Jointly Learnable Behavior and Trajectory Planning for Self-Driving Vehicles

TL;DR

This paper introduces a jointly learnable behavior and trajectory planning framework for self-driving cars, improving safety and human-likeness by integrating decision-making and trajectory generation into a unified, interpretable model.

Contribution

It proposes a novel joint learning approach with an interpretable cost function that unifies behavior and trajectory planning in self-driving vehicles.

Findings

Jointly learned planner outperforms baselines in safety metrics.

Improves similarity to human driving behaviors.

Demonstrates effectiveness on real-world data.

Abstract

The motion planners used in self-driving vehicles need to generate trajectories that are safe, comfortable, and obey the traffic rules. This is usually achieved by two modules: behavior planner, which handles high-level decisions and produces a coarse trajectory, and trajectory planner that generates a smooth, feasible trajectory for the duration of the planning horizon. These planners, however, are typically developed separately, and changes in the behavior planner might affect the trajectory planner in unexpected ways. Furthermore, the final trajectory outputted by the trajectory planner might differ significantly from the one generated by the behavior planner, as they do not share the same objective. In this paper, we propose a jointly learnable behavior and trajectory planner. Unlike most existing learnable motion planners that address either only behavior planning, or use an uninterpretable neural network to represent the entire logic from sensors to driving commands, our approach features an interpretable cost function on top of perception, prediction and vehicle dynamics, and a joint learning algorithm that learns a shared cost function employed by our behavior and trajectory components. Experiments on real-world self-driving data demonstrate that jointly learned planner performs significantly better in terms of both similarity to human driving and other safety metrics, compared to baselines that do not adopt joint behavior and trajectory learning.