Huei Peng, a researcher at the University of Michigan, commented that “Even the most advanced and largest-scale efforts to test automated vehicles today fall woefully short of what is needed to thoroughly test these robotic cars,”. To solve this problem, he and his colleague Ding Zhao have developed a new four-step accelerated approach to test autonomous vehicles.
Outlining the approach in their paper ‘From the Lab to the Street: Solving the Challenge of Accelerating Automated Vehicle Testing’, it’s claimed that their accelerated testing approach eliminates up to 99.9 percent of the cost and time in compiling enough data to achieve confidence that the tested vehicle is 90 percent safer than cars driven by human drivers on the road today.
Interestingly, both of the most common serious crash scenarios identified in the testing methodology involved a human driven car as well as an automated car. It seems to be that we need to promote AV uptake as much as possible to ensure the safest possible road conditions.
Become a Member
Enter our draw for a chance to test drive the StreetDrone ONE and spend a day with the SD team! Additionally, get acces to exclusive technical content, discounted events and conferences