After several incidents with autonomous driving and cars being hacked and taken over, a more dangerous side of the new technology appears.
Hacker attacks or faulty software could shift the burden of legal and regulatory liability toward makers of self-driving cars and away from customers, experts say, forcing regulators and insurers to develop new models.
Autonomous cars have the potential to reduce the rate of traffic accidents as sensors and software give a car faster and better reflexes to prevent a collision. However, a greater level of automation increases the need for cyber security and sophisticated software, experts said.
“Although accident rates will theoretically fall, new risks will come with autonomous vehicles,” said Domenico Savarese, Group head of Proposition Development and Telematics at Zurich Insurance.
“What should be done in the case of a faulty software algorithm? Should manufacturers be required to monitor vehicles post-sale in the case of a malfunction or a hacker attack?” Savarese asked.
While established models for assigning liability – such as holding the owner responsible for what the car does – will still be relevant, the onus may shift toward manufacturers.
Greater automation may also change consumer behavior and affect insurance costs if drivers become less vigilant and less practiced in their ability to avert an accident.
While self-driving cars are expected to have many positive impacts, there are also some potential drawbacks to consider.
Just as Wired magazine published a headline-grabbing story about hackers taking control of a new Jeep Cherokee with UConnect, engineers, computer programmers, professors, and lawyers were meeting in Ypsilanti, Michigan, to discuss evolution of the autonomous, connected automobile.
The Automated Vehicle Symposium is held every year by the Association for Unmanned Vehicle Systems International (AUVSI) and the Transportation Research Board (TRB), and so it measures advances in these technologies in increments. Questions and concerns about security, ethics, and who’s responsible for the first crash caused by an autonomous car are not new for this group.
This year’s AVS followed the University of Michigan’s grand opening of its Mcity autonomous vehicle test track, where some of the suppliers and automakers participating in the symposium gave demonstrations of their latest technologies. I stayed indoors and listened to presentations. While there were no revelations, there were some interesting ideas that give clear indications where the automobile industry and our transportation system are heading. Herewith, a few tidbits:
The University of Michigan is extending its three-year-old testing of smart cars on public roads into an “entire system of connected roads,” entailing some of the major roads between metro Detroit and Ann Arbor, according to John Maddox, assistant director for the U-M Mobility Transformation Center (MTC). It could rival Google’s autonomous car-testing efforts in Silicon Valley and adds the element of ice- and snow-covered roads, which semi-automated cars can’t handle very well so far.
The MTC essentially is an extension of the university’s test of smart car technology that began on public roads in 2012, with a $100 million budget funded by automakers and suppliers, state and federal governments, and the university.