Safety Metric Aware Trajectory Repairing for Automated Driving

TL;DR

The paper tackles autonomous driving under dynamic disturbances by reframing evasion as a safety-metric–aware trajectory repair rather than full replanning. It introduces the Feasible Time-to-React ($F$-$TTR$) and a binary-search framework to iteratively deform and refine a B-spline trajectory, ensuring collision-free, dynamically feasible repairs that largely preserve the original plan. A two-phase optimization (trajectory deformation and refinement) integrates smoothness, feasibility, and collision penalties, and its performance is benchmarked against a baseline Quadratic Programming Trajectory Repairing (QPTR) on CommonRoad scenarios, where QPTR can become infeasible when operating near constraints. The results demonstrate that the proposed method yields executable, safety-compliant repairs with competitive computation times, and highlight practical implications for real-time AV planning and potential extensions to robotics. Overall, the work provides a principled, integrated approach to safety-aware trajectory repair with clear paths for performance and applicability enhancements.

Abstract

Recent analyses highlight challenges in autonomous vehicle technologies, particularly failures in decision-making under dynamic or emergency conditions. Traditional automated driving systems recalculate the entire trajectory in a changing environment. Instead, a novel approach retains valid trajectory segments, minimizing the need for complete replanning and reducing changes to the original plan. This work introduces a trajectory repairing framework that calculates a feasible evasive trajectory while computing the Feasible Time-to-React (F-TTR), balancing the maintenance of the original plan with safety assurance. The framework employs a binary search algorithm to iteratively create repaired trajectories, guaranteeing both the safety and feasibility of the trajectory repairing result. In contrast to earlier approaches that separated the calculation of safety metrics from trajectory repairing, which resulted in unsuccessful plans for evasive maneuvers, our work has the anytime capability to provide both a Feasible Time-to-React and an evasive trajectory for further execution.

Figures (8)

• Figure 1: Trajectory repairing while computing safety metrics.
• Figure 2: Illustration of the bicycle model.
• Figure 3: Illustration of configuration space
• Figure 4: Flowchart of the proposed trajectory repairing framework
• Figure 5: Illustration of the trajectory deformation and refinement using B-spline