Generating Critical Scenarios for Testing Automated Driving Systems

Trung-Hieu Nguyen, Truong-Giang Vuong, Hong-Nam Duong, Son Nguyen, Hieu Dinh Vo · 2024

Read Generating Critical Scenarios for Testing Automated Driving Systems by Trung-Hieu Nguyen, Truong-Giang Vuong, Hong-Nam Duong, Son Nguyen, Hieu Dinh Vo (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-02574.txt.

Description

Trung-Hieu Nguyen and co-authors train a learning agent to invent the road situations most likely to crash a self-driving system in this 2024 paper. Real crash testing is slow and dangerous, so simulators carry the load, but random scenarios rarely find the edge cases that matter.

Their system reads both the car's inner state and the world's outer state, from weather to traffic. A reinforcement learning agent then turns the dials, weather, roads, and other drivers, hunting for setups that end in collision while staying believable under written safety rules. On two maps with four road layouts it produced thirty to over a hundred percent more collision scenarios than the leading prior method. Against random search it did up to nearly four times better. The authors offer it as a sharper crash-test dummy for virtual roads.

Catalog record

CREATORTrung-Hieu Nguyen, Truong-Giang Vuong, Hong-Nam Duong, Son Nguyen, Hieu Dinh Vo
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE87 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-02574

Browse the full library · About the Tower of Babel