The National Highway Traffic Safety Administration has launched an investigation into two Tesla crashes, each involving a motorcyclist, revealing a potentially harrowing hazard: the partially automated vehicles may not stop for motorcycles.
The crashes had several things in common. Each occurred in the dark, likely involved the use of the partially automated driver-assist system, occurred when the motorcyclist was struck from behind, and was fatal for the motorcyclist. The drivers of the Tesla's did not see the motorcyclists and collided with the rear of the motorcycles causing the motorcycle drivers to be ejected from their vehicles.
The NHTSA says that once it gathers more information, it may include the motorcycle crashes in a broader probe into Tesla's ramming into emergency vehicles parked along freeways. The issue stems from the "autopilot" and "full self-driving" misnomers assigned to the vehicles.1 Despite these names, the vehicles cannot drive themselves and should not be trusted to safely maneuver roads without human supervision. Since its release, "Autopilot" has been misused by Tesla drivers, who have been caught riding while napping, drunk, and even riding in the back seat while the car drove itself down the road, making timely intervention impossible.
Previously, Tesla used radar systems that were triggered by signals to stop the vehicles. Some of these signals were "false positives" which would stop the cars after determining that overpasses were obstacles. Tesla vehicles now use a network of cameras and thousands of images that the computer neural network uses to determine if there are objects in the way of the vehicle.
- Tesla is selecting car owners to test "Full Self-Driving" software, which is designed to complete a route on its own with human supervision. The software is expected to be available to all Tesla owners who request it by the end of 2022.
Editor's Note: The issue seems to arise when the vehicles are traveling too fast for fully stopped obstacles to register, or during dark hours when it may be more difficult to capture images of other vehicles. Tesla has said that even their "Full Self-Driving" cars cannot drive themselves, and that drivers should be ready and able to intervene at all times.