Adaptive Stress Testing with Reward Augmentation for Autonomous Vehicle Validatio

Anthony Corso, Peter Du, Katherine Driggs-Campbell, Mykel J. Kochenderfer

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

Abstract

Determining possible failure scenarios is a critical step in the evaluation of autonomous vehicle systems. Real world vehicle testing is commonly employed for autonomous vehicle validation, but the costs and time requirements are high. Consequently, simulation driven methods such as Adaptive Stress Testing (AST) have been proposed to aid in validation. AST formulates the problem of finding the most likely failure scenarios as a Markov decision process, which can be solved using reinforcement learning. In practice, AST tends to find scenarios where failure is unavoidable and tends to repeatedly discover the same types of failures of a system. This work addresses these issues by encoding domain relevant information into the search procedure. With this modification, the AST method discovers a larger and more expressive subset of the failure space when compared to the original AST formulation. We show that our approach is able to identify useful failure scenarios of an autonomous vehicle policy.

Original languageEnglish (US)
Title of host publication2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages163-168
Number of pages6
ISBN (Electronic)9781538670248
DOIs
StatePublished - Oct 2019
Event2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019 - Auckland, New Zealand
Duration: Oct 27 2019Oct 30 2019

Publication series

Name2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019

Conference

Conference2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019
CountryNew Zealand
CityAuckland
Period10/27/1910/30/19

ASJC Scopus subject areas

  • Artificial Intelligence
  • Management Science and Operations Research
  • Instrumentation
  • Transportation

Fingerprint Dive into the research topics of 'Adaptive Stress Testing with Reward Augmentation for Autonomous Vehicle Validatio'. Together they form a unique fingerprint.

  • Cite this

    Corso, A., Du, P., Driggs-Campbell, K., & Kochenderfer, M. J. (2019). Adaptive Stress Testing with Reward Augmentation for Autonomous Vehicle Validatio. In 2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019 (pp. 163-168). [8917242] (2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITSC.2019.8917242