Augmenting Safety-Critical Driving Scenarios while Preserving Similarity to Expert Trajectories

TL;DR

This paper introduces a trajectory augmentation method for imitation learning that preserves similarity to expert data and enhances safety-critical scenario performance by clustering, transforming, and selectively adding trajectories.

Contribution

It presents a novel augmentation approach that maintains expert trajectory similarity and focuses on safety-critical clusters to improve imitation learning in safety-sensitive contexts.

Findings

Augmented trajectories improve closed-loop safety performance

Clustering identifies safety-critical trajectory groups

Transformation-based augmentation enhances imitation learning

Abstract

Trajectory augmentation serves as a means to mitigate distributional shift in imitation learning. However, imitating trajectories that inadequately represent the original expert data can result in undesirable behaviors, particularly in safety-critical scenarios. We propose a trajectory augmentation method designed to maintain similarity with expert trajectory data. To accomplish this, we first cluster trajectories to identify minority yet safety-critical groups. Then, we combine the trajectories within the same cluster through geometrical transformation to create new trajectories. These trajectories are then added to the training dataset, provided that they meet our specified safety-related criteria. Our experiments exhibit that training an imitation learning model using these augmented trajectories can significantly improve closed-loop performance.