How Governments Can Clear the Road for Self-Driving Cars
There are smart steps that states and communities can take to encourage automated driving. There are also some potholes to avoid.
So you've decided that your state should have self-driving cars. How, then, do you catch the attention of the Googles, Volvos and Navyas of the world that are developing and even deploying these vehicles?
Passing a typical "autonomous driving" law will get your state noticed -- but not necessarily in a good way. Although Google pushed for these laws in Nevada (the first) and California (the most prominent), the company has since resisted legislation in other states as too restrictive or onerous. Many forms of automated driving already may be legal, and states from Texas to Massachusetts have attracted research activities without a specific regime for self-driving cars.
Meanwhile, some new laws are not helping. Michigan has expressly prohibited widespread use of these vehicles on public roads. California has conditioned automated driving on rules enacted by the state's Department of Motor Vehicles -- rules that are now 15 months overdue with no clear end in sight. Nevada requires companies testing their automated vehicles in the state to register them in Nevada regardless of any other registration. They must change license plates at the border.
Fortunately, there are many steps that a state, or a local government for that matter, can take to encourage automated driving. A new paper of mine, "How Governments Can Promote Automated Driving," describes nearly 50 strategies.
Some of these strategies are, admittedly, kind of boring. Filling potholes and painting faded lane markings, for example, will increase the roadway miles on which lane-keeping systems can operate. Since automated driving uses many of the same technologies as advanced safety systems, requiring new buses, taxis and government-owned vehicles to have the latest safety systems will increase demand for and decrease the cost of these technologies and potentially save lives in the process.
Merely recognizing that conventional driving is one of the most dangerous things we do can reframe the policy discussion about automated driving. When a self-driving car eventually harms someone -- and that will happen -- the public must not forget the more than 30,000 Americans who will die in vehicle crashes this year or the 2.3 million who will be injured. Indeed, millions will be hurt every year as long as drunk, drowsy, distracted, aggressive, hurried and otherwise reckless driving continues.
Expecting today's vehicle owners to behave more responsibly will allow automated driving to compete on fair terms. For example, if your state is like most, a driver who could cause millions of dollars of harm is allowed to buy liability insurance worth less than a hundredth of that amount. Raising these insurance minimums would convert safety from an abstract concept into a pocketbook priority. If automated driving is ultimately safer, then it may be cheaper as well.
Carefully auditing existing law can reveal other legal issues that should be addressed. Some states, for example, prohibit crossing a double yellow line even to give more space to a bicyclist; changing this prohibition could help self-driving cars operate more safely and reasonably. Automated-vehicle developers will seek any legal changes they deem necessary.
State governments can prepare for these discussions. For example, does your state have a point person for automated driving? This official should understand the technologies and coordinate with the federal government, state officials, local police, community groups and automated-vehicle developers.
Furthermore, do your administrative agencies have the flexibility to work with any developers that do reach out? Executive orders, legal opinions and policy guidance may be as useful as new legislation. Formalizing a robust mechanism for granting exceptions to existing laws-such as that prohibition on crossing a double-yellow line-can let policy develop in tandem with technology.
Your community can also play an important role in this innovation. Local conditions will matter most for truly driverless systems -- such as campus shuttles and downtown delivery robots -- that will be initially deployed in specific environments. These systems may be especially appropriate for business districts, shopping centers, airports, retirement homes and military bases, to name just a few.
A community that documents its needs and considers the role of automation in addressing them will be well positioned to compete for public and private opportunities. The U.S. Department of Transportation plans to fund intelligent transportation systems in a single "Smart City" somewhere in the United States. Meanwhile, developers of self-driving vehicles already are starting to identify communities for early deployments.
Your community may do everything right and still not make one of these short lists. But passing a superficial state law that purports to regulate automated driving is unlikely to help.
Fortunately, if your broader goal is to improve safety and mobility for those in your state or community, then there is much that you can do for them and for automated driving. For this, it's neither too early nor too late to begin.