I Know You Can't See Me: Dynamic Occlusion-Aware Safety Validation of Strategic Planners for Autonomous Vehicles Using Hypergames

TL;DR

This paper introduces a hypergame-based risk measure and a scenario-based validation framework to improve safety assessment of autonomous vehicle planners under dynamic occlusion, achieving significant speedups and broader coverage.

Contribution

It presents a novel hypergame theory approach for dynamic occlusion risk assessment and a white-box validation framework that accelerates safety testing for AV strategic planners.

Findings

Achieves 4000% speedup in safety validation process

Provides more diverse and comprehensive occlusion scenarios

Can generate common occlusion crash scenarios automatically

Abstract

A particular challenge for both autonomous and human driving is dealing with risk associated with dynamic occlusion, i.e., occlusion caused by other vehicles in traffic. Based on the theory of hypergames, we develop a novel multi-agent dynamic occlusion risk (DOR) measure for assessing situational risk in dynamic occlusion scenarios. Furthermore, we present a white-box, scenario-based, accelerated safety validation framework for assessing safety of strategic planners in AV. Based on evaluation over a large naturalistic database, our proposed validation method achieves a 4000% speedup compared to direct validation on naturalistic data, a more diverse coverage, and ability to generalize beyond the dataset and generate commonly observed dynamic occlusion crashes in traffic in an automated manner.