Research papers · 2017.03
Accelerated Evaluation of Automated Vehicles Safety in Lane Change Scenarios Based on Importance Sampling Techniques
Accelerated Evaluation of Automated Vehicles Safety in Lane-Change Scenarios Based on Importance Sampling Techniques
Overview
The method biases sampling of naturalistic cut-in behavior while retaining probability corrections for the original distribution. Cross-entropy optimization tunes sampling parameters, enabling accelerated conflict, crash, and injury estimation for a modeled automated vehicle.
Author-supplied source information
Author-supplied information on arXiv.
submitted to IEEE ITSC
Version: arXiv v2 — manuscript revision