Research Article
Deep Reinforcement Learning for Real-Time Obstacle Avoidance and Local Path Replanning of Robots in Complex Dynamic Scenes
@INPROCEEDINGS{10.4108/eai.22-5-2026.2365334, author={Borui Zhang}, title={Deep Reinforcement Learning for Real-Time Obstacle Avoidance and Local Path Replanning of Robots in Complex Dynamic Scenes}, proceedings={Proceedings of the 4th International Conference on Image, Algorithms, and Artificial Intelligence, ICIAAI 2026, 22-24 May 2026, Singapore, Singapore}, publisher={EAI}, proceedings_a={ICIAAI}, year={2026}, month={8}, keywords={Local Path Planning Deep Reinforcement Learning Mobile Robot Navigation Autonomous Navigation}, doi={10.4108/eai.22-5-2026.2365334} }- Borui Zhang
Year: 2026
Deep Reinforcement Learning for Real-Time Obstacle Avoidance and Local Path Replanning of Robots in Complex Dynamic Scenes
ICIAAI
EAI
DOI: 10.4108/eai.22-5-2026.2365334
Abstract
In recent years, local path planning methods have been widely used in real scenes. With the development of deep reinforcement learning technology, the local path planning method combined with various reinforcement learning algorithms significantly improves the working efficiency of the robot. This paper briefly introduces the modeling method and basic control principle of local path planning problem, reviews the research of local path planning method combined with deep reinforcement learning in recent years, and introduces the local path planning method combined with safety reinforcement learning. Finally, the paper summarizes the research results in recent years and points out the opportunities and challenges in this field, which provides optimization ideas and development guidance for future research.
