Testing Rare Downstream Safety Violations via Upstream Adaptive Sampling of Perception Error Models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Testing black-box perceptual-control systems in simulation faces two difficulties. Firstly, perceptual inputs in simulation lack the fidelity of real-world sensor inputs. Secondly, for a reasonably accurate perception system, encountering a rare failure trajectory may require running infeasibly many simulations. This paper combines perception error models -- surrogates for a sensor-based detection system -- with state-dependent adaptive importance sampling. This allows us to efficiently assess the rare failure probabilities for real-world perceptual control systems within simulation. Our experiments with an autonomous braking system equipped with an RGB obstacle-detector show that our method can calculate accurate failure probabilities with an inexpensive number of simulations. Further, we show how choice of safety metric can influence the process of learning proposal distributions capable of reliably sampling high-probability failures.
Original languageEnglish
Title of host publication2023 IEEE International Conference on Robotics and Automation (ICRA)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages12744-12750
Number of pages7
ISBN (Electronic)9798350323658
ISBN (Print)9798350323665
DOIs
Publication statusPublished - 4 Jul 2023
Event2023 IEEE International Conference on Robotics and Automation - London, United Kingdom
Duration: 29 May 20232 Jun 2023
https://www.icra2023.org

Conference

Conference2023 IEEE International Conference on Robotics and Automation
Abbreviated titleICRA 2023
Country/TerritoryUnited Kingdom
CityLondon
Period29/05/232/06/23
Internet address

Fingerprint

Dive into the research topics of 'Testing Rare Downstream Safety Violations via Upstream Adaptive Sampling of Perception Error Models'. Together they form a unique fingerprint.

Cite this