Autonomous vehicles (AVs) are coming. Every major manufacturer is testing them (as is Google). And some states, like California, Florida, and Nevada, have begun AV licensing. Experts expect driverless cars to be a mass market reality by 2025.
A new report from the Eno Center for Transportation looks at the potential implications. And it's largely positive about the prospect. AVs have great potential to reduce accidents, alleviate congestion, and improve fuel efficiency, the D.C.-based think tank says.
It's not surprising when you think about it. Computers are already better drivers than tired, drunk, and distracted humans. More than 40% of fatal crashes today are caused by alcohol, drugs, or fatigue. Up to 90% of crashes are the result of errors, rather than equipment or infrastructure failures. It may be a while before AVs can deal with complex or unusual conditions. But Eno expects it to happen. AVs could cut accidents to 1% of current rates, or nearer to aviation or rail levels, it estimates.
AVs will also do a better job of sensing and anticipating the movement of other cars, leading to smoother braking and better fuel usage. Eno expects AVs to use "existing lanes and intersections more efficiently through shorter headways, coordinated platoons, and more efficient route choices." Vehicle-to-vehicle and vehicle-to-infrastructure communication could allow cars to drive closer together (as seen here) and reduce stop-starting on freeways. Reducing accidents will also help cut congestion.
Putting all factors together, Eno estimates the potential benefits in reduced crashes, lives lost, and economic gains. The numbers are impressive. At a 10% AV penetration rate, it expects 1,100 lives saved a year, 211,000 fewer crashes, and $5.5 billion in economic savings. At 50%, it predicts 9,600 fewer fatalities, 1.8 million fewer crashes, and $48.8 billion in savings. At 90%, it says there could be 21,700 fewer deaths, 4.2 million fewer crashes, and $109.7 billion in savings.
And those are just the economic savings from reduced crashes. When you include fuel and congestion benefits, and more efficient parking, the number gets on for half a trillion dollars a year ($447 billion). That's not chump change.
Of course, getting to 90% AV adoption is going to take time and technological improvements. AVs are likely to be more expensive than standard vehicles, at least to start. States will have to develop new types of licensing systems and new infrastructure. There are likely to be a host of legal issues (who has liability when computers make a mistake?). And then, there's the hacking issue. It can't be long before we see the first AV spin out of control, because some idiot forgot to update the latest security patch on their operating system.
Still, Eno shows the potential of AVs outstrips the downsides. It's not like driving is that great today. There ought to be room for improvement, if we put our minds to it.