Validity Learning on Failures: Mitigating the Distribution Shift in Autonomous Vehicle Planning

TL;DR

This paper introduces Validity Learning on Failures (VL(on failure)), a novel approach to improve autonomous vehicle planning by learning from failure cases without expert annotations, effectively mitigating distribution shift issues.

Contribution

The paper proposes a new failure-based learning method that leverages a pre-trained planner to identify and learn from mistakes, enhancing autonomous driving performance.

Findings

VL(on failure) significantly improves closed-loop metrics.

The method outperforms state-of-the-art approaches on Bench2Drive.

Experimental results show increased success rate and progress.

Abstract

The planning problem constitutes a fundamental aspect of the autonomous driving framework. Recent strides in representation learning have empowered vehicles to comprehend their surrounding environments, thereby facilitating the integration of learning-based planning strategies. Among these approaches, Imitation Learning stands out due to its notable training efficiency. However, traditional Imitation Learning methodologies encounter challenges associated with the co-variate shift phenomenon. We propose Validity Learning on Failures, VL(on failure), as a remedy to address this issue. The essence of our method lies in deploying a pre-trained planner across diverse scenarios. Instances where the planner deviates from its immediate objectives, such as maintaining a safe distance from obstacles or adhering to traffic rules, are flagged as failures. The states corresponding to these failures are compiled into a new dataset, termed the failure dataset. Notably, the absence of expert annotations for this data precludes the applicability of standard imitation learning approaches. To facilitate learning from the closed-loop mistakes, we introduce the VL objective which aims to discern valid trajectories within the current environmental context. Experimental evaluations conducted on both reactive CARLA simulation and non-reactive log-replay simulations reveal substantial enhancements in closed-loop metrics such as \textit{Score, Progress}, and Success Rate, underscoring the effectiveness of the proposed methodology. Further evaluations against the Bench2Drive benchmark demonstrate that VL(on failure) outperforms the state-of-the-art methods by a large margin.