US lawmakers to hear from OEMs on autonomous-vehicle regulation
The US House Energy and Commerce Committee has scheduled a hearing to discuss autonomous-vehicle regulation, with GM, Toyota, Lyft, Volvo, and Rand scheduled to offer testimony. A review of the prepared remarks suggests some common fronts among them include a concern for having a single national standard.
IHS Markit Perspective:
- Significance:The US House Energy and Commerce Committee is holding a hearing on 14 February 2017 to discuss issues around testing and deployment of self-driving cars. Among those in attendance will be executives from GM, Lyft, Toyota, and Volvo.
- Implications: The participating companies have released planned remarks ahead of the hearing, with each stressing issues somewhat differently, though nearly all addressed the issues of a potential patchwork of rules from state to state, as well as the potential benefits of improved safety and traffic flow.
- Outlook:Largely, these comments are consistent with the kinds of testing that each company is doing. Automakers and Lyft are concerned about the issues of inconsistent standards from state to state, in the absence of a single standard applicable across the country. One outlier is the prepared testimony from think tank RAND, who suggests that on-road testing may simply not be enough, given the low instances of accidents and fatalities per mile driven. However, the hearing's outcome could be months or years in coming. The US agency with responsibility for vehicle safety oversight, the National Highway Safety Transportation Administration, takes overall direction from the US Congress. While the agency can move forward on several fronts, congressional mandates are part of the process for rulemaking.
The US House Energy and Commerce Committee will convene a panel on 14 February 2017 to discuss deployment of self-driving cars and regulations; ahead of the meeting, statements previewing testimony from various participants have been posted to the committee's website. Testimony is expected from GM (Mike Abelson, VP of global strategy); RAND (Dr Nidhi Kalra, co-director and senior information scientist); Volvo (Anders Karrberg, VP of government affairs); Lyft (Joseph Okpaku, VP of public policy); and the Toyota Research Institute (Gill Pratt, executive technical adviser and CEO).
GM makes the case that development of self-driving cars should accelerate from a position of improving safety. Abelson calls for Congress to give the Secretary of Transportation authority to grant exceptions for testing, as well as noting that faster changes to Federal Motor Vehicle Safety Standards (FMVSS) will be necessary to deploy the technology. Without revision to the rules, deployment could be slowed by years. "At the same time, we understand that we must be able to prove to our customers, our regulators, and the American public that our vehicles are safe," said Abelson. "NHTSA (the National Highway Safety Transportation Administration) has already begun a collaborative process with stakeholders to facilitate the safe testing and deployment of self-driving vehicles. While important regulatory work continues, it is imperative that manufacturers have the ability to test these vehicles in greater numbers to gather the safety data that will be critical to inform large-scale deployment of life-saving self-driving vehicles. One good way to accomplish this goal is to grant the Secretary of Transportation authority to grant specific exemptions for highly automated vehicle development," he said.
The representative from RAND, however, will suggest that extensive testing will not yield the necessary results, because of the relatively low rate of incidents of accidents and injuries. Kalra will testify that, "Data about the number and types of miles travelled and the number of crashes, injuries, and fatalities can be used to assess safety.
"However, this approach is largely impractical for pre-market testing; even though the number of crashes, injuries and fatalities from human drivers is high overall, the rate of these failures is low in comparison with the number of miles that people drive. Americans drive nearly eight billion miles every day and three trillion miles every year. The 35,092 fatalities and 2.44 million injuries in 2015 correspond to a rate of 1.12 fatalities and 78 injuries per 100 million miles driven," she will say.
"Given that current traffic fatalities and injuries are rare events compared with the vehicle-miles travelled, fully autonomous vehicles would have to be driven hundreds of millions of miles, and sometimes hundreds of billions of miles, to demonstrate their reliability in terms of fatalities and injuries. Under even aggressive testing assumptions, existing test fleets would take tens and sometimes hundreds of years to drive these miles - an impossible proposition if the aim is to demonstrate their performance prior to releasing them on the roads for consumer use."
Kalra will note, however, that the safety could probably be demonstrated once autonomous vehicle were on the road. Regarding partial test simulation, which both Tesla and Google have leveraged, Kalra's testimony seems to favour partial simulation testing, though she acknowledges that it is unproven. Although taking a different tack from GM, Kalra also notes that existing functional safety standards are not designed for autonomous vehicles, and recommends further work to adapt them to the unique challenges of autonomous vehicles.
"In sum," Kalra will say, "the transportation industry and policymakers do not yet have a method that is both practical and sound for testing autonomous vehicle safety … This does not mean their use should be prohibited; the technology has too much potential to save lives. Instead, it suggests that the race to develop autonomous vehicles needs a parallel race to develop methods for demonstrating and managing their safety. "Ultimately, the Rand group recommends feasible, sound methods of testing, to be implemented urgently and that the methods should be built into a flexible regulatory framework. While it is more cautious about the depths of effort preliminary testing can provide, its overall position supports moving faster rather than slower.
Volvo, which has begun testing with Uber and has created a new autonomous-driving (AD) software joint venture with supplier Autoliv called Zenuity, has set a target that no one is killed or seriously injured in a new Volvo by 2020. For Volvo, self-driving vehicles will assist in reaching that goal. Karrberg is also to note that self-driving cars have potential to create a smooth traffic flow, leading to improved fuel efficiency and improved air quality, as well as enabling better city planning because they could park themselves and that could happen outside a city centre. Volvo also notes its Drive Me testing programme in Sweden, which has already put autonomous test cars in the hands of consumers. In fact, Karrberg will testify, "[Volvo] believes it is critical to use real customers on public roads in order to capture all the human aspects of self-driving. Our goal is to define the technology based on the role of the driver - not the other way around." The company expects to offer self-driving cars by 2021, albeit in selected cities in China, Europe, and the United States. Intially, these will be geofenced, operating autonomously under normal traffic conditions on selected commuter roads. Karrberg also reiterated a statement Volvo has made previously, that the company believes product liability should rest with the manufacturer and that it will assume liability for any SAE Level 4 vehicles - if a crash or incident is a result of a defect in the AD technology and the car has not been misused. Volvo also plans to criticise the existing voluntary national plan for not effectively preventing a patchwork of state policies. Instead, Volvo will say, "Congress and NHTSA should set the laws and regulations with respect to vehicle technology and vehicle performance" and the NHTSA should promote inclusion of active safety systems by reflecting them in the US-NCAP (new-car assessment programme) safety rating system. "The current US patchwork of numerous and various state bills and laws on testing and deployment of self-driving cars could however discourage market entry and stifle autonomous driving development. It would also represent a competitive disadvantage for the USA in the global race to reap the benefits of job creation that this technology provides," Karrberg plans to say.
Lyft's interest in autonomous driving is about ride sharing and Okpaku is scheduled to say, "Lyft believes that the introduction of automonous vehicles via a ridesharing network will fundamentally transform cities and the way people move around them." Okpaku will also note that Lyft expects its model to help see a world with fewer cars on the road, less congestion, and increased positive environmental impacts. Lyft will also attack the potential of a variety of state-to-state rules, noting that the "greatest potential obstacle is constrictive legislation and regulations." For Lyft, this is an issue relevant to both the vehicles themselves and to its fundamental business model. The company suggests that Congress look at revising NHTSA's exemption authority to allow for a greater number of autonomous vehicles to be allowed on the road for testing, as well as for NHTSA to begin updating current FMVSS rulemaking sooner rather than later.
The key message of the head of the Toyota Research Institute, Gill Pratt, based on a review of his prepared remarks, focuses on who is determining what is an acceptable level of safety. Pratt is expected to say, "Policymakers - working with industry and relevant stakeholders - must determine what constitutes a sufficient level of safety for autonomous vehicle technology. As we sit here today, it is not clear how this measure will be devised or by whom … However, before developers can complete testing of these systems and deploy the technology, this foundational question will need to be answered." On testing, Pratt will say, "Without making public roads available for extensive testing, we risk companies or entities deploying autonomous vehicle technology that is not yet ready for prime time." Pratt will also tackle the subject of a state patchwork of regulations, while suggesting that a primary objective of policymakers should be encouraging and incentivising manufacturers to test the technology on a wide variety of roads to ensure safety.
"Under a patchwork of inconsistent state laws, technology may meet performance requirements in one state and not in another state … slowing the development and deployment of the technology." Pratt will take issue with the Federal Automated Vehicle Policy, arguing that the policy does not sufficiently differentiate between testing and commercial deployment; he recommends that the FMVSS be updated to deal with commercial deployment. Pratt will make an interesting point on the number of miles and locations tested, noting that public testing is required not only for normal situations, but also for as many of those 'edge cases' as possible. "The truth is that all testing miles are not created equal, and developers should be focused on testing scenarios where driving is challenging or even exceedingly difficult."
Outlook and implications
Largely, these comments are consistent with the kinds of testing that each company is doing. Automakers and Lyft are concerned about the issues of inconsistent standards from state to state, in the absence of a single standard applicable across the country. One outlier is the prepared testimony from think tank RAND, who suggests that on-road testing may simply not be enough, given the low instances of accidents and fatalities per mile driven. The outcome of the hearing could yet be months or years in coming. The US agency with responsibility for vehicle safety oversight, the NHSTA, takes overall direction from the US Congress. While the agency can move forward on several fronts, Congressional mandates are part of the process for rulemaking. Regulation is one of the most significant issues facing self-driving vehicle development; this hearing can influence activity, but there will be further steps before any rules are changed.
About this article
The above article is from IHS Automotive Same-Day Analysis of automotive news, events and trends, and is a deliverable of the World Markets Automotive Service. The service averages thirty stories per day and also provides competitor and country intelligence. Get a free trial.
- 2018 CES – A confluence of technologies to drive automotive innovation
- Autonomous vehicle sales to surpass 33 million annually in 2040, enabling new autonomous mobility in more than 26% of new car sales
- Renault Group on course for strong 2017 as global sales grow 4.8% y/y during November
- Geely becomes largest stakeholder in AB Volvo, enters commercial vehicle sphere
- Indonesian new vehicle sales decline by 4.6% y/y in November, production up by 4.5% y/y
- NAIAS 2018: Honda to introduce all-new Insight, Jeep to reveal updated Cherokee
- Thai new vehicle sales surge 20.6% y/y in November; Toyota Thailand starts EV sharing service
- NAIAS 2018: Chevrolet teases all-new Silverado pick-up truck