Trajectory Planning for Autonomous Vehicles Using Hierarchical Reinforcement Learning

TL;DR

This paper introduces a hierarchical reinforcement learning framework combined with a PID controller for autonomous vehicle trajectory planning, improving safety, robustness, and efficiency in dynamic environments.

Contribution

The paper presents a novel HRL-based trajectory planning method with sub-goals and LSTM integration, enhancing generalization, safety, and computational efficiency over existing approaches.

Findings

Reduces convergence time in trajectory planning.

Generates smoother and safer trajectories.

Handles noisy perception and dynamic environments effectively.

Abstract

Planning safe trajectories under uncertain and dynamic conditions makes the autonomous driving problem significantly complex. Current sampling-based methods such as Rapidly Exploring Random Trees (RRTs) are not ideal for this problem because of the high computational cost. Supervised learning methods such as Imitation Learning lack generalization and safety guarantees. To address these problems and in order to ensure a robust framework, we propose a Hierarchical Reinforcement Learning (HRL) structure combined with a Proportional-Integral-Derivative (PID) controller for trajectory planning. HRL helps divide the task of autonomous vehicle driving into sub-goals and supports the network to learn policies for both high-level options and low-level trajectory planner choices. The introduction of sub-goals decreases convergence time and enables the policies learned to be reused for other scenarios. In addition, the proposed planner is made robust by guaranteeing smooth trajectories and by handling the noisy perception system of the ego-car. The PID controller is used for tracking the waypoints, which ensures smooth trajectories and reduces jerk. The problem of incomplete observations is handled by using a Long-Short-Term-Memory (LSTM) layer in the network. Results from the high-fidelity CARLA simulator indicate that the proposed method reduces convergence time, generates smoother trajectories, and is able to handle dynamic surroundings and noisy observations.