Autonomous Cars Testing in U.S. Have Crashed More Often – But Still Might Be Safer

Ten companies are now licensed to drive autonomous vehicles in various states on U.S. roads, and three of them have been involved in crashes in California.

To examine things at this stage, the University of Michigan Transportation Research Institute examined the yet-limited miles and data to draw what conclusions it could.

Compared to conventional vehicles, lead author Brandon Schoettle reported autonomous vehicles have a higher crash rate per million miles traveled – but he offered qualifiers suggesting more time will be needed to gain a greater grasp of things.

Among the companies with crashes reported – all crashes must be reported under California law – Google had the most crashes, and crashes were reported also by Delphi and Audi.

SEE ALSO: Is Google Moving Toward Mass Producing Autonomous Cars?

In all cases, however, the autonomous vehicles were not at fault, but merely they were involved in the crash, so that was reported.

Google has the most miles traveled at around 1.2 million, and its drivers have suffered four injuries reported out of 11 crashes. These and all crashes counted for the study were while the vehicle was in autonomous mode, not conventional manual mode.

Qualifiers are in order however, says the researcher. Autonomous vehicles are being driven mainly in less-taxing situations, and avoiding things like snow. Therefore, the autonomous vehicles’ exposure has not been representative of the exposure for conventional vehicles. Also, the number of miles traveled in total is relatively scant, at just around 1.2 million. Carmakers other than Google have accrued miles only in the low thousands.

By contrast, the aggregate fleet in the U.S. in 2013 shows drivers travel about 3 trillion miles per year, so one might say the autonomous record is a drop in the bucket.

As it is, the researcher said it was worth it to examine the initial results.

NHTSA defines vehicle automation as having five levels:

No-Automation (Level 0): The driver is in complete and sole control of the primary vehicle controls – brake, steering, throttle, and motive power – at all times.

Function-specific Automation (Level 1): Automation at this level involves one or more specific control functions. Examples include electronic stability control or pre-charged brakes, where the vehicle automatically assists with braking to enable the driver to regain control of the vehicle or stop faster than possible by acting alone.

Combined Function Automation (Level 2): This level involves automation of at least two primary control functions designed to work in unison to relieve the driver of control of those functions. An example of combined functions enabling a Level 2 system is adaptive cruise control in combination with lane centering.

Limited Self-Driving Automation (Level 3): Vehicles at this level of automation enable the driver to cede full control of all safety-critical functions under certain traffic or environmental conditions and in those conditions to rely heavily on the vehicle to monitor for changes in those conditions requiring transition back to driver control. The driver is expected to be available for occasional control, but with sufficiently comfortable transition time. The Google car is an example of limited self-driving automation.

Full Self-Driving Automation (Level 4): The vehicle is designed to perform all safety-critical driving functions and monitor roadway conditions for an entire trip. Such a design anticipates that the driver will provide destination or navigation input, but is not expected to be available for control at any time during the trip. This includes both occupied and unoccupied vehicles.


Another preliminary finding is he cannot rule out the possibility that the actual crash rates for self driving vehicles are lower than for conventional vehicles.

And, said an abstract for the study, “the overall severity of crash-related injuries involving self-driving vehicles has been lower than for conventional vehicles.”

Companies that have now been licensed under California’s autonomous driving code are BMW, Bosch, Cruise Automation, Delphi Automotive, Google, Honda, Mercedes-Benz, Nissan, Tesla Motors, and Volkswagen Group of America which includes Audi.

SEE ALSO: Nevada Issues Driver’s License To A Driverless Google Prius

Presently, no company has released autonomous vehicles for public sale that are considered true full autonomous, or “level four,” by NHTSA definition.

Tesla’s autopilot has been described as level 2 by California authorities.

As for the study, only 50 vehicles in all constitute the U.S. fleet compared to 269 million conventional vehicles.

In sum, the record so far shows no red flags, more work is needed, and it is being done.