University of Michigan White Paper Aims to Build Accelerated Evaluation Processes for Autonomous Vehicles
The University of Michigan recently published a white paper aimed at accelerating the development of autonomous vehicles through more efficient testing processes. The paper identifies a significant problem in getting autonomous vehicles on the road: how to develop tests to accurately represent real-world driving situations? The approach suggested by the paper essentially focuses on “break[ing] down difficult real-world driving situations into components that can be tested or simulated repeatedly.” According to the paper, the suggested approach will cut the time to evaluate safety by 300 to 100,000 times for autonomous vehicles.