Safety benchmarking of automated driving systems

Current Automated Driving Systems (ADS) have reached the point of fully driverless deployments with passengers, so questions are being asked such as: “Are ADS safer than humans?" and "Are ADS safe enough?" But answering these questions requires solutions to two critical challenges:

1. What metric(s) should we use for assessment?

2. What human benchmark(s) should we be comparing to?

This talk will cover the use and validation of surrogate metrics as well as the challenges of apples-to-apples comparison to human benchmarks.