The White House also weighed in last week. President Obama’s proposed budget for the next fiscal year includes $4 billion to be spent over 10 years for related research. Transportation Secretary Anthony Foxx also said the government would remove hurdles to developing autonomous vehicles and set further guidelines for them within six months.
Cars are beginning to drive on their own in certain situations, and in the coming years, they will do increasingly more under computer control. They will follow curving roads, change lanes, pass through intersections, and stop and start.
But they will require human supervision. Significantly, on many occasions, the cars will in effect still tell their human drivers, “Here, you take the wheel,” when they encounter complex driving situations or emergencies.
In the automotive industry, this is referred to as the handoff problem, and automotive engineers acknowledge that there is no easy solution. Automotive designers have not yet found a way to make a driver who may be distracted by texting, reading email or watching a movie perk up and retake control of the car in the fraction of a second that is required in an emergency.
The danger is that by inducing human drivers to pay even less attention to driving, the safety technology may be creating new hazards.
“The whole issue of interacting with people inside and outside the car exposes real issues in artificial intelligence,” said John Leonard, a professor of mechanical engineering at the Massachusetts Institute of Technology. “The ability to know if the driver is ready, and are you giving them enough notice to hand off, is a really tricky question.”
The limitations of Autopilot, which Tesla describes as offering the ability to “automatically steer down the highway, change lanes and adjust speed in response to traffic,” were clearly visible in a recent test drive with Sebastian Thrun, a roboticist and artificial intelligence expert who led the Stanford University team that won the Pentagon’s autonomous vehicle Grand Challenge in 2005 and later founded Google’s self-driving effort.
Although Dr. Thrun left Google several years ago, he is still involved in the field of artificial intelligence. He describes himself as an enthusiastic Tesla owner. On a recent test drive, he cataloged the car’s limitations and errors, including those he described as “critical interventions” in which the driver must override the car’s behavior.