HiddenGems: Efficient safety boundary detection with active learning

TL;DR

HiddenGems is an active learning-based method that efficiently identifies safety boundaries in autonomous system scenarios, reducing simulation costs significantly while enabling critical safety analysis.

Contribution

The paper introduces HiddenGems, a novel sample-efficient active learning approach for safety boundary detection in autonomous systems, outperforming traditional parameter sweeps in efficiency.

Findings

Achieved boundary estimates with 6x fewer simulations than parameter sweeps.

Detected and fixed a failure mode in an unprotected turn with 86% fewer simulations.

Effectively mapped compliance domains for multiple safety metrics.

Abstract

Evaluating safety performance in a resource-efficient way is crucial for the development of autonomous systems. Simulation of parameterized scenarios is a popular testing strategy but parameter sweeps can be prohibitively expensive. To address this, we propose HiddenGems: a sample-efficient method for discovering the boundary between compliant and non-compliant behavior via active learning. Given a parameterized scenario, one or more compliance metrics, and a simulation oracle, HiddenGems maps the compliant and non-compliant domains of the scenario. The methodology enables critical test case identification, comparative analysis of different versions of the system under test, as well as verification of design objectives. We evaluate HiddenGems on a scenario with a jaywalker crossing in front of an autonomous vehicle and obtain compliance boundary estimates for collision, lane keep, and acceleration metrics individually and in combination, with 6 times fewer simulations than a parameter sweep. We also show how HiddenGems can be used to detect and rectify a failure mode for an unprotected turn with 86% fewer simulations.