Mustafa Ilhan Akbas
Scenario-based approaches are cost and time effective solutions to autonomous cyber-physical system testing to identify bugs before costly methods such as physical testing in a controlled or uncontrolled environment. Every bug in an autonomous cyber-physical system is a potential safety risk. This paper presents a scenario-based method for finding bugs and estimating boundaries of the bug profile. The method utilizes a nature-inspired approach adapting low discrepancy sampling with local search. Extensive simulations demonstrate the performance of the approach with various adaptations.
Goss, Quentin and Akbas, Mustafa
"A Nature-Inspired Approach for Scenario-Based Validation of Autonomous Systems,"
Beyond: Undergraduate Research Journal: Vol. 6
, Article 1.
Available at: https://commons.erau.edu/beyond/vol6/iss1/1