Towards Affordable On-track Testing for Autonomous Vehicle - A Kriging-based Statistical Approach

摘要

This paper discusses the use of Kriging model in Automated Vehicleevaluation. We explore how a Kriging model can help reduce the number ofexperiments or simulations in the Accelerated Evaluation procedure. We alsopropose an adaptive sampling scheme for selecting samples to construct theKriging model. Application examples in the lane change scenario are presentedto illustrate the proposed methods.