Adaptive speed planning for Unmanned Vehicle Based on Deep Reinforcement Learning

TL;DR

This paper introduces a deep reinforcement learning-based local navigation system for unmanned vehicles that improves obstacle avoidance and speed planning, reducing unnecessary deceleration and enhancing maneuverability.

Contribution

It develops a novel speed planning method using DQN and DDQN with improved reward functions for better obstacle adaptation in unmanned vehicle navigation.

Findings

Enhanced maneuvering capabilities without frequent deceleration

Effective obstacle avoidance in simulated environments

More stable and efficient path planning

Abstract

In order to solve the problem of frequent deceleration of unmanned vehicles when approaching obstacles, this article uses a Deep Q-Network (DQN) and its extension, the Double Deep Q-Network (DDQN), to develop a local navigation system that adapts to obstacles while maintaining optimal speed planning. By integrating improved reward functions and obstacle angle determination methods, the system demonstrates significant enhancements in maneuvering capabilities without frequent decelerations. Experiments conducted in simulated environments with varying obstacle densities confirm the effectiveness of the proposed method in achieving more stable and efficient path planning.