TY - GEN
T1 - STRETCH
T2 - 34th IEEE Intelligent Vehicles Symposium, IV 2023
AU - Scheuer, Franz
AU - Gambi, Alessio
AU - Arcaini, Paolo
N1 - Funding Information:
We thank Prof. Gordon Fraser (UniPassau) and Prof. Matthias Althoff and his group (TU Münich) for their support. This work was partially supported by ERATO HASUO Metamathematics for Systems Design Project (No. JPMJER1603), JST, Funding Reference number 10.13039/501100009024 ERATO; Engineerable AI Techniques for Practical Applications of High-Quality Machine Learning-based Systems Project (Grant Number JPMJMI20B8), JST-Mirai; and The EU Project Flexcrash (Grant Agreement n. 101069674).
Publisher Copyright:
© 2023 IEEE.
PY - 2023/7/27
Y1 - 2023/7/27
N2 - Collision avoidance systems are fundamental for autonomous driving and need to be tested thoroughly to check whether they safely handle critical scenarios. Testing collision avoidance systems is generally done by means of scenario-based testing using simulators and comes with the main challenge of generating situations that are realistic but avoidable. In other words, driving scenarios must stress the collision avoidance functionalities while being representative. Existing crash databases and accident reports describe observed accidents and enable to (re)create realistic collisions in simulations; however, as those data sources focus on the impact, their data do not generally lead to avoidable collision scenarios. To address this issue, we propose STRETCH, which generates realistic, critical, and avoidable collision scenarios by extending focused collision descriptions using a multi-objective optimization algorithm. Thanks to STRETCH, developers and testers can automatically generate challenging test cases based on realistic crash scenarios.
AB - Collision avoidance systems are fundamental for autonomous driving and need to be tested thoroughly to check whether they safely handle critical scenarios. Testing collision avoidance systems is generally done by means of scenario-based testing using simulators and comes with the main challenge of generating situations that are realistic but avoidable. In other words, driving scenarios must stress the collision avoidance functionalities while being representative. Existing crash databases and accident reports describe observed accidents and enable to (re)create realistic collisions in simulations; however, as those data sources focus on the impact, their data do not generally lead to avoidable collision scenarios. To address this issue, we propose STRETCH, which generates realistic, critical, and avoidable collision scenarios by extending focused collision descriptions using a multi-objective optimization algorithm. Thanks to STRETCH, developers and testers can automatically generate challenging test cases based on realistic crash scenarios.
KW - autonomous driving
KW - avoidable collisions
KW - collision avoidance systems
KW - reactive planner
KW - search-based testing
U2 - 10.1109/IV55152.2023.10186634
DO - 10.1109/IV55152.2023.10186634
M3 - Conference contribution
AN - SCOPUS:85167990259
T3 - IEEE Intelligent Vehicles Symposium, Proceedings
SP - 1
EP - 6
BT - IV 2023 - IEEE Intelligent Vehicles Symposium, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 4 June 2023 through 7 June 2023
ER -