NTSB Report Reveals Uber Disabled Emergency Braking Before Fatal Crash
Federal investigators reveal that the Uber self-driving car involved in a fatal Arizona crash detects the pedestrian six seconds before impact but fails to brake because the company intentionally disables the vehicle's factory emergency braking system.
The National Transportation Safety Board releases a preliminary report revealing critical details about the fatal Uber self-driving crash in Tempe, Arizona. Investigators determine that the vehicle's sensors, including radar and laser instruments, detect a woman walking her bicycle across the road approximately six seconds before impact. Despite having ample time to stop a vehicle traveling at 43 mph, the autonomous system fails to prevent the collision.
The report explains that the software struggles to correctly identify the victim as she crosses the street. As the vehicle and pedestrian paths converge, the system cycles through different classifications, labeling her first as an unknown object, then as a vehicle, and finally as a bicycle. It takes until just 1.3 seconds before impact for the self-driving system to finally determine that an emergency braking maneuver is necessary to mitigate the crash.
Even after recognizing the need for emergency braking, the vehicle does not stop on its own because Uber intentionally disables the Volvo's factory-equipped automatic emergency braking function when its own self-driving system is in control. The human safety driver inside the car survives without injuries but only begins to apply the brakes less than a second before the crash, meaning the intervention occurs after the fatal impact already happens.