2020/03/17 by Daniel J. Fremont, Fremont, Daniel J., Edward Kim +17 · 3 citations
Computer Science · #D.2.4 #D.2.5 #FOS: Computer and information sciences #FOS: Electrical engineering #Formal Methods in Verification #I.2.9 #Logic in Computer Science (cs.LO) #Machine Learning (cs.LG) #Model-Driven Software Engineering Techniques #Software Engineering (cs.SE) #Software Testing and Debugging Techniques #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2003.07739
openalex publication_date 2020/03/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We present a new approach to automated scenario-based testing of the safety\nof autonomous vehicles, especially those using advanced artificial\nintelligence-based components, spanning both simulation-based evaluation as\nwell as testing in the real world. Our approach is based on formal methods,\ncombining formal specification of scenarios and safety properties, algorithmic\ntest case generation using formal simulation, test case selection for track\ntesting, executing test cases on the track, and analyzing the resulting data.\nExperiments with a real autonomous vehicle at an industrial testing facility\nsupport our hypotheses that (i) formal simulation can be effective at\nidentifying test cases to run on the track, and (ii) the gap between simulated\nand real worlds can be systematically evaluated and bridged.\n