Louisiana MICHAEL F. DOYLE, Pennsylvania
ROBERT E. LATTA, Ohio JANICE D. SCHAKOWSKY, Illinois
CATHY McMORRIS RODGERS, Washington G.K. BUTTERFIELD, North Carolina
GREGG HARPER, Mississippi DORIS O. MATSUI, California
LEONARD LANCE, New Jersey KATHY CASTOR, Florida
BRETT GUTHRIE, Kentucky JOHN P. SARBANES, Maryland
PETE OLSON, Texas JERRY McNERNEY, California
DAVID B. McKINLEY, West Virginia PETER WELCH, Vermont
ADAM KINZINGER, Illinois BEN RAY LUJAN, New Mexico
H. MORGAN GRIFFITH, Virginia PAUL TONKO, New York
GUS M. BILIRAKIS, Florida YVETTE D. CLARKE, New York
BILL JOHNSON, Ohio DAVID LOEBSACK, Iowa
BILLY LONG, Missouri KURT SCHRADER, Oregon
LARRY BUCSHON, Indiana JOSEPH P. KENNEDY, III,
BILL FLORES, Texas Massachusetts
SUSAN W. BROOKS, Indiana TONY CARDENAS, California
MARKWAYNE MULLIN, Oklahoma RAUL RUIZ, California
RICHARD HUDSON, North Carolina SCOTT H. PETERS, California
CHRIS COLLINS, New York DEBBIE DINGELL, Michigan
KEVIN CRAMER, North Dakota
TIM WALBERG, Michigan
MIMI WALTERS, California
RYAN A. COSTELLO, Pennsylvania
EARL L. ``BUDDY'' CARTER, Georgia
7_____
Subcommittee on Digital Commerce and Consumer Protection
ROBERT E. LATTA, Ohio
Chairman
GREGG HARPER, Mississippi JANICE D. SCHAKOWSKY, Illinois
Vice Chairman Ranking Member
FRED UPTON, Michigan BEN RAY LUJAN, New Mexico
MICHAEL C. BURGESS, Texas YVETTE D. CLARKE, New York
LEONARD LANCE, New Jersey TONY CARDENAS, California
BRETT GUTHRIE, Kentucky DEBBIE DINGELL, Michigan
DAVID B. McKINLEY, West Virgina DORIS O. MATSUI, California
ADAM KINZINGER, Illinois PETER WELCH, Vermont
GUS M. BILIRAKIS, Florida JOSEPH P. KENNEDY, III,
LARRY BUCSHON, Indiana Massachusetts
MARKWAYNE MULLIN, Oklahoma GENE GREEN, Texas
MIMI WALTERS, California FRANK PALLONE, Jr., New Jersey (ex
RYAN A. COSTELLO, Pennsylvania officio)
GREG WALDEN, Oregon (ex officio)
(ii)
C O N T E N T S
----------
Page
Hon. Robert E. Latta, a Representative in Congress from the State
of Ohio, opening statement..................................... 1
Prepared statement........................................... 3
Hon. Janice D. Schakowsky, a Representative in Congress from the
State of Illinois, opening statement........................... 4
Hon. Greg Walden, a Representative in Congress from the State of
Oregon, opening statement...................................... 5
Prepared statement........................................... 6
Hon. Frank Pallone, Jr., a Representative in Congress from the
State of New Jersey, opening statement......................... 7
Prepared statement........................................... 8
Witnesses
Jeff Klei, President, North America Automotive Divisions,
Continental AG................................................. 9
Prepared statement........................................... 12
S. William Gouse, Diretor, Federal Programs Development, SAE
International.................................................. 15
Prepared statement........................................... 18
David S. Zuby, Executive Vice President and Chief Research
Officer, Insurance Institute for Highway Safety................ 24
Prepared statement........................................... 25
Kay Stepper, Ph.D., Vice President for Automated Driving and
Driver Assistance Systems, Robert Bosch LLC.................... 34
Prepared statement........................................... 36
Submitted Material
Report, ``A Roadmap to Safer Driving Through Advanced Driver
Assistance Systems,'' Motor & Equipment Manufacturers
Association, \1\ submitted by Mr. Costello
Letter of March 27, 2017, from Jacqueline S. Gillan, President,
and Catherine Chase, Vice President of Governmental Affairs,
Advocates for Highway & Auto Safety, to Mr. Latta and Ms.
Schakowsky, submitted by Mr. Costello.......................... 70
Response to Request for Comment, DOT Docket No. NHTSA 092016
090090, Advocates for Highway & Auto Safety, December 2, 2016,
\2\ submitted by Mr. Costello
Statement of National Safety Council, March 28, 2017, submitted
by Mr. Costello................................................ 74
Statement of John Bozzella, President and Chief Executive
Officer, Global Automakers, March 28, 2017, submitted by Mr.
Costello....................................................... 83
Response to Request for Comment, DOT Docket No. NHTSA 092016
090090, U.S. Chamber of Commerce, November 22, 2016, submitted
by Mr. Costello................................................ 87
----------
\1\ The information has been retained in committee files and also is
available at http://docs.house.gov/meetings/IF/IF17/20170328/105790/
HHRG-115-IF17-20170328-SD011.pdf.
\2\ The information has been retained in committee files and also is
available at http://docs.house.gov/meetings/IF/IF17/20170328/105790/
HHRG-115-IF17-20170328-SD004.pdf.
Statement of American Car Rental Association, March 28, 2017,
submitted by Mr. Costello...................................... 97
Statement of Dan Galves, Senior Vice President, Chief
Communications Officer, Mobileye, March 28, 2017, submitted by
Mr. Costello................................................... 100
Letter of March 27, 2017, from Marc Rotenberg, President, et al.,
Electronic Privacy Information Center, to Mr. Latta and Ms.
Schakowsky, submitted by Mr. Costello.......................... 105
Letter of March 28, 2017, from Honda North America, Inc., to Mr.
Latta and Ms. Schakowsky, submitted by Mr. Costello............ 108
SELF-DRIVING CARS: LEVELS OF AUTOMATION
----------
TUESDAY, MARCH 28, 2017
House of Representatives,
Subcommittee on Digital Commerce and Consumer
Protection,
Committee on Energy and Commerce,
Washington, DC.
The subcommittee met, pursuant to call, at 10:05 a.m., in
Room 2322 Rayburn House Office Building, Hon. Robert E. Latta
(chairman of the subcommittee) presiding.
Members present: Representatives Latta, Harper, Lance,
McKinley, Kinzinger, Bilirakis, Mullin, Walters, Costello,
Walden (ex officio), Schakowsky, Clarke, Cardenas, Dingell,
Matsui, Welch, Kennedy, Green, and Pallone (ex officio).
Staff present: Ray Baum, Staff Director; Blair Ellis, Press
Secretary/Digital Coordinator; Melissa Froelich, Counsel,
Digital Commerce and Consumer Protection; Adam Fromm, Director
of Outreach and Coalitions; Giulia Giannangeli, Legislative
Clerk, Digital Commerce and Consumer Protection/Communications
and Technology; Paul Nagle, Chief Counsel, Digital Commerce and
Consumer Protection; Olivia Trusty, Professional Staff Member,
Digital Commerce and Consumer Protection; Madeline Vey, Policy
Coordinator, Digital Commerce and Consumer Protection; Hamlin
Wade, Special Advisor for External Affairs; Michelle Ash,
Minority Chief Counsel, Digital Commerce and Consumer
Protection; Jeff Carroll, Minority Staff Director; Lisa
Goldman, Minority Counsel; Caroline Paris-Behr, Minority Policy
Analyst; Matt Schumacher, Minority Press Assistant; Andrew
Souvall, Minority Director of Communications, Member Services,
and Outreach.
Mr. Latta. Well, good morning. I would like to welcome you
all to our subcommittee meeting of the Digital Commerce and
Consumer Protection this morning. I really appreciate our
witnesses being here. We are going to have members coming in.
There is a meeting going on downstairs, and so more folks will
be coming in. We see a couple more coming in right now. But I
really again appreciate you so for being here, and to get
started I would like to recognize myself for 5 minutes for an
opening statement.
OPENING STATEMENT OF HON. ROBERT E. LATTA, A REPRESENTATIVE IN
CONGRESS FROM THE STATE OF OHIO
Again, good morning. And last month, this subcommittee
examined how automakers and other entities are testing self-
driving vehicles and preparing for the development of this
lifesaving technology. While projections for the development of
self-driving vehicles remains years out, advanced driver
assistance systems that offer self semi-autonomous driving
capabilities are entering the marketplace today.
Advanced driver assistance systems are crash avoidance
technologies that can protect drivers, reduce crashes, and
enhance the convenience of driving. Forward collision warning,
blind spot detection, and lane departure warnings are examples
of advanced driver assistance systems. These systems help
drivers make safer decisions on the road by providing real-time
information about surrounding roadway activity. The driver can
receive this information through audible tones, steering wheel
vibrations, or small flashing lights on side mirrors alerting
the driver of potential safety hazards on the road.
Increasingly, advanced driver assistance systems now
entering the market are capable of taking a more active role in
the driving task. Innovative systems such as automatic
emergency braking and lane departure prevention can temporarily
take control over parts of the vehicle's critical safety
functions such as braking or steering. This can occur by the
system either applying the brakes without input from the driver
or steering the vehicle back into marked lanes following
unintended drifting.
Automakers and equipment suppliers have announced
additional innovative driver assistance systems that are
currently in line for deployment. Traffic jam assist can take
control of a vehicle's functions in low speed, stop and go
traffic. Autonomous valet parking can park itself and retrieve
itself when summoned by the owner. And highway autopilot with
lane changing is being developed to change lanes and pass other
vehicles without the input of the human driver.
The deployment of the advanced driver assistance systems is
demonstrating significant safety benefits across the country.
Studies are showing that advanced driver assistance systems and
crash avoidance technologies are reducing crashes, roadway
injuries, and insurance claims. Advanced driver assistance
systems are also an essential part in laying the groundwork for
the deployment of fully self-driving vehicles.
Through technological advances by manufacturers and
equipment suppliers, basic driver assistance systems are taking
on more advanced capabilities that assume greater control of
the vehicle's critical safety functions throughout a driving
trip. The progression of these technologies is incrementally
removing the human driver from the driving task and paving the
way to full autonomy.
To provide consistency in the development of driver
assistance safety technologies, standards-setting organization
SAE International developed a classification system to define
six different levels of driving automation. SAE levels of
automation establish the general scope of the driver assistance
system and the role of the human driver in vehicles taking on
increasing autonomous driving capabilities.
The levels span from a vehicle with no automation all the
way to a vehicle with full automation or a fully self-driving
vehicle. Last September, the National Highway Traffic Safety
Administration adopted SAE's levels of automation for its own
use in its Federal Automated Vehicles Policy.
As we discuss the levels of vehicle automation today, I
look forward to learning more about the capabilities of
advanced driver assistance systems currently on the market and
how these technologies are increasing vehicle safety and
protecting America's motorists. I look forward to examining how
these systems are informing the development of fully self-
driving vehicles and how the auto industry is working to make
these systems available across all models and fleets.
I also look forward to hearing from witnesses about how
consumers are adopting these technologies and how they are
helping to build consumers' confidence in automated driving
systems. And with that I will end my opening statement.
[The prepared statement of Mr. Latta follows:]
Prepared statement of Hon. Robert E. Latta
Good morning. Last month, this subcommittee examined how
automakers and other entities are testing self-driving vehicles
and preparing for the deployment of this life-saving
technology. While projections for the deployment of self-
driving vehicles remain years out, advanced driver assistance
systems that offer semiautonomous driving capabilities are
entering the marketplace today.
Advanced driver assistance systems are crash avoidance
technologies that can protect drivers, reduce crashes, and
enhance the convenience of driving. ``Forward Collision
Warning,'' ``Blind Spot Detection,'' and ``Lane Departure
Warning'' are examples of advanced driver assistance systems.
These systems help drivers make safer decisions on the road by
providing real-time information about surrounding roadway
activity. The driver can receive this information through
audible tones, steering wheel vibrations, or small flashing
lights on side mirrors, alerting the driver to potential safety
hazards on the road.
Increasingly, advanced driver assistance systems now
entering the market are capable of taking a more active role in
the driving task. Innovative systems such as ``Automatic
Emergency Braking'' and ``Lane Departure Prevention'' can
temporarily take control over parts of the vehicle's critical
safety functions such as braking or steering. This can occur by
the system either applying the brakes without input from the
driver or steering the vehicle back into marked lanes following
unintended drifting.
Automakers and equipment suppliers have announced
additional innovative driver assistance systems that are
currently in line for deployment. ``Traffic jam assist'' can
take control of a vehicle's functions in low-speed, stop and go
traffic. ``Autonomous valet parking'' can park itself and
retrieve itself when summoned by the owner. And, ``highway
autopilot with lane changing'' is being developed to change
lanes and pass other vehicles without the input of a human
driver.
The deployment of advanced driver assistance systems is
demonstrating significant safety benefits across the country.
Studies are showing that advanced driver assistance systems and
crash avoidance technologies are reducing crashes, roadway
injuries, and insurance claims.
Advanced driver assistance systems are also an essential
part in laying the groundwork for the deployment of fully self-
driving vehicles. Through technological advancements by
manufacturers and equipment suppliers, basic driver assistance
systems are taking on more advanced capabilities that assume
greater control of the vehicle's critical safety functions
throughout a driving trip. The progression of these
technologies is incrementally removing the human driver from
the driving task and paving the way to full autonomy.
To provide consistency in the development of driver
assistance safety technologies, standards-setting organization,
SAE (S-A-E) International, developed a classification system
that defines six different levels of driving automation. SAE's
levels of automation establish the general scope of the driver
assistance system and the role of the human driver in vehicles
taking on increasing autonomous driving capabilities. The
levels span from a vehicle with no automation all the way to a
vehicle with full automation or a fully self-driving vehicle.
Last September, the National Highway Traffic Safety
Administration adopted SAE's levels of automation for its own
use in its Federal Automated Vehicles Policy.
As we discuss the levels of vehicle automation today, I
look forward to learning more about the capabilities of
advanced driver assistance systems currently on the market and
how these technologies are increasing vehicle safety and
protecting America's motorists. I look forward to examining how
these systems are informing the development of fully self-
driving vehicles and how the auto industry is working to make
these systems available across all models and fleets. I also
look forward to hearing from witnesses about how consumers are
adopting these technologies and how they are helping to build
consumers' confidence in automated driving systems.
Mr. Latta. I would like to recognize for 5 minutes the
gentlelady from Illinois, the ranking member, for 5 minutes.
Good morning.
OPENING STATEMENT OF HON. JANICE D. SCHAKOWSKY, A
REPRESENTATIVE IN CONGRESS FROM THE STATE OF ILLINOIS
Ms. Schakowsky. Good morning and thank you, Mr. Chairman
and our witnesses. Today's hearing continues our subcommittee's
series on autonomous vehicles. In last month's hearing, several
of our witnesses referenced different levels of automation and
today we will better define those levels and we will also ask
about the effectiveness of existing safety technologies.
Self-driving cars are part of a long-term vision to
minimize accidents due to human error. Automated features are
becoming increasingly common in our cars, but we still have a
long way to go to reach full automation, Level 5, as SAE would
call it. Technology must be sufficiently tested and ensure that
we don't replace human error with system error. In addition,
the Takata and Volkswagen scandals raised serious questions
about how much we can trust industry to do the right thing on
safety.
Volkswagen ordered its supplier to write software to cheat
on emissions testing. With software increasingly integral to
our vehicles, proper oversight becomes that much more
challenging. Ultimately, the success of autonomous features and
self-driving cars relies on consumers trusting the technology.
Trust must be earned. Once technologies are put in new vehicles
it takes decades for technology to become widespread among all
vehicles on the road.
Just look at backup cameras. I worked to require backup
cameras after I met and talked to parents who were devastated
after their children were injured or killed in backover
accidents. We passed that law in 2008. Parents and advocates
came to DC regularly during the rulemaking process, and NHTSA
finally established the standard in 2014. And backup cameras
will now be required in all vehicles starting in model year
2018, 10 years after the bill passed.
It will still be years before the passenger vehicles
without backup cameras cycle out of use. A car sold today may
be on the road for another 2 decades. That is why it is
critical we look not only at safety improvements in the long
term, but also at which technologies can be effectively
deployed right now to save lives.
A lot of safety technologies are out there. However, some
are more effective than others. Automatic braking for instance
has proven very effective in reducing accidents. The evidence
on lane departures systems is more mixed. Today we will hear
from the suppliers that develop safety technologies. We will
hear about the testing data that is essential to lawmakers as
we consider what should be standard, and we will learn about
classifying levels of automation, a useful framework as we
think about how we move from today's cars to the self-driving
cars of the future.
It is a long road ahead, but as I have seen in my years on
the subcommittee we have to push forward at every step in the
process to make safety improvements a reality. I thank all of
our witnesses for being here today, and I look forward to your
testimony. And now I would like to yield the remaining time to
Representative Matsui.
Ms. Matsui. Thank you very much, Ranking Member Schakowsky.
Innovation and AV vehicle technology is moving at an ever-
accelerating pace. We are seeing major investments from
traditional auto manufacturers, suppliers like our witnesses
from Bosch and Continental, and new entrants like technology
companies and ride-sharing platforms. I believe we will make
big leaps forward in this space sooner than any of us would
have anticipated.
Different companies are pursuing different levels of
automation and we know that they do not need to move
sequentially through each level of automation. Some companies
are choosing to incorporate certain individual features of
automation while others are investing in a more integrated
Level 4 automation systems today.
In my district in Sacramento we are looking aggressively to
the future to lay the foundation for fully autonomous vehicles
to be tested on our roads. We are rapidly moving towards a time
when truly driverless cars will be on our roads and will
coexist with human drivers and other vehicles with different
levels of automation.
I look forward to hearing more from our witnesses today and
working with all of you to accelerate the testing and
deployment of this exciting technology which holds so much
promise for improving safety on our roads. I thank you and I
yield back.
Ms. Schakowsky. And I yield back.
Mr. Latta. Thank you very much. The gentlelady yields back
the balance of her time, and at this time the Chair recognizes
the gentleman from Oregon, the chairman of the full committee,
for 5 minutes for an opening statement.
OPENING STATEMENT OF HON. GREG WALDEN, A REPRESENTATIVE IN
CONGRESS FROM THE STATE OF OREGON
Mr. Walden. I thank the chairman and I welcome our
witnesses and look forward to your delivery of your testimony
which I have read and appreciate.
Following years of declining traffic fatalities, we have
seen tragically a sharp rise in vehicle-related deaths over the
past 2 years. According to early estimates, more than 40,000
Americans, 40,000 people, lost their lives on our Nation's
roads last year. That marks a 6 percent increase from 2015. And
in my own State of Oregon, 2016 was the deadliest year on the
roads in more than a decade, up 20 percent from the year
before.
These are sobering numbers. The development of self-driving
cars could be a solution to this uptick in danger facing the
driving public, the main question is how do we get there? Last
month, this subcommittee examined how automakers and other
entities are testing self-driving cars and that we are still
years away from getting them into hands of consumers.
But that has not stopped the automotive industry from
laying the foundation for a complete vehicle autonomy. Today,
many cars on the market, including one that my wife owns, are
equipped with active safety features or semi-autonomous driving
systems. It is pretty impressive to see them in action. These
systems have the potential to keep a vehicle within its
designated lane; accelerate to pass another vehicle; change
lanes, brake, and park all without the input of a human driver.
These advanced driver assistance systems or crash avoidance
technologies represent the building blocks to a fully self-
driving car. Gradually allowing the vehicle to perform parts of
the driving task absent human control means that vehicles are
steadily learning how to operate alone and consumers are
progressively becoming more familiar and more comfortable with
automated driving systems. The advancement of driver assistance
systems over the last decade, it is already demonstrating this
progression as this technology is minimizing crashes, reducing
injuries, and decreasing insurance claims.
In recognition of the safety benefits provided by these
systems, the National Highway Traffic Safety Administration has
begun to formally incorporate many of these technologies in its
5-Star safety ratings program. Today's hearing will look more
closely at many of the advanced driver assistance systems and
crash avoidance technologies that are on the road. Our
witnesses will also help us to understand the different levels
of driving automation, how these technologies are improving
safety, and how the development of driver assistance systems
and technologies is paving the way for fully self-driving cars.
We often say the development of self-driving cars is a
lifesaving endeavor. Following a devastating year on our
Nation's roads this could not be any more true. I look forward
to a thoughtful and engaging discussion on the levels of
driving automation and how advanced driver assistance systems
can lead us to the future of a full vehicle autonomy on our
road systems.
So thanks for the work you all are doing, thanks for
sharing your comments with us. We want to make sure to advance
this innovation and technology and save lives on our roads and
in our communities.
[The prepared statement of Mr. Walden follows:]
Prepared statement of Hon. Greg Walden
Following years of declining traffic fatalities, there has
been a sharp rise in vehicle-related deaths over the past 2
years. According to early estimates, over 40,000 people lost
their lives on our Nation's roads last year, marking a six
percent increase from 2015. In Oregon, 2016 was the deadliest
year on the roads in more than a decade, up 20 percent from the
year before. These are sobering numbers.
The development of self-driving cars could be a solution to
this uptick in danger facing the driving public. The main
question is: how do we get there?
Last month, this subcommittee examined how automakers and
other entities are testing self-driving cars and preparing this
innovative safety technology for commercial deployment. Just
about everyone concedes that fully self-driving cars are still
years away from getting into the hands of consumers; but, that
has not stopped the automotive industry from laying the
foundation for complete vehicle autonomy.
Today, many cars on the market are equipped with active
safety features or semi-autonomous driving systems. These
systems have the potential to keep a vehicle within its
designated lane; accelerate to pass another vehicle; change
lanes; brake; and park--all without the input of a human
driver. These advanced driver -assistance systems or crash-
avoidance technologies represent the building blocks to a fully
self-driving car.
Gradually allowing the vehicle to perform parts of the
driving task absent human control means that vehicles are
steadily learning how to operate alone and consumers are
progressively becoming more familiar and more comfortable with
automated driving systems.
The advancement of driver assistance systems over the last
decade is already demonstrating this progression, as this
technology is minimizing crashes, reducing injuries, and
decreasing insurance claims. In recognition of the safety
benefits provided by these systems, the National Highway
Traffic Safety Administration has begun work to formally
incorporate many of these technologies into its 5-Star Safety
Ratings program.
Today's hearing will look more closely at many of the
advanced driver assistance systems and crash avoidance
technologies on the road. Our witnesses will also help us to
understand the different levels of driving automation; how
these technologies are improving safety; and how the
development of driver assistance systems and technologies is
paving the way for fully self-driving cars.
We often say that the development of self-driving cars is a
life-saving endeavor. Following a devastating year on our
Nation's roads, this could not be truer now. I look forward to
a thoughtful and engaging discussion on the levels of driving
automation and how advanced driver assistance systems can lead
us to a future of full vehicle autonomy.
Mr. Walden. With that Mr. Chairman, I don't know if anybody
else on our side--I would yield to the gentleman from
Mississippi for the remainder of my time.
Mr. Harper. Thank you, Mr. Chairman. Thank you, Chairman
Latta, for calling this hearing today to continue the
subcommittee's efforts to explore the world of self-driving
cars. As I have mentioned at our previous hearings, this topic
is of particular interest to me because of the potential
opportunities that self-driving cars would provide to Americans
with disabilities, including those with intellectual
disabilities.
In the disability community lack of transportation is
widely viewed as the top impediment to advancement and success
in society. Self-driving cars could offer the disability
community a new method of transportation to potentially remove
this roadblock and provide them additional independence that
would open the doors to access new job markets and
opportunities to have an even more active role in our society,
which benefits us all.
I am looking forward to learning more about the
capabilities of advanced driver assistance systems and crash
avoidance technologies that are currently on the market and how
these capabilities will advance the future of self-driving
cars. And with that I yield back.
Mr. Latta. Thank you. The gentleman yields back, and the
Chair now recognizes for a 5-minute opening statement the
gentleman from New Jersey, the ranking member of the full
committee.
OPENING STATEMENT OF HON. FRANK PALLONE, JR., A REPRESENTATIVE
IN CONGRESS FROM THE STATE OF NEW JERSEY
Mr. Pallone. Thank you, Chairman Latta. Today's hearing
gives us a our first true opportunity to talk about what is
happening now in automated technology. While learning about the
potential technologies of the future is exciting, understanding
that there are products currently available that are saving
lives and reducing injuries is paramount.
For the foreseeable future, human drivers are going to be
driving vehicles on our roads and so efforts to prevent crashes
or protect drivers and passengers in a crash are vital. For
example, advances such as the addition of airbags and
electronic stability control to our cars have saved thousands
of lives. As I mentioned at this subcommittee's November
hearing on self-driving cars, we see technologies in today's
marketplace such as automatic braking that have enormous
benefits.
So today I urge all automakers to expedite the deployment
of these braking systems into all new vehicles. According to
the Highway Loss Data Institute it takes 25 years for a new
feature to be on 95 percent of cars on our roads. Therefore,
when we see something that works we need to get it on vehicles
quickly and it needs to be made standard on all models and
makes, not just the most expensive ones.
Witnesses today will discuss other advances such as in
lighting and blind spot detection that have promise, and I hope
these technologies can help prevent injuries and fatalities.
And as with automatic braking, I encourage rapid deployment of
any new features that are proven to be beneficial. I also look
forward to hearing about research into pedestrian and bicycle
rider safety. As we learned at last week's hearing on smart
communities, the number of people living in urban areas is
rising and those areas have unique transportation challenges.
I am also interested in hearing what new technologies can
reduce injuries to rear seat passengers. While injuries to
drivers are still the most common, often our most vulnerable
passengers are in the back. Unfortunately, data on back seat
passengers is still limited which hampers efforts to determine
the effectiveness of features intended to protect them.
Therefore, I encourage NHTSA and all other stakeholders to
collect and share all relevant data on road safety. We need to
be able to see transit opportunities for safety improvements
for people riding in the back seats as well as drivers, front
seat passengers, and others on the road. More information will
also encourage innovation of new safety technologies.
And finally, I will close by continuing my push for
security by design and privacy by design where security and
privacy are not afterthoughts but built into the products from
day 1. I don't think anybody else wants my time, so I will
yield back, Mr. Chairman.
[The prepared statement of Mr. Pallone follows:]
Prepared statement of Hon. Frank Pallone, Jr.
Today's hearing gives us our first true opportunity to talk
about what is happening now in automotive technology. While
learning about the potential technologies of the future is
exciting, understanding that there are products currently
available that are saving lives and reducing injuries is
paramount.
For the foreseeable future, human drivers are going to be
driving vehicles on our roads, and so efforts to prevent
crashes or protect drivers and passengers in a crash are vital.
For example, advances such as the addition of airbags and
electronic stability control to our cars have saved thousands
of lives.
As I mentioned at this subcommittee's November hearing on
self-driving cars, we see technologies in today's marketplace,
such as automatic braking, that have enormous benefits. So
today, I urge all automakers to expedite the deployment of
these braking systems into all new vehicles.
According to the Highway Loss Data Institute, it takes 25
years for a new feature to be on 95 percent of cars on our
roads. Therefore, when we see something that works, we need to
get it on vehicles quickly and it needs to be made standard on
all makes and models, not just the most expensive ones.
Witnesses today will discuss other advances such as in
lighting and blind-spot detection that have promise. I hope
these technologies can help prevent injuries and fatalities.
And as with automatic braking, I encourage rapid deployment of
any new features that are proven to be beneficial.
I also look forward to hearing about research into
pedestrian and bicycle rider safety. As we learned at last
week's hearing on smart communities, the number of people
living in urban areas is rising, and those areas have unique
transportation challenges. I am also interested in hearing what
new technologies can reduce injuries to rear-seat passengers.
While injuries to drivers are still the most common, often our
most vulnerable passengers are in the back.
Unfortunately, data on back-seat passengers is still
limited, which hampers efforts to determine the effectiveness
of features intended to protect them. Therefore, I encourage
NHTSA, and all other stakeholders, to collect and share all
relevant data on road safety. We need to be able to see trends
and opportunities for safety improvements, for people riding in
the back seats as well as drivers, front seat passengers, and
others on the road. More information will also encourage
innovation of new safety technologies.
Finally, I will close by continuing my push for ``security
by design'' and ``privacy by design,'' where security and
privacy are not afterthoughts but built into the products from
day one.
Mr. Latta. Thank you. The gentleman yields back, and that
will conclude our opening statements from our members. The
Chair would like to remind Members that, pursuant to committee
rules, all Members' opening statements will be made part of the
record.
At this time I also want to again thank our witnesses for
being with us today. We really appreciate their taking the time
to testify before the subcommittee. Today's witnesses will have
the opportunity to give opening statements followed by a round
of questions from our members.
Our witness panel for today's hearing will include Mr. Jeff
Klei, president of Continental Automotive Systems North America
at Continental AG; Mr. Bill Gouse, director of Federal Programs
at SAE International; Mr. David Zuby, executive vice president
and chief research officer at Insurance Institute for Highway
Safety; and Dr. Kay Stepper, vice president for Automated
Driving and Driver Assistance Systems at Robert Bosch.
We appreciate you all being here with us today and I would
like to just mention that we have another subcommittee so we
have members coming and out from both subcommittees today. But
we look forward to your opening statements and, Mr. Klei, you
are recognized for 5 minutes.
STATEMENTS OF JEFF KLEI, PRESIDENT, NORTH AMERICA AUTOMOTIVE
DIVISIONS, CONTINENTAL AG; S. WILLIAM GOUSE, DIRECTOR, FEDERAL
PROGRAMS DEVELOPMENT, SAE INTERNATIONAL; DAVID S. ZUBY,
EXECUTIVE VICE PRESIDENT AND CHIEF RESEARCH OFFICER, INSURANCE
INSTITUTE FOR HIGHWAY SAFETY; AND KAY STEPPER, PH.D., VICE
PRESIDENT FOR AUTOMATED DRIVING AND DRIVER ASSISTANCE SYSTEMS,
ROBERT BOSCH LLC
STATEMENT OF JEFF KLEI
Mr. Klei. Thank you very much and good morning, Chairman
Latta, Ranking Member Schakowsky, and members of the
Subcommittee on Digital Commerce and Consumer Protection. I
thank the committee for the opportunity to testify today on
behalf of Continental. My name is Jeff Klei and I am the
president of Continental Automotive Systems in North America.
Continental is a leading tier 1 supplier to develop safe,
sustainable, and affordable mobility technology and solutions
for our customers. In 2016, we generated more than $43 billion
in sales within our automotive tire and specialty rubber
groups. Continental employs more than 20,000 employees in the
U.S. in more than 80 facilities located in 26 States and has
more than 220,000 employees in 55 countries worldwide.
In 2015, there were more than 35,000 lives lost in the U.S.
due to traffic crashes. Projections for 2016 are the dismal
increase to more than 40,000 fatalities, a level we haven't
seen in a decade. More troubling is that on a global scale,
roughly 1.2 million people die in roadway crashes and another
50 million are injured each year. This is unacceptable and
changing this is what motivates each and every employee at
Continental.
In the last 45 years, the U.S. has experienced a relatively
declining trend in traffic fatalities due in large part to
vehicle safety technology like seatbelts in the '70s, the
introduction of anti-lock brake systems and airbags in the
'80s, and finally electronic stability control in the '90s. As
the auto industry moves towards more widespread implementation
of advanced driver assistance systems, Continental projects
these technologies will once again reverse the recent increase
in fatalities.
Continental and our dedicated employees are committed to
developing safe and dynamic driving technologies that
contribute to what we call our Vision Zero, a future with zero
traffic fatalities, zero injuries, and ultimately zero
accidents. Such a future can only be achieved with the help of
innovative active and passive safety, advanced driver
assistance systems, and automated driving technologies.
With building block technologies like automatic emergency
braking, adaptive cruise control, and rear backup assist that
are available in vehicles today, we believe we can continue to
pursue our Vision Zero and achieve higher levels of automated
driving. When we ultimately achieve fully automated driving we
believe that we can reduce the number of fatalities by more
than 90 percent, the percentage of accidents caused by human
error.
The world and the behavior of drivers within it are ever-
changing and the vehicles must adapt to these changing trends.
Our children seem to rely more on smart phones to stay
connected with one another and living in a world of
distractions has been commonplace. Automotive technology must
develop accordingly.
That is why Continental has put a great deal of effort into
human-machine interface technology. We want the driver to be
aware of their surroundings, be aware of what systems in the
vehicle are doing, and be aware of when it is safe to
relinquish control of the vehicle and when it is necessary to
re-engage with the vehicle. In addition, we are heavily focused
on securing the systems of the vehicle with cybersecurity
enhancements as well as the redundancy of safety systems.
Since 2011, we have continued a pursuit of developing and
testing highly automated driving with next generation
technologies like automated parking, Cruising Chauffeur, and a
complete self-driving vehicle in combination with V2X
technology. We were the first supplier in the U.S. to be
awarded a testing license in the State of Nevada for automated
vehicles and are currently testing our third generation
automated vehicle on highways and roads throughout the country
and around the world.
But our continued efforts in this direction would benefit
greatly from an investment in infrastructure that promotes
vehicle to X communication, a dedicated spectrum communication
band that can be utilized by current and future safety systems,
and harmonization of safety laws that allows for the full real
world testing of these technologies. The safe commercial
deployment of potential lifesaving technology depends on the
ability to extensively test on public roads under all
conditions.
Finally, we need an update of Federal motor vehicle safety
standards to accommodate automated driving technology in a
legal framework that supports a new system of mobility. The
world of mobility has the capability of expanding to
unimaginable independence and personal freedom while enhancing
the safety of future generations. Continental stands at the
ready alongside our industry colleagues to work with the
committee and Congress in helping construct laws and
regulations that foster innovation, enable mobility, and create
a safer environment for our public.
Thank you again, Chairman Latta, Ranking Member Schakowsky,
members of the Subcommittee on Digital Commerce and Consumer
Protection, and staff for the opportunity to testify at today's
hearing.
[The prepared statement of Mr. Klei follows:]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
Mr. Latta. Well, thank you very much for your testimony
today and the Chair recognizes Mr. Gouse for 5 minutes. Thanks
again for being here.
STATEMENT OF S. WILLIAM GOUSE
Mr. Gouse. Thank you, Chairman Latta, Ranking Member
Schakowsky, and distinguished members of the subcommittee. SAE
International thanks you for the opportunity to participate in
this hearing, Self-Driving Cars: Levels of Automation. SAE
International is a global society founded in 1905 with more
than 140,000 engineers, scientists, related technical experts,
and students in over a hundred countries in the aerospace,
automotive, motorcycle, commercial, construction, agricultural,
and specialized vehicle industries.
Some notable members were aviation and automotive pioneers
Orville Wright, Henry Ford, Amelia Earhart and Ransom Olds,
motor sports legends such as Andy Granatelli and Dan Gurney,
along with celebrities like Jay Leno. SAE members from
Government, academia, and industry have testified at this
subcommittee or at previous hearings in both chambers. All four
of us on this panel today testifying are SAE members, as I see
are many of my colleagues in the audience.
My SAE experience began even before I was a freshman
mechanical engineering student at Georgia Tech when my
professor and SAE Student Chapter advisor Professor Williams
signed me up as a student member and gave me this membership
pin. My initial exposure to SAE was before college because my
father was or actually still is an SAE member.
SAE's core competencies are voluntary consensus standards
development with nearly 30,000 experts across the globe
contributing to a continually growing standards portfolio of
over 10,000 active and 25,000 historical standards. These are
used to increase safety, performance, quality and productivity
of personal commercial transportation services while optimizing
cost of products and product life cycles. This is an important
point as this standard I will discuss in more detail in a
moment is a product, as all standards are, of our members and
other volunteers' efforts.
In addition to the standards activities SAE holds dozen of
conferences and symposia, including the Government/industry
meeting held in January in conjunction with the Washington Auto
Show Mobility Talks, and next week is the SAE World Congress in
Detroit where my colleagues are also presenting and
participating. These events plus other mutually beneficial
Government/industry academic networking opportunities provide
information for the formation of sound public policy positions
and affiliated programs, products, and services that add value
and encourage innovation.
SAE standards are referenced in Government regulations,
procurement documents, recommendations, and guidelines issued
by the U.S. DOT, the U.S. EPA, Department of Energy, the NTSB,
in regulations in our States, Commonwealths, inhabited
territories, and local jurisdictions. In addition, SAE
standards are used internationally, Canada, elsewhere in the
Americas, overseas, and by the UNECE.
SAE believes that incorporating voluntary consensus
standards by reference as directed in the National Technology
Transfer Advancement Act and the Office of Management and
Budget Circular-A119 improves the efficiency and effectiveness
of Government, whether a Federal, State, municipal body, or
global harmonization activity, it saves time and money while
increasing the efficacy of policy, legislation, and/or
regulation. This is critical in order to respond to the policy
or regulatory needs brought about by the rapid technology
developments we are witnessing.
These developments are progressing significantly faster,
potentially orders of magnitude faster, than the regulatory
process. In addition, the competitiveness of products and
services increased in the global marketplace because of the
higher quality, value, and customer confidence achieved through
conformity with SAE standards. SAE has several standards
published and many documents in development by a variety of
car, motorcycle, pedestrian, and truck and bus committees
relating to increasing the safety and efficiency of transport.
While work continues to improve passive safety and
crashworthiness of vehicles, the potential of implementing
technological solutions to avoid or reduce the severity of
crashes is a major focus of our SAE committee activities.
Details of these efforts, standards and documents, and progress
were submitted to the subcommittee in written testimony. In
summary, they encompass active safety systems, driver
assistance systems, cybersecurity, vehicle connectivity and
communications, measurement and test devices, vehicle testing
including safe on-road testing of automated driving systems,
and specific to today's hearing, title SAE International
Standard J3016: Taxonomy and Definitions for Terms Related to
Automated Driving. I believe there is a flyer in front of all
of you of this standard.
This recommended practice originally published in 2014 and
revised last September and referenced in the Federal Automated
Vehicles Policy provides stakeholders including Federal, State,
and local/municipal regulators, policy makers with a taxonomy
describing the full range of six levels, SAE 0 through 5, of
driving automation in on-road motor vehicles. These six levels
span from no automation to full automation.
I want to point out the key distinction. You see a dark
green break in the handout here is between Level 2 where the
human driver performs part of the dynamic driving task and
Level 3 where the automated driving system performs the entire
dynamic driving task under various conditions. The document,
J3016, also contains functional definitions for advanced levels
of driving automation and over a dozen related terms and
definitions.
Additional terms and definitions of active safety systems
are contained in another standard, J3063 that was published in
November of 2015. Importantly, what these standards do not
provide are specifications or otherwise imposed requirements on
driving automation systems or active safety systems, nor does
it imply any particular order of market introduction or
adoption. One vehicle might have multiple driving automation
features such that it could operate at or different levels
depending upon the features that are engaged or other
consideration.
Standardizing levels of driving automation and supporting
terms serve several purposes particularly clarifying the role
of the human driver, if any, during driving automation system
engagement; providing a useful framework for driving automation
specifications and technical requirements; providing clarity,
consistency, and stability in communications on the topic of
driving automation, as well as a useful shorthand that saves
considerable effort and time. The document is designed to be
useful to many beyond the engineering community, such as
legislators, regulators, others in the legal profession, the
general and trade media, and consumers and the public that are
buying, riding in, or having freight delivered in a vehicle
with some level of driver assistance or automation.
The levels I will go through very briefly are 0, with no
automation; 1, a driver assistance system to a specific mode
such as keeping steering or accelerating/decelerating; Level 2,
partial automation, one or more driver assistance systems, both
steering and acceleration/deceleration using information about
the driving environment. The human driver is still expected to
perform all remaining aspects.
That break down to automated driving systems that monitor
the driving environment for Level 3 conditional automation,
driving mode-specific performed by an automated driving system
in all aspects of the dynamic driving task which define the
standard, with the expectation that the human driver will
respond appropriately with a request to intervene; 4, high
automation, the driving mode-specific performance by an
automated driving system of all aspects of the driving task
even if a human driver does not respond appropriately to a
request to intervene; and 5, full automation, full-time
performance by an automated driving system of all aspects of
the dynamic driving task under all roadway and environmental
conditions that can be managed by a human driver.
SAE has been and will continue to work with organizations
and entities to reference SAE standards as we learn of their
policy, regulatory, and legislative activities regarding both
the public on-road testing, and the deployment of vehicles with
driver assistance and automation systems. We are members of the
Commonwealth of Pennsylvania----
Mr. Latta. Pardon me, Mr. Gouse, if you could just wrap up,
please.
Mr. Gouse. All right. We are members of the Pennsylvania
Department of Transportation Task Force; we work with the
associated motor vehicle manufacturers and other groups. SAE
levels of automation were adopted in the Declaration of
Amsterdam and they are used as we spoke earlier of the U.S. DOT
and the Federal Automated Vehicles Policy. Prior to this, the
Government used separate terms and retired their classification
so now we have this consistent usage.
Driving assistance and automated driving systems have the
potential to provide substantial benefits to all customers of
road transport. And I thank you very much for this opportunity
to provide this statement and answer any questions.
[The prepared statement of Mr. Gouse follows:]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
Mr. Latta. Well, thank you very much. And Mr. Zuby you are
recognized for 5 minutes and thank you very much for being
here.
STATEMENT OF DAVID S. ZUBY
Mr. Zuby. Good morning, Chairman Latta, Ranking Member
Schakowsky, and distinguished members of the subcommittee. On
behalf of the Insurance Institute for Highway Safety, thank you
for the opportunity to testify today on vehicle automation and
crash avoidance technologies.
The Insurance Institute for Highway Safety and its sister
organization, the Highway Loss Data Institute, are nonprofit
research institutes that identify ways to reduce deaths,
injuries, and property damage on our highways. We are wholly
supported by voluntary contributions from companies that sell
automobile insurance in the United States and Canada.
The United States has made enormous progress in reducing
the toll from motor vehicle crashes. The death rate per billion
vehicle miles traveled is one quarter of what it was in 1973
when crash deaths peaked at 54,589. While changes in traffic
laws and their enforcement combined with changes in road and
vehicle designs all contributed to that decline, our research
has shown that improvements in vehicle safety have been the
largest contributor to road safety since the 1990s. We are
convinced that further improvement in vehicle safety will
remain an important strategy to make travel on U.S. roads even
more safe in the future.
Past improvements in vehicle safety largely focused on
mitigating and preventing injuries when crashes occurred. The
newest tool in the vehicle safety toolbox is automation of the
vehicle controls that can prevent crashes in the first place
and reduce the severity of those that aren't prevented.
Electronic stability control which helps prevent sideways
skidding and loss of control, reduces the risk of a fatal
single vehicle crash by 49 percent and cuts the risk of a fatal
multiple vehicle crash by 20 percent.
More recently, front crash prevention systems which help
drivers avoid front to rear crashes with warnings or automatic
braking reduce these crashes by 26 percent for warnings by
itself and by 50 percent for warnings combined with auto
braking. Reductions for crashes with injuries are even larger.
These are large reductions and count as wins for automation
of vehicle control, but neither ESC nor front crash prevention
systems prevent all the crashes they target. In addition, there
are other new crash avoidance technologies like those that aim
to prevent crashes precipitated by inadvertent lane drifts for
which we have not yet found definitive benefits. There are
reasons to be skeptical of the claims that driving automation
will eliminate all crashes currently caused by human error.
This is especially true in the near term technologies which
will continue to involve human driver to a large extent.
The design of these technologies and how drivers interact
with them will be an important factor in their success. For
example, we have found that on average across multiple
implementations from various automakers, lane departure warning
and other lane maintenance systems are used by only 50 percent
of drivers whose cars have them. There is a wide variation in
the use rate and that seems to be influenced by system design.
As technology allows further automation of the driving
task, we are concerned that some human drivers will fail to
understand the limitations of these systems on their vehicles
and crash because they are overly reliant on them. The design
of driving automation systems will be key to helping drivers
understand how systems work including the limitations of the
technology. It will be important to continually monitor the
effects of safety on new technologies entering the market.
The studies mentioned above were only possible with close
cooperation of a few automakers who helped us identify by
vehicle identification number the specific vehicles that were
equipped with a range of optional features. Unfortunately,
there was no comprehensive database linking VINs to information
about what features are present on a given vehicle. Government
policies aimed at ensuring the availability of such highway
safety data are important to enhance highway safety research on
the effectiveness of these emerging technologies.
Thank you again to the members of the subcommittee for
inviting me to share what IIHS and HLDI have learned about the
effectiveness of crash avoidance technologies. I would be happy
to answer any questions.
[The prepared statement of Mr. Zuby follows:]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
Mr. Latta. Well, thank you very much. And Dr. Stepper, you
are recognized for 5 minutes for your opening remarks. Thank
you very much for being here.
STATEMENT OF KAY STEPPER
Dr. Stepper. Thank you Chairman Latta, Ranking Member
Schakowsky, and members of the committee for the opportunity to
testify before you today. My name is Kay Stepper, vice
president with responsibility for the Driver Assistance and
Automated Driving Systems for Bosch in the United States. At
Bosch we are proud to be inventive for life, and I am honored
to discuss an issue that is one of the pillars of our everyday
work at Bosch: to save lives.
Bosch has a long history in the United States. Robert Bosch
himself established the first office in the United States in
New York City in 1906. Now in 2017, Bosch companies operate
more than 100 sites across the country. Bosch believes that
automated driving is the future of mobility, and leading the
way to safe, agile, and automated driving is our guiding
principle. Worldwide, Bosch has more than 2,500 engineers and
researchers working on the topics of automated driving and
advanced driver assistance in our autonomous driving tests that
is conducted in the United States, Germany, Japan, and
Australia.
Preliminary 2016 data from the National Safety Council
projects that as many as 40,000 people died in motor vehicle
crashes last year. The magnitude of the safety crisis is such
that we must seek active means to increase deployment of
technologies that can support drivers and reduce accidents and
injury rates. Driver assistance systems such as automatic
emergency braking and blind spot detection can assist in
reducing the rising fatality and injury numbers that we are
facing in the United States today.
In the near term, it is critical that Government and
industry continue to work together to help increase consumer
access to and understanding of these advanced technologies. I
commend the committee for calling this hearing and for focusing
its attention on two topics that lie at the heart of this
transformation in vehicle mobility: the levels of automation
and the importance of the deployment of driver assistance
systems as a foundation for automated driving.
Unfortunately, these topics are often overlooked in the
overall dialogue about automated driving. The truth is that
many drivers and passengers are already experiencing the
benefits of vehicle automation every single day. The active
safety system electronic stability control is integrated into
every new light-duty vehicle sold in the United States today.
This revolutionary technology invented by Bosch engineer Dr.
Anton van Zanten has saved thousands of lives. A 2014 report
from NHTSA found that ESC saved close to 4,000 lives during the
5-year period from 2008 to 2012.
Automated driving will bring great benefits and pave the
paths forward a new vision of personal and collective
transportation. However, it will take time to achieve fully
automated driving and it will be an evolutionary process,
building up on the stepping stones of active safety, driver
assistance, and crash avoidance system.
In discussing the evolution toward automated driving I want
to emphasis that Bosch strongly supports NHTSA's decision to
adopt the SAE J3016 framework for levels of automation as part
of the Federal Automated Vehicle Policy. This is a major step
toward harmonizing and establishing a common set of definitions
across the various stakeholders involved in these efforts.
Bosch wishes to highlight automatic emergency braking as one
clear example of how drivers are being introduced to automation
in a gradual manner, and also of how automation intervention by
the vehicle can provide the greatest benefit in terms of
accident reduction.
Suppliers play an important role in the innovation cycle
and many suppliers such as Bosch conduct extensive testing in
the lab on test tracks and on public roads. Suppliers presently
face several obstacles in carrying out this testing on public
roads, and we respectfully request that the committee consider
extending the FAST Act exemption to include suppliers with
active and established research and development programs in the
United States.
Bosch position on the need for improved consumer education
is well known. We have urged NHTSA and the U.S. Department of
Transportation for many years to include crash avoidance system
as a key component of the vehicle 5-Star rating and to provide
additional information to consumers through the Monroney label.
Bosch believes that displaying crash avoidance systems as part
of the official safety portion of the Monroney label and
particular in the form of 5-Star rating, as the most effective
means to help driver consumer awareness and eventually consumer
demand for such technologies. Without the clear presence of
crash avoidance and mitigation technologies on the most
recognizable feature for consumers, the physical Monroney label
as affixed to the vehicle, consumer education will continue to
lag.
The adaption of crash avoidance technologies into NCAP
would be a very significant improvement and one which we
believe will bring about immediate benefits as well as paving
the path toward the attainment of automated driving in the
future. Bosch encourages Congress and NHTSA to cooperate a path
forward for the U.S. NCAP to become an effective means of
encouraging the enhanced adoption of these lifesaving systems.
Bosch truly believes that a 5-Star rating is the most effective
means to translate the presence and performance of crash
avoidance technologies into an easy-to-understand indicator for
consumers.
Thank you again for the opportunity to speak before the
committee. I welcome any questions you may have.
[The prepared statement of Dr. Stepper follows:]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
Mr. Latta. Well, thank you very much for your testimony and
that will conclude our opening statements from our witnesses.
Again we appreciate you being here, and I will begin the
questions if I may.
And if I could, Mr. Zuby, I would like to just follow up
what you said what you said. A lot of the drivers out there
driving the vehicles that have a lot of this technology are not
using it. Is it because, you know, is it too difficult for them
to understand maybe from reading the instructions in the manual
or they just don't want to bother with doing it, or what are
you finding out there why people aren't using that technology?
Mr. Zuby. Right. So we think that one of the reasons that
people aren't using lane departure warning technology is
because they find it annoying. The way that technology works
today is that it basically gives you a warning which may be an
audible beeping or a vibrating of the steering wheel or
vibrating of the seat when you transgress a lane line without
signaling your intention to do so.
So one way to think of the current technology is it is sort
of a turn signal nanny rather than warning the driver about an
imminent danger. And when we interview, or rather survey
drivers with the technology that is one of the things that they
tell us is the lane departure warning is very annoying. Systems
that interact with the driver less frequently like front crash
prevention are much more likely to be left turned on. In the
studies that we have done we find that AEB and front crash
warnings are left on in 90 percent or more of the vehicles,
whereas we only see about 50 percent of lane departure systems
left on.
The other thing that our research is finding is that the
design of the lane departure warning seems to have an
influence. So people don't like the audible alerts, but when
the system alerts them about crossing the lane line with a
vibrating steering wheel or a vibrating seat they are much more
likely to leave it on. And we also find that if the car takes
some steering action in response to, you know, transgressing
the line that too leads to higher use rates than the original
systems which only warned the driver with an audible warning.
Mr. Latta. Well, thank you very much.
Mr. Klei, if I could ask you a little bit about especially
on the cyber side, in your testimony you mentioned how driver
assistance systems will require sensors to gather data about a
vehicle's surrounding environment in order to adequately assist
that driver. How is Continental thinking about the privacy and
security of the advanced driver assistance systems and crash
avoidance systems, and what is Continental doing to secure
those systems against cyber threats?
Mr. Klei. Thanks for the question, Chairman Latta, and it
is a great question and it is something that at Continental we
have been thinking about for many years. Cybersecurity is not
new with automated driving or the advanced driver assistance
systems. It has been a discussion point and a key development
area for us for many, many years ever since, really,
electronics started to come into the car.
I would say the connection to the cloud, the connection
with all the 4G connections that are now available open up a
new opportunity for those cybersecurity threats. We have
developed an entire competency center in our company that is
used extensively for cybersecurity and we are trying to install
all the different protections that we can from known
cybersecurity attacks.
But many people say should we have a cybersecurity
specification it is dynamic. Every day there is new threats.
Every day there is new opportunities that emerge. So we have to
work together with our OEM partners, suppliers, and the
Government to look at ways we can work together to identify and
eliminate those cybersecurity attacks. But we clearly have a
competency center, we think very much about it, and it is
clearly a challenge as we bring many of these technologies into
market. But it is not new. It has been thought about and
developed for many, many years.
Mr. Latta. Well, if I could also, Dr. Stepper, would you
like to comment on that on what Bosch is doing in this area on
the cyber side?
Dr. Stepper. Yes. Thank you, Chairman, for the question.
Bosch has been very active on this topic for cybersecurity
protection. We believe very much in a layered approach, layered
in a sense that there is hardware layer, software layers, and
architectural layers that need to be introduced. We actually
established a center of competency for cybersecurity back in
2010, and we already established additional units within Bosch
that work specifically on software solution to help our OEM
partners to protect against cybersecurity threats.
Mr. Latta. Well, thank you very much.
And also, Dr. Klei, could I ask a real quick question
because my time is running out here, commenting on SAE levels
of automation and why they are important to the industry
standard of fully self-driving cars.
Mr. Klei. Certainly we very much support the adoption of
the SAE standards. We think a standard that clearly defines
what the levels of automation are, are very useful as we start
to develop and deploy these technologies. The consumers are
often confused by the various naming and the various levels.
And I think we as an industry have a lot of work to do to
improve that communication and education of the consumers.
Suppliers have a role in this. The OEMs have probably the
largest role because they are the ultimate touch point with
consumers. And then of course any assistance from the
Government and other outside agencies are very, very
beneficial. So we very much support it and we think everyone
has a role in educating so that the naming of these
technologies really describe what it can do and people don't
get confused.
Mr. Latta. Well, thank you very much. And my time is
expired and I will now recognize the gentlelady from Illinois,
the ranking member of the subcommittee, for 5 minutes.
Ms. Schakowsky. Thank you, Mr. Chairman.
Dr. Stepper, your testimony mentions rear automatic
emergency braking systems and I am wondering if you could
discuss how that could help prevent backover accidents.
Dr. Stepper. Yes. Thank you, Ranking Member Schakowsky, for
the question. The rear automated emergency braking is a
relatively recent addition to the automatic emergency brake
suite of functions that we have. We already have a mandate in
the United States starting in 2018 for backover legislation to
have a rearview camera installed in each and every vehicle.
So we have already a basis of the technology in there, and
we also see that especially with pedestrian incidents that we
see in rear backover situations this technology could really
help not only to protect from material damage but saves lives
and prevent injuries.
Ms. Schakowsky. And is this feature available today in any
makes or models?
Dr. Stepper. It is available today but still in very, very
small numbers. There are a few select vehicles in the United
States today sold with this. The installation rate overall is
less than five percent, in contrast to forward-looking
automatic emergency braking where you look more between a 20 to
25 percent installation rate today already.
Ms. Schakowsky. You also mentioned pedestrian automatic
emergency braking. Is that any different from AEB when another
car is in front of the vehicle?
Dr. Stepper. It is another progression and another step in
the full AEB suite. The automatic emergency braking for
vehicles was invented first and brought to market. Pedestrian
automatic emergency braking has a little bit of a different
requirement in the sense that you need to have a very wide
field of view to recognize crossing pedestrians and not only at
higher speeds, but especially in urban scenarios at lower
speeds. So and therefore it is different in the sense that the
requirements on the technology are different and it is already
part of Euro NCAP in the European Union as a requirement moving
forward.
Ms. Schakowsky. Thank you. Mr. Zuby, I wonder if you have
looked into these technologies and if you have any comments on
that.
Mr. Zuby. Yes. We have been looking into these technologies
and we have worked up a series of tests that we intend to start
using to promote the idea of reversing automatic braking. We
think that that may be an additional thing that is needed to
address backover crashes because the experiments that have been
run using cameras show that while they definitely improve the
situation and help drivers avoid running into things that are
behind their vehicle that they don't expect to be behind their
vehicle, they are not a hundred percent effective because the
driver needs to be looking at the camera at the same time that
the person or object behind them is in the view of the camera.
So automatic braking, I think, can augment the benefits
that we get from the technology looking rearward in the camera
during reversing maneuvers. We are also looking at pedestrian--
by the way, my guys have identified, I think, 14 models of cars
sold in the current model year that are equipped with reversing
AEB. We are also looking at pedestrian detection. And it is a
slightly more difficult problem for the technology to solve
because of the field-of-view issue and the fact that
pedestrians can change direction and change their movement very
quickly.
Ms. Schakowsky. Because I have been so involved in the
issue of the cameras and you say it is not a hundred percent,
have you estimated how effective it is or how many times it
does fail to prevent an accident?
Mr. Zuby. Well, so in experiments we find that it reduces
the likelihood that you are going to back over something that
is in your path by about two thirds.
Ms. Schakowsky. OK. So you have done years of research on
AEB systems. Can you give us more details on how these systems
work and why they save lives?
Mr. Zuby. So the current AEB systems mainly prevent front-
to-rear crashes. They are effective at preventing those kinds
of crashes, and even when they don't prevent the crash they
reduce the risk of injury. Front-to-rear crashes don't result
in a lot of fatalities. It is in the neighborhood of about 800,
900 people a year out of the nearly 40,000 die in front-to-rear
crashes. So even if a technology were to prevent all of the
rear crashes, it would have a small dent on fatalities.
But the sensors that are needed for AEB are sensors that
will be needed to address other types of crashes, you know,
leverage the technology to address other kinds of crashes that
do account for more fatalities.
Ms. Schakowsky. Thank you. I see I am out of time, I yield
back. Thank you.
Mr. Latta. Well, thank you very much. The gentlelady yields
back, and the Chair now recognizes the gentleman from Illinois
for 5 minutes.
Mr. Kinzinger. Thank you, Mr. Chairman. Thank you all for
being here and taking some time with us today. It is an
important hearing on the future of self-driving cars and
specifically the opportunity to learn more about the advanced
driver assistance systems that is saving lives today and it is
also paving the way to fully autonomous vehicles.
Dr. Stepper, in your testimony you highlighted the
importance of the SAE framework for the various stakeholders in
autonomous vehicles and the lack of common language for
advanced driver assistance systems. How has this lack of a
voluntary standard impacted Bosch's ability to bring technology
to the market?
Dr. Stepper. Thank you for the question, Congressman. Very
clearly, the lack of clear language and common taxonomy has
resulted in some confusion at the consumer side: What is really
my car doing with the different technologies that we have? So
as Mr. Gouse has very graphically illustrated in his chart,
there is well defined levels of 0 to 5 for automation, and
coupled with a very active consumer education campaign we can
really educate consumers what they can expect.
Is it just a warning that my vehicle will provide or is it
actually an actual intervention like an active braking
situation or can I take my hands and my feet off the controls
and the car will drive by itself? And what we have found
clearly is that the lack of such common language really has led
to confusion on the consumer end, and we really commend the
National Safety Council together with the University of Iowa
joining the Road to Zero campaign and actually establishing a
Web site that is called mycardoeswhat.org to educate consumers
of what is actually in their vehicles today because it can be
so confusing.
Mr. Kinzinger. We should do a my-congressman-does-what. Mr.
Gouse, what are the challenges to adopting a voluntary
consensus standard and what efforts are underway to provide a
common language for advanced driver assistance systems?
Mr. Gouse. Thank you, Congressman, for the question. It is
an emotional question internally, because it is very difficult
to raise awareness that our documents even exist to a variety
of stakeholders that don't traditionally know that they even
use this. We were working with the American Association of
Motor Vehicle Administrators and they didn't even know that the
license plate geometry was our standard. So that was our
beginning point. And we told them we had this document in works
at the same time NHTSA had their levels of automation in works,
and with differing vocabulary and differing levels it confused
the issue a lot. Fortunately, NHTSA decided to adopt the SAE
language, and then AAMVA and through the States that
proliferated. That is one example.
The same thing is happening all over the world. For the
driver assistance systems, same thing, we have a standard that
is called Active Safety Systems Terms and Definitions. It is a
fairly easy read. It is not really riveting like a novel, but
it is a fairly easy read and we are trying to get that language
adopted too. And as you hear today, we even use different terms
ourselves and I agree it is confusing.
Mr. Kinzinger. Let me add on. Are there any policies,
developing policies that you are concerned with as you are
seeing them right now?
Mr. Gouse. The States that are unaware or choosing not to
use a common terminology and the common taxonomy, I believe,
will result in a patchwork of very difficult to understand and
operate in environments. This is happening now at the testing
level where they are passing regulations permitting testing of
various levels of automation in nonsalable vehicles. So it is a
concern.
Mr. Kinzinger. And then we will go with Dr. Stepper on this
one. When you look at educating the public about the benefits
and the limitations of various systems, especially for systems
like automatic emergency braking that provides a lot of value
to the customer, but the customer, the consumer may not be
aware that the technology is assisting the driver. Mr. Zuby
mentioned that lane maintenance systems were only turned on in
51 percent of the vehicles that IHS observed. How do your
companies, how does your company work with the consumers to
build confidence in the technology so it is being fully
utilized?
Dr. Stepper. So thank you for the question, Congressman.
Clearly we work with activities like the Road to Zero and the
activities from the National Safety Council as well as the
University of Iowa. We work very closely with our OEM
customers, for example, in joined co-marketing campaigns to
educate dealers, because at the end of the day new vehicles are
being bought from dealerships and consumers are being consulted
by dealership personnel and that is really your first touch
point of a new vehicle purchase and understanding of what this
vehicle really has on board in terms of technology.
So we work very actively with several OEM customers on this
topic to make tours to make joint marketing campaigns around
the country to educate dealerships on this topic so they can
explain what is installed on the vehicle. Again I want to
emphasize an additional mention of these crash avoidance
technologies. In a 5-Star rating, incorporating crash avoidance
technologies could also very much help in that regard because
now the dealership personnel would have the Monroney label
right in front of them to help them guide the consumer through
the purchase.
Mr. Kinzinger. Thank you. And I have some more questions; I
will submit them for the record. Mr. Chairman, I yield back.
Mr. Latta. Thank you very much. The gentleman yields back,
and the Chair now recognizes the gentlelady from Michigan for 5
minutes.
Mrs. Dingell. Thank you for the recognition, Mr. Chairman,
and for your continued interest in the automated vehicles. As
you all know it is a subject that I really care a great deal
about. As I stated in the last hearing on this issue, I believe
it is critical that the Congress, the administration, the
industry, and safety advocates all come together on a common
framework for automated vehicles. Too much is at stake and we
have got to get it right.
Legislation will be needed to facilitate the deployment of
higher level automated vehicles, and I support raising the
statutory exemption caps as an interim solution while directing
NHTSA to amend existing vehicle safety standards as they relate
to human operated controls. And I think a lot of people don't
understand what some of the regulations are because they have
been there for so long.
Great strides in vehicle automation are being made. Proud
of it that a lot of it is in my district in developing safety
technologies that have the potential to reduce roadway deaths,
and I believe helping them get to market could have a
significant impact on public safety, and I have got some
questions to help the committee examine these issues.
My first questions are for Mr. Gouse of SAE, and if you
could just do yes or no, please. Is it correct that SAE Levels
0 to 2 contemplate that a human driver will perform all or some
aspects of what is known as the dynamic driving task?
Mr. Gouse. Yes.
Mrs. Dingell. Is it correct that SAE Level 3 contemplates
that a human driver must be in the loop and prepared to respond
to a request by the vehicle to take over the dynamic driving
task?
Mr. Gouse. Yes.
Mrs. Dingell. Now is it true that an SAE Level 4 vehicle is
one that is capable of performing all aspects of the dynamic
driving task in a given situation also known as the operational
design domain?
Mr. Gouse. Yes.
Mrs. Dingell. And a Level 5 vehicle can handle all aspects
of driving under all conditions?
Mr. Gouse. Yes.
Mrs. Dingell. Thank you. Now these questions are for all
four witnesses. Is it true that companies like FCA, Ford, and
GM in Michigan are developing and currently deploying SAE
Levels 1 and 2 systems? Anyone can say yes or no.
[Chorus of yeses.]
Mrs. Dingell. Thank you. Is it true that these traditional
automakers and others like Waymo are developing Level 4 systems
at the same time?
Mr. Klei. Yes.
Mrs. Dingell. In other words, these companies aren't
necessarily pursuing a sequential progression through the SAE
Levels to full vehicle automation; is that correct?
Mr. Klei. No.
Mrs. Dingell. That is not correct. So you think they are
going 1, 2, 3, 4 or are they going from 2 to 4?
Dr. Stepper. If I may jump on this one, Congresswoman
Dingell, it depends on the automaker. Some absolutely proceed
along the path, Level 0, 1, 2, 3, 4, and 5; some other ones may
skip Level 3. There is no common answer. But some of them that
you mentioned are indeed following exactly along the path of
what Mr. Gouse has presented.
Mrs. Dingell. And others are skipping. Is it true that a
number of existing NHTSA safety standards require human
operation of vehicle controls that may not be necessary if
there is no human driver, such in Level 4 or 5?
Mr. Zuby. Yes.
Mr. Gouse. Yes.
Mrs. Dingell. Do you all have good--I don't know if I am--
my staff wants me to keep moving. But I think people don't know
that a NHTSA requirement requires a foot on a brake and it is
not necessary at 4 or 5, so----
Dr. Stepper. That is correct.
Mrs. Dingell. Thank you. Should NHTSA amend existing safety
standards to clarify how they apply to higher level automated
vehicles without drivers?
[Chorus of yeses.]
Mrs. Dingell. Do all of you agree on that?
Mr. Gouse. Yes.
Mrs. Dingell. Well, I am running out of time, so I am going
to--I have lots of questions but--and, for the record, I may
submit some more, Mr. Chairman. But I want to commend the
chairman for holding this important hearing to help educate
members on the issues because it is really important that we
get it right. Automated vehicles are going to be developed and
they are going to be developed internationally if we don't take
the lead on making sure we do it, develop them here and that
these technologies are developed in the United States of
America. So I look forward to working with my colleagues on
both sides of the aisle in a bipartisan manner to achieve this
goal.
Thank you all for being here today. Thank you, Mr.
Chairman. I yield back my 15 seconds.
Mr. Latta. Thank you very much. The gentlelady yields back
the balance of her time and the Chair now recognizes for 5
minutes the gentleman from Mississippi, the vice chairman of
the subcommittee.
Mr. Harper. Thank you, Mr. Chairman. And again thanks to
each of you. This is, you know, it is just mind boggling the
possibilities and we have just barely scratched the surface.
And, you know, I can't imagine what it will be like we come
back in 5 years and just discuss what we are doing next. I mean
this is really remarkable. So thanks for the involvement that
each of you and each of your companies have.
And Mr. Klei, thank you very much. We are excited about the
presence of the new Continental Tires facility that will be
opening in Mississippi. I think that was a great decision. We
are honored to have a part of your company that will be there,
and I wanted to talk to you for just a minute.
Obviously, the intellectual disabilities issue is
important. My wife and I have a son who is 27 years old who has
Fragile X syndrome. He graduated from a special program at
Mississippi State University. He works Monday through Friday.
My wife has to drive him every day and drop him off and pick
him up. So it is something for many families, this is an
important issue. So are advanced driver assistance systems at a
point where they are able to provide new transportation
opportunities to the disabled community?
Mr. Klei. Certainly is it an important topic, and thank
you for the question, Congressman. It is something that I
think, as an industry we are working very hard, and it is not
just for the automated driving technologies in general. We are
trying to make mobility more available and safer for all, and I
think the advancements in automated driving are clearly going
to move that forward.
Are they ready today to take over all driving tasks for
someone that can't drive today? Not necessarily; over time,
absolutely. We believe when we get to Level 4 and Level 5,
absolutely it is going to provide mobility for many people that
today don't have that mobility. The Waymo development, their
first example that they showed was someone that was blind. And
that is a huge statement for the potential mobility promise for
the elderly, the blind, and every disabled person in the United
States will have mobility, and it is an important step for
them, but also for society.
Mr. Harper. Well, we are excited that Continental is taking
that into consideration in the development of this.
Dr. Stepper, will you also comment on that as well?
Dr. Stepper. Yes. Thank you for the question, Congressman.
We are actually working very, very intensively on the aspect of
human factors because as we have learned before, on some of the
levels of automation the interaction of the human being is
still very, very important and part of the requirement for both
SAE all the way to Level 3 as we heard earlier.
So in human factors we have done a number of research for
user, human-machine interaction perspective, but we have also
worked in augmented reality experiences. And that is a topic I
just want to make the comment that we are actually going to
show a demonstration of augmented reality for automated
driving. It is an upcoming experience here on the Hill as the
event that is CES on the Hill on April 5th, where all of you of
course are invited to experience some of the human factors
aspect and how important it is as part of the automated driving
equation.
Mr. Harper. We are expecting self-driving cars to be at
Level 5 tomorrow, when most drivers are not Level 5 drivers.
Mr. Klei, what do you think Congress should do to facilitate
this development in deployment of advanced driver assistance
systems at a point where we can assist and not be, let's say, a
roadblock to that development?
Mr. Klei. Thank you, Congressman, a very important question
and one that I think we look at a couple different areas. One
is the Federal Automated Vehicles Policy that was issued last
September. While we commend the NHTSA organization and all the
work that they did we think there is a lot more to do.
First of all, when it comes to that policy it really more
talks about deployment rather than development, and we think
development is an important part of bringing these technologies
to market safely and with real world testing.And only through
an improvement in that policy can we get there. For example,
the policy requires for every software change or every change
that we make we have to submit a new exemption. The time to
develop those and the time to get the approvals will
significantly delay the implementation of this.
I think the other thing is the model State policy. To have
a patchwork of State regulations is clearly hindering our
ability to test and develop and ultimately commercially deploy
these technologies. So there is two examples. I could go on and
on about other examples, but clearly there is opportunity to
work closer together between ourselves as suppliers, the OEMs,
and the Government to really bring these forward in a safe and
effective way.
Mr. Harper. Thanks to each of you. I yield back.
Mr. Latta. Well, thank you very much. The gentleman yields
back, and the Chair now recognizes the gentlelady from
California for 5 minutes.
Ms. Matsui. Thank you very much, Mr. Chairman, and thank
you very much for the witnesses for being here today. As many
of you know, the FAST Act mandated that self-driving cars could
be introduced into commerce solely for the purposes of testing,
but only by companies that had at the time of the law's
enactment already manufactured and distributed motor vehicles
in the United States.
In addition, legislation has been proposed in some States
that would allow only traditional car manufacturers to test and
deploy AVs. Some have even speculated that NHTSA's deployment
exemptions also could be limited to car manufacturers that
already build and distribute motor vehicles in the United
States, and I believe we started down this path already.
But Dr. Stepper and Mr. Klei, I know that you have been
working with AV components that could benefit from direct
testing. What are the barriers to your companies doing testing
on your own?
Mr. Klei. From the Continental side certainly we have
talked a little bit about some of those barriers with the
ability to test without concern for all the different State
regulations. I mean, since the Federal Automated Vehicles
Policy came out there has been 48 different bills in 20 States
that complicate our development of these technologies. We
believe that as suppliers we also need to have the ability to
test and develop these. It can't be just the OEMs that in fact
do certify vehicles for FMVSS. We as suppliers don't certify
vehicles. We develop technologies, we work with our OEM
partners to bring them in safely, but we need the ability to
develop and test those ourselves, not as a certifying FMVSS
body but as one that really looks to develop those.
Ms. Matsui. Certainly. Dr. Stepper?
Dr. Stepper. Congresswoman Matsui, thank you for the
question. As I mentioned earlier in my testimony, suppliers
play a very important role in the innovation cycle. And as a
matter of fact, often innovations like electronic stability
control, the required sensors like radars, video cameras,
ultrasonic sensors, and many of the other active systems, for
example, the braking and the steering in the vehicle, is
actually coming from the suppliers.
So we do our utmost of course to develop and test and
verify these components and systems in the lab with artificial
methods like modeling and simulation, but there comes the point
where we suppliers need to take these technologies on the road
to ensure that they are fully verified and validated before
they ever go into consumers' hands. So it is really limiting
our ability to test on public roads.
And we understand very clearly that the expansion of the
exemption must be handled very carefully and cautiously, but we
are very happy to engage actively with the committee on this
point.
Ms. Matsui. Thank you. I understand that different
companies are pursuing different strategies in terms of the
level of automation in the vehicles they plan to deploy. And as
we have been reminded, often it is human drivers that can cause
and contribute to accidents with automated vehicles.
Mr. Zuby, are there particular concerns we should consider
during a transition when vehicles from all different levels of
automation will be on the roads?
Mr. Zuby. Yes. I think we are already seeing in studying
work that Waymo are doing and other automakers that even when
the automated cars are driving at a very high level of
competency they often are involved in crashes caused by human
drivers. And so I think as the testing develop it is important
to make sure that there are safeguards that the testing be done
in safe ways and not endanger other people and the public, but
it will be absolutely necessary to test these things in the
real situation because that is where they need to work.
Ms. Matsui. Right. As companies continue to expand testing
of autonomous vehicles, they are all gathering an enormous
amount of data about these vehicles. Mr. Gouse, are there any
efforts in place to standardize the data that is being
collected so that we can learn best practices regardless of
where the autonomous vehicles is being tested?
Mr. Gouse. Ma'am, there are very early efforts going on.
You have to understand that it is a very proprietary
environment. While these gentlemen are cordial here, they
probably want to kill each other sometime over a product.
Ms. Matsui. I hope not.
Mr. Gouse. No, no, no. So there are discussions going
underway with the associations that they belong to on this and
how to collect the data and use it.
Ms. Matsui. So we are at the very early stages of that
right now but it would be very helpful to have the data. So
anyway I will yield back my remaining time.
Mr. Latta. Thank you very much. The gentlelady yields back,
and the Chair now recognizes the gentleman from West Virginia
for 5 minutes.
Mr. McKinley. Thank you, Mr. Chairman. And last month when
we met I said then that I think this is, this whole process is
probably inevitable. And as one of just two licensed engineers
in Congress, I am intrigued with the problem-solving
possibilities that we have with this. I am fascinated with the
developments that have occurred so far in lane movement as you
referred to it or the braking.
But I am a huge skeptic of driverless cars and I am not
buying this one iota yet. I will go with all the others. I can
see the possibilities of that. But at the last meeting I raised
some questions about IV&V and everyone on the panel had no idea
what we were talking about, so I ask you because you are four
different people. Are you using IV&V for confirmation of the
various steps that we are going through so far?
I am seeing a no all the way around again. If we send a
ship to Mars or when we send a satellite into space we run
through all the steps to test it for individual verification
and validation and make sure that it is going to work because
we don't want to rely on competitive peer pressure without
having some third party validate what we are doing. And that is
what we are looking for, I am going to looking for is third
party, because I know companies are going to be under a lot of
pressure to skip steps 2 and 3 and go right to 4 if possible or
skip 1 and go to 3, whatever that might be they are going to
move that because of competitive pressures.
We talked a little bit when one of the things since that
time--because I am fascinated with this. Again, it is the
engineering. I know this is inevitable. How can we work with
this thing to do everything but driverless? So when I have
asked the question when I have been back in my district, it is
wherever it is we are excited. In fact we are going to have a
summit meeting about this, about driverless cars.
But when I have raised the question, Would you put your 6-
year-old granddaughter in the car and let her go 40 miles to
meet her brother, perhaps, every one of them says no. Now, I
know it is going to be evolutionary. They will develop more
confidence with it. But when I was hearing about if something
goes wrong they are going to transfer operation back over to
the person in the car, what happens if it is indeed someone
that is intellectually impaired or is inebriated and we have
allowed them to get in that car to be able to get home, and
then they are turning the transportation over to them when they
are doing 60 miles an hour, and they say, ``OK, driver, it is
your car''?
I have a series of questions about it. I am going to remain
a skeptic on this. I want to follow the money. I don't
understand other than insurance companies who is really going
to benefit for this, but as an engineer let me skip to my last,
so ask a question of this. If when we get to steps 4 and 5,
because I have designed a lot of bridges, a lot of highways,
culverts, I don't know how this is functioning yet, so is there
something I should be working in in my old company in
engineering that starts to get ready so the cars when we are at
steps 4 and 5 there is something, is there a wire in the road,
is there something along the guardrail, or is this something
merely sensing it? Is this all GPS driven?
I need to have a lot more information before we get
anywhere close to that. Because if we are designing all these
roads, why aren't we taking those things into consideration now
especially with this infrastructure bill that it is going to
have? So with that can you tell me what should we be doing in
our highways to be ready for steps 4 and 5?
Mr. Klei. In terms of the highways themselves we have to
adapt to the highways, we can't expect the highways to adapt to
these systems. That is why real world testing around the world
has to happen.
Mr. McKinley. So in that case, Mr. Klei, is it GPS driven
or is it sensing the side of the highway?
Mr. Klei. It is both. It is GPS, it is sensing.
Mr. McKinley. It goes through a tunnel, and in West
Virginia, where we have almost 50 percent of the State does not
have service, I lose my signal constantly and no one knows
where we are. And I don't know what happens at that point, so
you are going to have to rely on a lot better control if you
are going to use GPS. So if it is going to be sensing how do we
do that?
Mr. Klei. Obviously, the sensory development is a key part
of that. But it is not just sensing it is also GPS. It is also
vehicle-to-vehicle, vehicle-to-infrastructure, DSRC, all of
that coming together will unable that Level 4 and Level 5.
Mr. McKinley. Thank you very much, I have run out of my
time. But I want some engineering answers on this, not the 90
percent savings of accidents, because I think it is BS. It is
not going to happen, just like we have had the debates here
over my 7 years in Congress that, if we stop using coal, we
would eliminate 80 percent of the asthma attacks in this
country. We know that is false. So I don't want to use a
technique or a topic that says we are going to save 90 percent
of accidents if we adopt this, I want to have more facts. The
engineer in me says I need more facts. So thank you, and I
yield back.
Mr. Latta. Thank you very much. The gentleman yields back
and the gentleman from Texas is now recognized for 5 minutes.
Mr. Green. Thank you, Mr. Chairman, for both you and our
ranking member, Ms. Schakowsky, for having the hearing today.
While the technology behind autonomous vehicles continues to
evolve at a rapid pace it is important that industry and
Congress continue to examine safety standards to ensure
consumer safety. Not all the safety innovations are willingly
accepted by the public with the history of airbags and
seatbelts has shown. Continued open discussion on these new
technologies are essential moving forward so that consumers can
be familiar with both benefits and the limits of autonomous
features. Frankly, my wife is probably the most supporter of me
not being in an autonomous vehicle when I am driving. She
complains all the time about my driving.
Mr. Zuby, in your testimony you state that your research
has shown that the driver acceptance of technology varies. Can
you tell us more about the varying level of acceptance of new
technology and what can be done to increase the public's
acceptance?
Mr. Zuby. Yes. For one of the things that we found for lane
departure warning systems, the mode of the warning made a big
difference in whether or not the drivers accepted them. When we
interview drivers what we find is they complain about audible
warnings being annoying. Another important aspect of lane
departure warning and lane maintenance is that the systems
respond to truly dangerous situations and not be perceived by
the driver as simply being a nanny about use of the turn
signal.
So I think the technology needs to go a ways beyond where
it is today in order to sort out what are the real dangerous
situations that we need to inform the driver about versus those
things that might be dangerous, but a lot of drivers aren't
going to perceive them as such.
Mr. Green. OK. At this point, is it known why one warning
system is so effective and another ineffective?
Mr. Zuby. One of the issues is if the warning system can be
heard by other people in the vehicle drivers tend not to like
it. So the vibrating steering wheels, the vibrating seats tend
to have higher levels of acceptance than audible warnings
themselves.
Mr. Green. Thank you. How can we better study the
effectiveness of these safety claims to ensure technology is
living up to its promise?
Mr. Zuby. It is super important I think that we work out
ways to make sure that data about which cars have which systems
and how the systems are working is available to independent
researchers. Obviously, the companies who are developing the
systems are going to want to make claims about their high
levels of effectiveness, but I think people in Government and
independent evaluators need to be able to verify those claims.
Mr. Green. I would like to ask this question of the entire
panel. Would enhanced Government regulation on the collection
of the crash data with specific regard to what autonomous
technologies were in each vehicle improve both public safety
and efficiency, the AV technology? I will start with Mr. Klei.
Mr. Klei. Yes. Certainly when you look at things like the
Auto ISAC, which has been developed as an industry coalition to
really share data on cybersecurity, it is a good example where
data sharing can really benefit. We think there is an
opportunity as well to do something similar for some of the
crash data and some of the activity around autonomous,
automated driving vehicles. We think that the sharing is very
powerful, but it needs to be the edge cases and it needs to be
things that can help all of us develop and deploy these
technologies.
Mr. Green. Mr. Gouse.
Mr. Gouse. In our committees, sir, there is quite a bit of
sharing going on of technical information that is not
proprietary to build the standards to design test
specifications, test devices, and what not to build good
product, so there is a quite a bit ongoing already at that
level.
Mr. Green. OK. Mr. Zuby.
Mr. Zuby. Definitely, I think regulations prescribing what
kind of data needs to be saved and under what kind of
circumstances and with whom that data can be shared will help
all of us achieve a greater level of comfort that the
technology is being developed in a safe way.
Mr. Green. Dr. Stepper.
Dr. Stepper. Definitely a yes, Congressman. Bosch has been
working very adequately to actually get NHTSA more resources
for data for crash reconstruction. Why, because we have used
NHTSA's NASS database for our own research in understanding how
many percent of collisions with injuries and fatalities with
rear-end crashes, how many drivers failed to, for example, even
after they received the warning to even apply the brakes in the
first place. So it is very valuable data for us for our
development purposes.
Mr. Green. OK. Thank you, Mr. Chairman. I yield back.
Mr. Latta. Thank you. The gentleman yields back, and the
Chair now recognizes the gentleman from Florida for 5 minutes.
Mr. Bilirakis. Thank you, Mr. Chairman. I appreciate it.
Dr. Stepper, some driver assistance systems on the market use
audible tones, steering wheel vibrations, and flashing lights
to alert the drivers to impending hazards. We are also facing
high levels of driver distraction as you know. As Bosch works
to develop these technologies how are you working with
automakers to ensure that these technologies aren't pulling
drivers' attention away from the task of driving and causing
more distraction?
Dr. Stepper. Thank you, Congressman, for the question.
Mr. Bilirakis. Sure.
Dr. Stepper. We work very intensively with our OEM partners
on the human factors element. For example, evaluating what is a
really effective and efficient means of alerting the driver of
getting the attention from the driver back? Is it audible, is
it visual, is it maybe haptic?
As Mr. Zuby has answered before what we have found is that
haptic feedback is actually very, very efficient when it is
related to a specific action that is wanted. For example, if
there is a hazard approaching from the rear left, if your seat
vibrates on the left side of the driver's seat there is a
haptic feedback that alerts you that something is happening to
the left of the vehicle. Or if it is intended that you are, for
example, departing your road lane, the vibration of the
steering wheel is directly related to something that is going
on with the steering system that the driver should pay
attention to.
We have formed our own group to work on human factors to
specifically look at the human-machine in action and we work
very intensively not only with our OEM customers but also with
academia on this topic.
Mr. Bilirakis. Mr. Klei, do you want to comment on that as
well?
Mr. Klei. Yes, I think similar to the Bosch development we
also have a very significant investment in the human-machine
interface technologies. We have been one of the leaders in
displays, in clusters, and in warning systems for vehicles for
many, many years. We think that is an important part of
bringing these technologies to market safely.
Clearly, when it comes to the audible versus haptic, we
have done a lot of research as well. We actually have driver
monitoring cameras that we are looking where the driver is
seeing, or looking, where the driving task should be. And we
sometimes use LED lights or other ways to try and bring the
driver's attention back to the driving task. That is a big
question.
As you talk about Level 3 technologies that is the biggest
question and the biggest area of development is how do you get
the driver disengaged and then re-engaged fast enough to resume
the driving task. And I think that is a challenge for the
industry. That is why you see some developing from Level 2 to
Level 4, some are going to go through Level 3. But that is
probably one of the biggest challenges and we are investing
heavily in this area.
Mr. Bilirakis. OK. As a follow-up, are consumers able to
manually turn off these alerts or warnings or customize them to
their individual preferences?
Mr. Klei. So that is really a question for the OEM to
determine what they would like to do. And it happened as well
with ABS and electronic stability control and the various
traction control systems, the OEMs for many years could
determine which could be turned on and off. So it is something
that some allow, some don't. We believe that ultimately when it
is proven that the safety technologies are really going to save
lives that it shouldn't be turned off. It should be developed
over time to be very easy to understand, very easy to use, and
will ultimately save lives.
Mr. Bilirakis. OK. I have a question with regard to
actually a follow-up on the gentleman from West Virginia. I
mean, we want to help a lot of the elderly, maybe physically
disabled people get around. We don't have in my area, in the
Tampa Bay area we really don't have a mass transit system, so
this could be extremely beneficial to people getting to
doctors' appointments, what have you, these automated cars.
But you anticipate them having a standard driver's license;
is that correct? I mean they have to qualify for this. For
example, if you have a visual disability, if you are visually
impaired and you don't qualify. I am visually impaired but I
qualify at this particular time. I have a standard driver's
license. I don't drive at night, but 5 years from now, who
knows? Will I be able to drive one of these cars even though I
am visually impaired? That is just an example there. Can I hear
from one of you? What do you anticipate?
Mr. Klei. Certainly we believe like we have talked a lot
about the improvements in mobility for disabled and then
certainly we think these technologies will offer significant
improvements here. But it takes time and it takes really more,
the systems that are developed with that in mind. And that is
why we are working hard as a company with our OEM partners to
make sure that these systems are developed with all
considerations in mind. It is not just for the driver that has,
you know, zero disabilities. It is to provide mobility for
everyone. And we think there is a clear promise and they are
being developed with this in mind.
Mr. Bilirakis. Anyone else?
Mr. Gouse. May I, please. We have been working with AAMVA,
the American Association of Motor Vehicles Administrators, on
that exact topic for both cars and trucks. And a simple example
would be some States require that parallel parking is required
to get your initial driver's license, but in some vehicles the
vehicle itself can parallel park without the assistance, with
the assistance----
Mr. Bilirakis. If you could put the mike a little closer.
Mr. Gouse. So we have been working with them trying to
define what features are in place or are possibly in place in
the future and they can design their driving tests and their
ratings or perhaps certification levels like a commercial
driving license has or something that says you can operate a
Level 3 vehicle with these features, but you can't do a
completely manual one. You can't drive a manual transmission
anymore. So it is a complicated question, but it is being
worked on.
Mr. Bilirakis. And there will be a State issue, obviously,
as far as that is concerned. OK, well, that is important. I
mean, we have got to know that ,because we want to help out our
constituents. But again, you know, if you have a standard
driver's license you qualify. And the gentleman asked about
someone that is intellectually impaired. You know, would that
person qualify? More than likely they couldn't get a license.
So anyway that is something we have to resolve, so I appreciate
that. I have one more question if I have time. I don't have
time.
Mr. Latta. Yes. If you would like to submit it in writing
that would be great.
Mr. Bilirakis. Yes, I will submit it. Thank you very much.
I yield back. Thank you.
Mr. Latta. Thank you very much. The gentleman yields back,
and the Chair now recognizes the gentlelady from New York for 5
minutes.
Ms. Clarke. Thank you, Mr. Chairman, and I thank our
ranking member. I thank our expert panelists for a very
important and stimulating examination of autonomous cars.
Some experts have raised particular concerns regarding
Level 3 automation and you have discussed it here today where a
vehicle can drive itself but the driver must be ready to take
over at a moment's notice. There is some evidence that Level 3
may lead to an increase in traffic collisions. During recent
test drives, Ford reportedly noticed that even their engineers
trained to monitor autonomous vehicles had trouble staying
alert at the wheel while the car was driving. Volvo's
autonomous vehicle program is skipping Level 3 altogether and
planning to go straight from Level 2 to Level 4.
Mr. Zuby and Mr. Gouse, do you agree that complications of
Level 3 automation are an example of why it is important to
monitor autonomous technology to make sure that it is actually
making driving safer?
Mr. Zuby. Yes. Thank you for the question. Absolutely, I
think the important thing will be to be able to monitor these
developments as they are put out into the fleet. There is a
long history of human factors research that says things like
Level 3 are potential problems for human monitors, and I think
that is why you find some automakers and some technology
developers deciding that they aren't going to mess around with
Level 3.
I am not expert enough to know that Level 3 is impossible
to do successfully, but definitely there is a concern that if
the car is too highly capable at the dynamic driving task that
the driver will discontinue his monitoring activities and not
be able to resume control when it is necessary because the
system is no longer capable handling a situation.
Mr. Gouse. I would just second what David said, but I would
like to caveat with, bear in mind that people working on this--
I am just awed when I go to committee meetings and listen in at
the experts, the level of knowledge that is behind all this and
the amount of consideration that is going on for all the
aspects. Whether it be taking over control immediately or
changes in weather conditions or road issues or anything at all
these levels, it is very impressive the level of expertise and
the care that is going into this.
Ms. Clarke. The only factor that I guess is challenging to
sort of pin down is human error, right?
Mr. Gouse. Well, there are other challenges too, just like
in our normal driving that we have unexpected issues that
arise. The deer jumps out that you never saw before and how do
you react to that? Or there is some sort of a failure in the
vehicle or in the infrastructure that is unanticipated and how
do you react to that? Or someone else who has not got
automation or not got assistance and makes a grave error and
how do you react to that?
Ms. Clarke. But the reaction is the human being, right, not
necessarily the vehicle? Or is it that the vehicle would be
programmed to react to the jumping deer or the change in
weather conditions?
Mr. Gouse. Well, that goes back to the level of automation,
whose job it is, who is it assigned and----
Ms. Clarke. So Level 3 then becomes the challenge in terms
of what the standard would be for automation versus human
participation.
Mr. Gouse. The expectations between Level 2 and 3, it is a
big step.
Ms. Clarke. OK. As we have heard, semi-autonomous features
can have significant safety benefits but they may also be
confusing, especially to drivers who are unfamiliar with the
technology or fail to use it correctly. Consumer education will
be essential to ensuring that the full advantages of these
technologies are realized.
Mr. Zuby, why is it so important that drivers understand
that limits of semi-autonomous features and are aware of what
exactly their cars can and cannot do?
Mr. Zuby. Yes, for exactly the issues that we have been
discussing about Level 3. I mean it will be important for
drivers to understand how close attention they need to pay to
the driving situation in order to be ready to take over and
wonder what situations the system is likely to hand control
back to them.
But we would say that I think it is important to try to
figure out how to design these things so that the limitations
and the way they work is as intuitive as possible because I
don't think we can rely on people to spend extra time to learn
how to drive their cars. I mean how many people in this room
have read their owner's manual from front to start? There is a
lot of really important information in there, but I for one
have not read the owner's manual from start to finish for any
of the vehicles I have ever owned.
Ms. Clarke. Very well. Mr. Chairman, I yield back.
Mr. Costello [presiding]. Mrs. Walters.
Mrs. Walters. Thank you, Mr. Chair.
Mr. Gouse, we know that many States and localities have
developed legislation aimed at regulating self-driving cars.
Can you go into further detail on the State localities
implementing SAE's level of driving automation into their laws?
Mr. Gouse. I am most familiar as a staff person with
Pennsylvania and Michigan and California. But there are, as
Jeff said earlier, there are two or three dozen States, and at
each State or Commonwealth there is an upper chamber and a
lower chamber and also there may be a regulatory agency, or two
of them that are working in concert or in parallel paths. So
there are quite a few going on.
And our members who are active are picking up things. I
know New Jersey is talking about it. I heard that from a member
yesterday. North Dakota is a State, I believe. So it is not our
main business as SAE to monitor State activities, but we want
them to adopt the SAE language so there is consistency across
all the States and territories.
Mrs. Walters. Yes. I think that is going to be an issue.
The consistency is going to be obviously very, very important.
And then the same question for you again is a number of groups
have developed classification systems to define automated
driving systems, and can you discuss why SAE determined the
J3016 standard to be the most optimal way of defining the
different automated driving systems?
Mr. Gouse. I would just like to say probably that the
committee leadership and members worked very hard on this over
quite a bit of time with a tremendous amount of input from
various different stakeholders. And it is not just a committee
of technology developers, there are policy folks in there,
NHTSA was part of it, motor carriers, Federal Motor Carriers
was part of it.
So it was an ongoing process. It was in fact adopted
internationally before NHTSA did even at the Amsterdam
convention in April of '16, I believe. So it is becoming a
global standard and it is being validated that way across the
globe and in the States as being the preferred choice. It is
also a living document. It has been revised already once since
it was issued. In fact, the name was even changed a little bit
to clarify it. So it will go through revisions and additional
references to discuss some of the issues that were brought up
here in questions to add to it.
Mrs. Walters. OK, all right. Thank you very much, and I
yield back the balance of my time.
Mr. Costello. Mr. Cardenas, you are now recognized for 5
minutes.
Mr. Cardenas. Thank you very much, Mr. Chairman. Something
just occurred to me. Are we likely going to see in the near
future--I grew up learning how to drive on a stick shift. A lot
of today most drivers in America probably don't know how to use
a manual or a stick shift vehicle, these automatic gear
shifting vehicles. Are we looking at possibly in the near
future where people get in their car and they push a button,
today I am going to use automation 1, 2, or 3 Level, and maybe
that is the new gear shifting or shifting of the vehicle that
we are going to be driving in the future? Does that make any
sense, or is that probably likely what we are going to be
looking at?
Mr. Klei. I think one of the things that we look at when we
are looking into development is you never take the fun away
from driving your car. We still like the ability for people to
drive their cars when they want to drive their cars. But there
is many driving tasks, there is many opportunities for disabled
to provide mobility, and that is where we think the big benefit
will be. We never want to take the fun away though.
So it could be someone gets in a car and says yes, I want
to go from point A to point B in an automated way or it could
be that I want to drive myself on the windy country roads. So I
think there is going to be some opportunities there over time
for people to still have fun, but in certain circumstances
still get the mobility that they need and they want and to be
able to do other things in the car.
Mr. Cardenas. Well, speaking of taking the fun away
driving, I can envision if we are going to be appropriate as a
Government, and maybe in the future what we have is a speed
limit technology where if you are going to be driving an
automated vehicle then the speed limit is 35 miles an hour.
Your car is not going to be allowed to go over 35 miles an hour
on that piece of the road.
Mr. Klei. Yes. I mean, I think these are things that we
need to consider, but quite frankly we believe that if you do
that you could actually introduce more challenges because
everyone will try and go around the car. You want the car to
flow naturally with traffic with other automated vehicles as
well as nonautomated vehicles, so you want it to be very
natural, and through testing and development that is what we
are developing for. So to limit a car and limit the mobility
and limit the functionality is going to limit the testing and
deployment of such technologies and potentially lifesaving
benefits.
Mr. Cardenas. For those of you who are on the panel from
private industry, I mean how do you feel about your
relationship right now with Federal departments when it comes
to reporting and expectations of, you know, obviously
nonproprietary progress and letting them know what you are
looking for as long as timing of introducing products, et
cetera?
Mr. Klei. I think, Congressman, it is a great question. It
is one that through the Federal Automated Vehicles Policy that
was rolled out last September from NHTSA it is a great start to
bringing the collaboration together between industry and
Government. And we think it is a big step forward but there is
more work to do.
In that policy it requires significant reporting between
the industry and NHTSA and that reporting needs to be better
defined, it needs to be more expedited, and the exemption rules
that we are all looking for especially in the development side
need to be improved. And so we are working closely with that
agency, with NHTSA to try and improve that and make sure that
when it is officially rolled out and deployed that really it
is, in fact, usable and it is going to drive this technology
forward and potentially save lives when deployed.
Mr. Cardenas. What country right now seems to be more, I
don't want to use the word advanced, but more ready and willing
to allow their constituents to drive the highest class of
automated vehicle right now?
Mr. Klei. Every country has certain limitations and certain
regulations and there is no one country that is easy. Every
country has different----
Mr. Cardenas. I mean, is there a particular country right
now that--I am thinking of Germany. I am wondering if they are
allowing a little bit more than we are so far.
Mr. Klei. I don't know that there is one country that says
it is easy to do. Every country has certain limitations and for
good reason.
Mr. Cardenas. Anybody know what is going on around the
world?
Dr. Stepper. The same as Mr. Klei said, from my side
sometimes it is not even regulated by a specific country law.
You know, also in Germany, you mentioned Germany as an example,
the different States have different laws and different
regulations and the regards of allowing or not allowing
different levels of automation. There may be some States that
are really fostering the rollout so that companies like Bosch
can go on public roads and test and validate the systems which
is very helpful for our development to be allowed to do that.
Mr. Cardenas. OK. Thank you, Mr. Chairman.
Mr. Costello. Mr. Mullin, you are now recognized for 5
minutes.
Mr. Mullin. Thank you, Mr. Chairman.
Doctor, is it Stepper?
Dr. Stepper. Yes.
Mr. Mullin. Thank you for being here. You talk about the
technology and moving forward with the technology of going out
and testing the vehicles. But can you explain a little bit more
how that works with the technology of the vehicle versus the
GPS----
Dr. Stepper. Yes.
Mr. Mullin [continuing]. That the vehicle I am assuming has
to be programmed into a GPS and it has got to take you from
point A to point B; is that correct?
Dr. Stepper. So it depends on the level of automation,
Congressman. So if you would go all the way to a Level 5
automated driving, for example, which really takes the driver
out of the loop and there is no longer a driver required to
operate the machine that it would exactly the scenario that you
would dial in a particular destination and the vehicle will
take you there, for example, door to door.
Mr. Mullin. Well, what is Level 1?
Dr. Stepper. In Level 1, this is what we call today's
driver assistance systems where there is----
Mr. Mullin. Where your seat vibrates and it tells you and
does all that stuff?
Dr. Stepper. For example, there would be a warning that
there is an impending front-to-rear-end collision or there is a
lane departure that is about to happen.
Mr. Mullin. And 2?
Dr. Stepper. Two combines the longitudinal and lateral
control of the vehicle so, for example, we still call it the
assistance functions. It is functions like a traffic jam assist
where the vehicle in that particular scenario in a traffic jam
would automatically take the control for the longitudinal and
the lateral perspective of the vehicles but the driver is still
fully responsible and fully in the loop, whereas in Level 3,
for example, you take that as one example to a traffic jam
pilot where you can take your hands and your feet off for a
well-defined scenario.
You need to be on a Class 1 road. On a traffic jam pilot,
for example, you need to have preceding traffic, and then for
this stop and go traffic the machine would take over the
control of the vehicle until it handles it back to the human
being.
Mr. Mullin. And 5 is what we started the conversation with.
Do we see the advancement of the vehicles catching up or going
to surpass the GPS? Because everybody uses their road maps and
their GPSs on their phones and I am sure I am not the only one
that it takes me to the wrong place all the time.
Dr. Stepper. Yes, yes.
Mr. Mullin. So they would have to work simultaneously,
wouldn't they?
Dr. Stepper. Yes, so they actually, Congressman, there is
additional technology that is required. So what we know today
as GPS, also standard definition maps, for Level 4, Level 5
automated driving to a certain extent even for Level 3, we have
the need for high resolution, highly dynamic maps that really
exceed the requirements that we see from the map requirements
from today's navigation system. And that is actually coupled in
a process called data fusion with onboard sensing via radio
cameras, your radars, your other sensing technology you may
have on board on the vehicle that will recognize certain
landmarks like a fire hydrant, like a bridge, like a certain
exit, and it combines the GPS information----
Mr. Mullin. That is more of an eyesight on it.
Dr. Stepper. As well as nonvisible electromagnetic base
like radar, for example, or LiDAR technology which uses laser
light.
Mr. Mullin. So would this be one entity or would each
company be responsible for their own technology for the GPS to
which their vehicle is going to be operating by?
Dr. Stepper. It really comes together at the end at the
vehicle manufacturer. There may be different suppliers for
certain sensing technologies or GPS technology. What really is
the trick to have the competency in bringing all this data
together in this data fusion process and derive driving policy
decisions out of that.
Mr. Mullin. What I am talking about is somebody working on
this end of the GPS as you guys are working up with the
vehicle, are they going to meet? Or when the technology for the
vehicle gets to that point, then we start diving into the
precise GPS?
Dr. Stepper. Yes, so that is already available today in a
system that is called differential GPS systems that increases
the resolution. Most companies, actually, out there testing and
validating automated driving today use differential GPS system
to get them to the resolution that they need, which in essence
is a centimeter resolution as opposed to a couple meters that
we see today. So that technology is already available today.
The challenge in the development is going to be to bring the
prices down and the costs down of such an advanced GPS system
for use in every vehicle.
Mr. Mullin. Is there one company that is leading that?
Dr. Stepper. There are several companies that are working
on that exact topic. There is not one company that stands out.
Mr. Mullin. Do you have one particular one that you are
working with?
Dr. Stepper. We work really with all of them at the moment.
There is no particular one that I can point out at the moment,
Congressman.
Mr. Mullin. All right, thank you. Thank you for your time.
Mr. Chairman, I yield back.
Mr. Costello. Thank you. I will now recognize myself for 5
minutes and ask a question to all panelists, two-part question:
One, how is the development and testing of these systems
different from the development and testing of fully self-
driving technologies; and second, how much can be learned from
the development and testing of advanced driver assistance
systems?
Mr. Klei. So first, what is different, I don't really think
there is so much difference in the way we develop and we test
technologies, everything from ABS through electronic stability
control and all the way to fully automated driving. It is a
very rigorous, long testing process. It starts with the
technology itself. It starts with bench testing, then in
contained track environments, and we evolve all the way to,
ultimately, the real road and real world testing.
So the process is very similar. Obviously, the conditions
by which we test are going to be different depending on the
technology. But in terms of the rigorous, you know, Six Sigma,
continuous improvement mindset that we have to make sure the
products are safe is no different regardless of what the
technology is. The challenges are bigger the higher levels of
automation you go to, but the testing process itself is always
very much the same, safety first.
When it comes to the implementation of these and across the
various product portfolio again everyone is going to be
different, and ultimately it is the OEM that decides when it is
safe to deploy in the vehicle. We work with OEM customers and
they ultimately are the ones that certify for FMVSS.
Mr. Gouse. I would like to just briefly add a couple
things. Prior to the beginning of testing, there are some tools
you put in place, what are called a design failure mode effects
analysis and failure mode effects analysis, where you look at
all different ways a system might fail and then you design a
test procedure to encompass that and then you look at when
something fails, whether it is part of the system or something
external or you are testing an automated vehicle, but the
engine conks out or something or you get a flat tire, you have
to build all of that into your test procedures. And so you have
got a complete, very comprehensive, and carefully designed
program to execute as part of the process.
Mr. Zuby. Yes. I would agree with Mr. Klei and Mr. Gouse
that the process is similar. But I think one of the things that
we need to keep in mind that as we deploy increasingly evolving
technologies we do need to watch them very carefully and see
how they perform in the real world. And when they fail to
perform try to understand whether or not they are failing to
perform because of a deficiency in the technology, a deficiency
in the logic behind the technology, or because the circumstance
in which they failed is just outside the design domain of that
particular technology.
Again, consequently, I think information about what is
happening in the real world as these technologies deploy is
going to be vitally important to making sure that this stuff is
developed in a safe way.
Dr. Stepper. And if I just may add a few points. Number one
is what we didn't have available in the past when we started
developing ABS or ESP, for example stability control, was an
international standard specifically designed for the different
safety assessments and different safety levels. And that
standard is called ISO 26262 which was specifically developed
for use in the automotive space to define different safety
levels and also define how to get to and what you have to meet
in order to get to the different levels of this safety.
Number two, what we didn't have available when we are
deploying ABS or electronic stability control or early in
driver assistance is the vehicle being connected to the rest of
the world, being connected to servers. If we would just proceed
with conventional validation as we have in all these decades it
would really be cost and time prohibitive. We would; literally,
in order to fully validate a fully automated vehicle we would
have to drive a distance that equals the average distance
between the sun and the earth which is not feasible from a cost
and time perspective.
So what we continue to deploy is the advantages of being
connected and having vehicles deployed in the field that
collect for us very valuable data of real world traffic
situations that we then can take back to analyze and develop
and adjust our software, for example, accordingly.
Mr. Costello. Thank you. Seeing there are no further
members seeking to ask questions for the first panel, I would
like to thank all of our witnesses again for being here today.
Before we conclude, I would like to include the following
documents to be submitted for the record by unanimous consent:
a report from MEMA; Advocates for Highway and Auto Safety's
FAVP comments in a March 27th letter to Chairman Latta and Ms.
Schakowsky; a statement from the National Safety Council; a
statement from Global Automakers; a letter from the U.S.
Chamber of Commerce, Technology; a statement from American Car
Rental Association; a statement from Mobileye; a statement from
EPIC; and a letter from Honda.
[The information appears at the conclusion of the hearing
and at http://docs.house.gov/Committee/Calendar/
ByEvent.aspx?EventID=105790.]
Mr. Costello. In pursuant to committee rules, I remind
members they have 10 business days to submit additional
questions for the record and I ask that witnesses submit their
response within 10 business days upon receipt of the questions.
Without objection, the subcommittee is adjourned.
[Whereupon, at 12:00 p.m., the subcommittee was adjourned.]
[Material submitted for inclusion in the record follows:]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
[Additional information from AHAS is available at http://
docs.house.gov/meetings/IF/IF17/20170328/105790/HHRG-115-IF17-
20170328-SD004.pdf.]
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]
[all]