Stackelberg Driver Model for Continual Policy Improvement in Scenario-Based Closed-Loop Autonomous Driving
This addresses the challenge of continual policy improvement for autonomous vehicles in safety-critical scenarios, though it appears incremental as it builds on adversarial generation methods.
The paper tackles the problem of autonomous vehicle (AV) performance being hindered by rare corner cases in driving scenarios by proposing the Stackelberg Driver Model (SDM), which uses a leader-follower game to iteratively improve AV policies and generate challenging scenarios, resulting in superior performance compared to baselines in higher-dimensional scenarios.
The deployment of autonomous vehicles (AVs) has faced hurdles due to the dominance of rare but critical corner cases within the long-tail distribution of driving scenarios, which negatively affects their overall performance. To address this challenge, adversarial generation methods have emerged as a class of efficient approaches to synthesize safety-critical scenarios for AV testing. However, these generated scenarios are often underutilized for AV training, resulting in the potential for continual AV policy improvement remaining untapped, along with a deficiency in the closed-loop design needed to achieve it. Therefore, we tailor the Stackelberg Driver Model (SDM) to accurately characterize the hierarchical nature of vehicle interaction dynamics, facilitating iterative improvement by engaging background vehicles (BVs) and AV in a sequential game-like interaction paradigm. With AV acting as the leader and BVs as followers, this leader-follower modeling ensures that AV would consistently refine its policy, always taking into account the additional information that BVs play the best response to challenge AV. Extensive experiments have shown that our algorithm exhibits superior performance compared to several baselines especially in higher dimensional scenarios, leading to substantial advancements in AV capabilities while continually generating progressively challenging scenarios. Code is available at https://github.com/BlueCat-de/SDM.