MAER-Nav: Bidirectional Motion Learning Through Mirror-Augmented Experience Replay for Robot Navigation

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Wang, Shanze, Tan, Mingao, Yang, Zhibo, Huang, Biao, Shen, Xiaoyu, Huang, Hailong, Zhang, Wei
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Robotics
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

_version_	1866913768277016576
author	Wang, Shanze Tan, Mingao Yang, Zhibo Huang, Biao Shen, Xiaoyu Huang, Hailong Zhang, Wei
author_facet	Wang, Shanze Tan, Mingao Yang, Zhibo Huang, Biao Shen, Xiaoyu Huang, Hailong Zhang, Wei
contents	Deep Reinforcement Learning (DRL) based navigation methods have demonstrated promising results for mobile robots, but suffer from limited action flexibility in confined spaces. Conventional DRL approaches predominantly learn forward-motion policies, causing robots to become trapped in complex environments where backward maneuvers are necessary for recovery. This paper presents MAER-Nav (Mirror-Augmented Experience Replay for Robot Navigation), a novel framework that enables bidirectional motion learning without requiring explicit failure-driven hindsight experience replay or reward function modifications. Our approach integrates a mirror-augmented experience replay mechanism with curriculum learning to generate synthetic backward navigation experiences from successful trajectories. Experimental results in both simulation and real-world environments demonstrate that MAER-Nav significantly outperforms state-of-the-art methods while maintaining strong forward navigation capabilities. The framework effectively bridges the gap between the comprehensive action space utilization of traditional planning methods and the environmental adaptability of learning-based approaches, enabling robust navigation in scenarios where conventional DRL methods consistently fail.
format	Preprint
id	arxiv_https___arxiv_org_abs_2503_23908
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	MAER-Nav: Bidirectional Motion Learning Through Mirror-Augmented Experience Replay for Robot Navigation Wang, Shanze Tan, Mingao Yang, Zhibo Huang, Biao Shen, Xiaoyu Huang, Hailong Zhang, Wei Robotics Deep Reinforcement Learning (DRL) based navigation methods have demonstrated promising results for mobile robots, but suffer from limited action flexibility in confined spaces. Conventional DRL approaches predominantly learn forward-motion policies, causing robots to become trapped in complex environments where backward maneuvers are necessary for recovery. This paper presents MAER-Nav (Mirror-Augmented Experience Replay for Robot Navigation), a novel framework that enables bidirectional motion learning without requiring explicit failure-driven hindsight experience replay or reward function modifications. Our approach integrates a mirror-augmented experience replay mechanism with curriculum learning to generate synthetic backward navigation experiences from successful trajectories. Experimental results in both simulation and real-world environments demonstrate that MAER-Nav significantly outperforms state-of-the-art methods while maintaining strong forward navigation capabilities. The framework effectively bridges the gap between the comprehensive action space utilization of traditional planning methods and the environmental adaptability of learning-based approaches, enabling robust navigation in scenarios where conventional DRL methods consistently fail.
title	MAER-Nav: Bidirectional Motion Learning Through Mirror-Augmented Experience Replay for Robot Navigation
topic	Robotics
url	https://arxiv.org/abs/2503.23908