OBJECTIVE: The objective of this investigation was to evaluate the interaction between an SAE level 2 automated vehicle and the driver, including the limitations imposed by the vehicle on the driver. METHODS: A case study of the first fatal crash involving a vehicle operating with an automated control system was performed using scene evidence, vehicle damage, and recorded data available from the vehicle, and information from both drivers, including experience, phone records, computer systems, and medical information, was reviewed. RESULTS: System performance data downloaded from the car indicated that the driver was operating it using the Traffic-Aware Cruise Control and Autosteer lane-keeping systems, which are automated vehicle control systems within Tesla's Autopilot suite. As the car crested the hill, a tractor trailer began its left turn onto a crossing roadway. Although reconstruction of the crash determined that there was sufficient sight distance for both drivers to see each other and take action, neither responded to the circumstances leading to the collision. Further, based on the speeds of the vehicles and simulations of the truck's path, the car driver had at least 10.4 s to detect the truck and take evasive action. Neither the car driver nor the Autopilot system changed the vehicle's velocity. At the time of the crash, the system performance data indicated that the last driver interaction with the system was 1 min 51 s prior when the cruise control speed was set to 74 mph. The driver was operating the vehicle using the Autopilot system for 37 of the 41 min in the last trip. During this period, the vehicle detected the driver's hands on the steering wheel for a total of 25 s; each time his hands were detected on the wheel was preceded by a visual alert or auditory warning. CONCLUSIONS: The National Transportation Safety Board (NTSB) determined that the probable cause of the Williston, Florida, crash was the truck driver's failure to yield the right of way to the car, combined with the car driver's inattention due to overreliance on vehicle automation, which resulted in the car driver's lack of reaction to the presence of the truck. Contributing to the car driver's overreliance on the vehicle automation was the car's operational design, which permitted the driver's prolonged disengagement from the driving task and his use of the automation in ways inconsistent with guidance and warnings from the manufacturer.
OBJECTIVE: The objective of this investigation was to evaluate the interaction between an SAE level 2 automated vehicle and the driver, including the limitations imposed by the vehicle on the driver. METHODS: A case study of the first fatal crash involving a vehicle operating with an automated control system was performed using scene evidence, vehicle damage, and recorded data available from the vehicle, and information from both drivers, including experience, phone records, computer systems, and medical information, was reviewed. RESULTS: System performance data downloaded from the car indicated that the driver was operating it using the Traffic-Aware Cruise Control and Autosteer lane-keeping systems, which are automated vehicle control systems within Tesla's Autopilot suite. As the car crested the hill, a tractor trailer began its left turn onto a crossing roadway. Although reconstruction of the crash determined that there was sufficient sight distance for both drivers to see each other and take action, neither responded to the circumstances leading to the collision. Further, based on the speeds of the vehicles and simulations of the truck's path, the car driver had at least 10.4 s to detect the truck and take evasive action. Neither the car driver nor the Autopilot system changed the vehicle's velocity. At the time of the crash, the system performance data indicated that the last driver interaction with the system was 1 min 51 s prior when the cruise control speed was set to 74 mph. The driver was operating the vehicle using the Autopilot system for 37 of the 41 min in the last trip. During this period, the vehicle detected the driver's hands on the steering wheel for a total of 25 s; each time his hands were detected on the wheel was preceded by a visual alert or auditory warning. CONCLUSIONS: The National Transportation Safety Board (NTSB) determined that the probable cause of the Williston, Florida, crash was the truck driver's failure to yield the right of way to the car, combined with the car driver's inattention due to overreliance on vehicle automation, which resulted in the car driver's lack of reaction to the presence of the truck. Contributing to the car driver's overreliance on the vehicle automation was the car's operational design, which permitted the driver's prolonged disengagement from the driving task and his use of the automation in ways inconsistent with guidance and warnings from the manufacturer.
Entities:
Keywords:
Automated vehicle; autopilot; crash; recorded data