Continual Driving Policy Optimization with Closed-Loop Individualized Curricula

arXiv:2309.14209v3 Announce Type: replace
Abstract: The safety of autonomous vehicles (AV) has been a long-standing top concern, stemming from the absence of rare and safety-critical scenarios in the long-tail naturalistic driving distribution. To tackle this challenge, a surge of research in scenario-based autonomous driving has emerged, with a focus on generating high-risk driving scenarios and applying them to conduct safety-critical testing of AV models. However, limited work has been explored on the reuse of these extensive scenarios to iteratively improve …

abstract arxiv autonomous autonomous driving autonomous vehicles challenge continual cs.ai cs.lg cs.ro distribution driving focus loop optimization policy research risk safety safety-critical stemming type vehicles