The safety implications of driverless cars 1California is now the third state to legalize driverless cars, joining the ranks of Nevada and Florida in paving the way for what might be the new technology’s largest obstacles yet: testing and regulation.

With an estimated 80 percent of all accidents occurring as a result of human error, the safety implications of reducing that variable from the equation are substantial. However, concerns about the fallibility of computers are likely to leave driverless vehicles in need of a human behind the wheel for the foreseeable future.

The new legislation requires the Department of Motor Vehicles to establish safety regulations “as soon as practicable,” with an ultimate deadline of January 2015 at the latest. It also allows manufacturers to immediately begin testing autonomous vehicles on public roadways, provided there’s a driver ready to take over in the event of an emergency. Manufacturers will also have the ability to seek permits from the DMV to start building and selling driverless cars to consumers. Although with many experts conceding, “there’s much work still to be done before driverless cars are proved irrefutably safe and reliable in traffic,” it could be a while before driverless cars are available to the public.

But what exactly are the safety implications of autonomous vehicles?

Autonomous vehicles present an entirely new set of safety issues. The typical accidents resulting from distracted driving, following too close, failing to signal, or driving recklessly should all be avoidable with driverless cars. Unfortunately, as University of Washington assistant professor Ryan Calo notes, there’s a whole new set of safety issues that need to be considered. As an example, Calo cites that a driverless vehicle could be programmed to avoid baby strollers and shopping carts, yet potentially misprioritize the two if put in a situation where it was simultaneously presented with both.

It’s the vast series of unknowns that’s going to provide the biggest hurdle for driverless cars being accepted as a safe means of transportation. Google issued a statement this week saying, “Self-driving cars have the potential to significantly increase driving safety.” And, although this seems undoubtedly true, Calo cautions that even if the new technology proves to be statistically safer, it may face unfair amounts of scrutiny as soon as something goes wrong.

The DMV is now in the difficult position of defining how safe driverless vehicles need to be, laying out guidelines for liability, and regulating a continuously evolving new technology. Still, with Google’s self driving car recently exceeding 300,000 miles without logging an accident, the emerging technology has undeniable potential to make roads safer for everyone. The real challenge is going to be hammering out the logistics and exploring the potential hazards that might not yet be accounted for.

Photo credit:

Driverless car accidents are increasing in the United States

The National Highway Traffic Safety Administration (NHTSA) has released new statistics that reveal almost 400 reported crashes of partially automated driver-assist systems. Of these, 273 were from Tesla drivers. The crashes occurred between July 2020 and May 2021. The numbers are not weighted by the number of vehicles on the road from each manufacturer or how many miles the vehicles travelled, so the data is not intended to compare the performance of the manufacturers. The other companies reporting crashes were Honda with 90, followed by Subaru with 10. The remaining automakers reported five or fewer. Five people were killed, and six people were seriously injured in the crashes involving the systems. NHTSA issued an order for more than 100 automakers and autonomous vehicle tech companies to report serious crashes within one day of learning about them and to disclose less-serious crashes within 15 days of the following month. The agency is currently evaluating how the systems perform and whether new regulations are necessary.

Tesla’s telematics, which allow for real-time crash reports, might contribute to the high number of reported crashes. Tesla accounted for nearly 70% of the 392 total crashes. Although Tesla calls its systems Autopilot and Full Self-Driving, the company insists that the vehicles cannot drive themselves and that the drivers must be ready to intervene at all times. The crashes occurred while the vehicles were using Autopilot, Full Self-Driving, Traffic Aware Cruise Control, or other driver-assist systems that have some control over speed and steering.

The NHTSA order also covered companies with fully autonomous vehicles, and 25 companies reported a total of 130 crashes. Waymo led with 62 crashes, followed by Transdev Alternative Services with 34 and Cruise LLC with 23. However, no serious injuries occurred in any of the fully autonomous vehicle crashes. Companies were not required to report how many vehicles have these systems or how far they travelled. This data will help investigators quickly identify potential defect trends that can emerge. These data will also help identify crashes that the NHTSA want to investigate and provide more information about how people in other vehicles interact with the vehicles. The data collected by the NHTSA alone is not enough to evaluate the safety of automated vehicle systems.

Key takeaways about driverless car accidents:

  • US safety regulators have released data showing nearly 400 crashes of vehicles with partially automated driver-assist systems, with 273 of them involving Teslas.
  • The National Highway Traffic Safety Administration has cautioned against using the numbers to compare automakers since the data is not weighted by the number of vehicles that use these systems.
  • NHTSA is assessing how these systems perform and whether new regulations may be necessary, after telling over 100 automakers and automated vehicle tech companies to report serious crashes within one day of learning about them.
  • Tesla’s crashes may be high due to its use of telematics to monitor its vehicles and get real-time crash reports.
  • Honda and Subaru were the next closest of a dozen automakers that reported crashes, with 90 and 10 crashes respectively, while all other automakers reported five or fewer.
  • NHTSA’s order also covered companies that are running fully autonomous vehicles, and 25 reported a total of 130 crashes, with Google spinoff Waymo leading with 62.
  • While the data collected by NHTSA isn’t sufficient by itself to evaluate the safety of automated vehicle systems, it has enabled the agency to find out about crashes much faster than before, which is helping them to identify potential defect trends that can emerge.