Tests showed camera-based systems best
But both types were prone to being tricked by drivers
New testing from the American Automobile Association has found that driver-facing cameras are the best at making sure motorists are driving when they’re supposed to be driving. But even those safety technology systems aren’t foolproof, with the group saying that both camera-based and steering-wheel-based systems are “prone to being intentionally fooled.”
The number of driver assistance systems that take over some of the tasks of driving is becoming increasingly common. But, as AAA engineering and industry relations director Greg Brannon points out, “regardless of brand names or marketing claims, vehicles available for purchase today are not capable of driving themselves.” Brannon said that “vehicle technology has the potential to improve roadway safety, but the last thing we want are ineffective features in the hands of uninformed or overconfident drivers.”
Active driver assistance systems include features like radar cruise control, which can handle acceleration and braking to maintain a set speed and the flow of traffic. Also included are systems that center the vehicle in the lane with the driver’s hands on the wheel, including Tesla’s Autopilot and Hyundai’s Highway Driving Assist, and hands-off systems that require only that the driver watch the road in certain situations, including GM’s Super Cruise.
GM’s Super Cruise, Subaru’s EyeSight with Driver Focus, Hyundai’s Highway Driving Assist, and Tesla Autopilot were tested. The first two use a driver-facing camera, the latter sensors that monitor the steering wheel for driver input.
Three tests were conducted on a limited-access toll road, with AAA researchers spotting and driving. The first method tested a driver hands-off, head facing up, eyes looking down. The second was hands-off, head and eyes aimed at the center console, and the third saw drivers “attempting to “beat the system” through a variation of gaze/head placement and periodic steering wheel input.”
AAA said its findings showed camera-based systems alerted to disengaged driving 50 seconds sooner in the first method and 51 seconds in the second. Nearly an entire minute of driving sooner than wheel-based systems. It also found that steering wheel monitoring allowed more than five and a half minutes of distracted driving while cameras allowed just 2.25, all in a 10-minute test. Both systems could be tricked, without using external devices, AAA said.
“Regardless of brand names or marketing claims, vehicles available for purchase today are not capable of driving themselves,” said Brannon. “Driver monitoring systems are a good first step to preventing deadly crashes, but they are not foolproof.” The group recommends camera-based monitoring systems for safety, but with more refinement needed to prevent misuse.