Automated Vehicle Guidance Launches Framework for Public Discussion, Vehicle Integration
Tweet
In July, speaking at AUVSI’s and TRB’s Automated Vehicles Symposium in San Francisco, Secretary of Transportation Anthony Foxx promised a federal automated vehicles policy that would “prepare our ecosystem to integrate these new types of vehicles into the bloodstream of American infrastructure.”
Today, the Department of Transportation released that policy, which provides guidance on monitoring and reporting vehicle performance, a model policy for states to avoid having a patchwork of regulation, and some new regulatory tools that will help regulatory agencies stay ahead of the new technology.
Speaking in front of a lineup of automated vehicles, including an Audi, Toyota and Cadillac, Foxx said today that “in the 50 years of the United States Department of Transportation, there has never been a moment like this. A moment where we can build a culture of safety as a new transportation technology emerges that harnesses the potential to save even more lives and that will improve the quality of life for so many Americans.”
Brian Wynne, President and CEO of AUVSI, said in a statement that the policy is “another example of industry and government working together to advance innovations. The guidelines create a flexible framework that is critical to safely accelerate the deployment of automated vehicles and accommodate future innovations. By defining the federal and state responsibilities in regulating automated vehicles, they also provide the regulatory clarity necessary to foster the advancement of this emerging technology.”
Most of the document focuses on highly automated vehicles, ones that can take full driving control in at least some circumstances, which is defined as Level 4 automation according to SAE International.
The policy leaves it up to automakers to determine the automation levels of their vehicles — where can the operate, under what conditions, and at what speeds — along with a fallback to a “minimal risk condition” if the automation should fail. That doesn’t necessarily mean giving control back to a human driver, who could be sleepy or impaired, but could mean just bringing the vehicle to a stop.
Automakers are to voluntarily provide reports on their systems and their adherence to the guidance, although the policy notes that this reporting could be made mandatory in the future.
If manufacturers make extensive hardware or software changes to a vehicle, such as ones that increase its speed or expand the types of conditions in which it can operate, the manufacturers would need to submit a new safety assessment for those new capabilities.
Automakers also need to have a documented process for testing, validating and collecting crash data or data on other road incidents, which can be used to establish the cause of any problems.
That would include “positive” incidents, where an automated vehicle successfully avoids an accident.
Vehicle makers will also need systems to allow vehicles to share data, such as on road conditions, but without violating the privacy of vehicles owners.
“Data sharing is a rapidly evolving area that requires more research and discussion among stakeholders to develop consensus on data standards,” the policy says.
Manufacturers also need to pay careful attention to the human-machine interface, so drivers know when the system is doing the driving and when they need to take over. Drivers also need to be educated on how the systems work, and DOT suggests giving them on-road or on-track, hands-on experiences.
Model State Policy
To avoid having more than 50 regulatory schemes across the country, DOT “strongly encourages states to allow DOT alone to regulate the performance of HAV technology and vehicles,” the policy says.
If a state goes ahead with its own rules, it should consult with the policy and with the National Highway Transportation Safety Administration. However, federal performance regulations will trump state rules, the policy says.
However, states can use their authority to set up highway safety programs, driver education and testing, vehicle inspection programs and other related policies.
Among other things, the model framework calls for states to set up a lead agency responsible for consideration of any highly automated vehicle testing, and should determine liability rules for such vehicles. If an automated vehicle crashes, who is liable?
“For example, states may determine that in some circumstances liability for a crash involving a human driver of an HAV should be assigned to the manufacturer of the HAV,” the policy says.
New Tools
While NHTSA already has a substantial number of regulatory tools at its disposure, it sometimes needs new ones as new technology arrives, the guidance says.
This could include pre-market approval authority for some HAV systems, the way that the Federal Aviation Administration uses it to pre-approve autopilot systems on commercial aircraft and drones.
Under that scenario, rather than having manufacturers self-certify to standards, “NHTSA would test vehicle prototypes to determine if the vehicle meets all such standards,” the document says. That could contribute to public confidence in the vehicles, according to the DOT.
Foxx said the proposed guidance isn’t the last word, and the agency expects significant input from the public.
“We do not intend to write the final word on highly automated vehicles here,” Foxx said in the document’s introduction. “Rather, we intend to establish a foundation and a framework upon which future agency action will occur.”
He also noted that it will be updated yearly, and said, “we very much look forward to the dialogues that will emerge in the coming weeks and months and thank you in advance for helping us.”
To read more about the new federal policy on highly automated vehicles, go here:
