Interpretable and Flexible Target-Conditioned Neural Planners For Autonomous Vehicles

TL;DR

This paper introduces an interpretable neural planning model for autonomous vehicles that predicts multiple potential goals using heatmaps, improving safety and flexibility in real-world driving scenarios.

Contribution

The authors propose a novel heatmap-based neural planner with adaptive kernels and specialized loss functions to capture multiple goals and enhance collision avoidance.

Findings

Achieves safer driving performance than prior methods

Demonstrates flexibility in handling diverse scenarios

Effectively models multiple potential goals

Abstract

Learning-based approaches to autonomous vehicle planners have the potential to scale to many complicated real-world driving scenarios by leveraging huge amounts of driver demonstrations. However, prior work only learns to estimate a single planning trajectory, while there may be multiple acceptable plans in real-world scenarios. To solve the problem, we propose an interpretable neural planner to regress a heatmap, which effectively represents multiple potential goals in the bird's-eye view of an autonomous vehicle. The planner employs an adaptive Gaussian kernel and relaxed hourglass loss to better capture the uncertainty of planning problems. We also use a negative Gaussian kernel to add supervision to the heatmap regression, enabling the model to learn collision avoidance effectively. Our systematic evaluation on the Lyft Open Dataset across a diverse range of real-world driving scenarios shows that our model achieves a safer and more flexible driving performance than prior works.