clock menu more-arrow no yes

Filed under:

Tesla’s Autopilot lulled driver into a state of ‘inattention’ in 2018 freeway crash

New, 38 comments

The National Transportation Safety Board weighs in

Illustration by Alex Castro / The Verge

The driver of a Tesla Model S who crashed into a fire truck on a California freeway last year was not paying attention to the road thanks to “overreliance” on Autopilot, the National Transportation Safety Board said Wednesday. In addition, investigators determined that another probable cause for the crash was the design of Autopilot, “which permitted the driver to disengage from the driving task.”

NTSB also cited the driver’s “lack of response to the fire truck parked in his lane” and “use of the system in ways inconsistent with guidance and warnings from Tesla” as among the probable causes for the crash.

No one was injured, but the unidentified driver’s use of Autopilot has brought regulator and media attention to the incident. There have been a number of reports of Tesla owners who were using Autopilot at the time of a crash, as well as a handful of people who have been killed while using Autopilot. Tesla has consistently said that drivers who use Autopilot are safer than those who don’t.

The crash took place January 22nd, 2018, when a Tesla Model S P85 smashed into a Culver City Fire Department truck that was parked diagonally across the southbound high-occupancy vehicle lane of Interstate 405. The fire truck was responding to a vehicle collision in the northbound lane.

NTSB graphic

The Tesla was traveling in the HOV lane behind another vehicle, but when that vehicle changed lanes to the right, the Tesla accelerated and struck the rear of the fire truck at a recorded speed of about 31 mph. Autopilot had been engaged for a total of 29 minutes and four seconds prior to the crash, but the driver’s hands were only detected on the steering wheel for 78 seconds of that time. Autopilot issued “several” hands-off alerts during the last 13 minutes before the crash.

“For most of the time the system was engaged, it did not detect driver-applied steering wheel torque (hands on the steering wheel),” NTSB states.

After the lead vehicle changed lanes, the Model S began to accelerate back up to its adaptive cruise control speed of 80 mph. Autopilot issued a forward collision warning alert 0.49 seconds prior to impact, but “the automatic emergency braking system did not engage,” NTSB concludes.

Autopilot is a Level 2 semi-autonomous system that combines adaptive cruise control, lane keep assist, self-parking, and, most recently, the ability to automatically change lanes. It uses a suite of sensors, including eight cameras, radar, and ultrasonic.

Car safety experts note that adaptive cruise control systems like Autopilot rely mostly on radar to avoid hitting other vehicles on the road. Radar is good at detecting moving objects, but not stationary objects. It also has difficulty detecting objects like a vehicle crossing the road not moving in the car’s direction of travel.

In the past, Tesla CEO Elon Musk has blamed crashes involving Autopilot on driver overconfidence. “When there is a serious accident it is almost always, in fact maybe always, the case that it is an experienced user, and the issue is more one of complacency,” Musk said last year. This seems to backup the NTSB conclusion that driver “inattention and overreliance” were at play in the January 2018 crash.

In a statement, a Tesla spokesperson notes that while Autopilot will turn off when a driver repeatedly ignores warnings to remain engaged with the driving, the automaker is continuing to roll out updates to make its advanced driver assist system “smarter, safer and more effective.”

“Since this incident occurred,” the spokesperson added, “we have made updates to our system including adjusting the time intervals between hands-on warnings and the conditions under which they’re activated.”