Discovering Avoidable Planner Failures of Autonomous Vehicles using Counterfactual Analysis in Behaviorally Diverse Simulation

TL;DR

This paper presents a novel simulation-based framework for testing autonomous vehicle planners by identifying avoidable failures through counterfactual analysis in diverse traffic scenarios, enhancing safety validation.

Contribution

It introduces a new method for systematically discovering avoidable planner failures using behaviorally diverse simulation and large-scale search techniques.

Findings

Successfully identified a wide range of critical planner failures.

Distinguished between unavoidable and avoidable accidents.

Demonstrated effectiveness in complex multi-agent intersection scenarios.

Abstract

Automated Vehicles require exhaustive testing in simulation to detect as many safety-critical failures as possible before deployment on public roads. In this work, we focus on the core decision-making component of autonomous robots: their planning algorithm. We introduce a planner testing framework that leverages recent progress in simulating behaviorally diverse traffic participants. Using large scale search, we generate, detect, and characterize dynamic scenarios leading to collisions. In particular, we propose methods to distinguish between unavoidable and avoidable accidents, focusing especially on automatically finding planner-specific defects that must be corrected before deployment. Through experiments in complex multi-agent intersection scenarios, we show that our method can indeed find a wide range of critical planner failures.