A Reinforcement Learning-Driven Algorithm for Rapid Path Replanning of Robot Navigation in Indoor Uncertain Discrete Environments
May 1, 2025·
,·
1 min read
H. Min
LI ZHOUJIAN
W. Chi
Image credit: UnsplashAbstract
This paper presents a reinforcement learning-driven algorithm for rapid path replanning in robot navigation within indoor uncertain discrete environments. The proposed approach enhances traditional path planning methods by incorporating adaptive learning mechanisms that enable robots to quickly respond to dynamic environmental changes and uncertainties.
Type
Publication
In 2025 37th Chinese Control and Decision Conference (CCDC), Xiamen, China, 2025, pp. 3998-4003
Publication Details
DOI: 10.1109/CCDC65474.2025.11090489
Conference: 2025 37th Chinese Control and Decision Conference (CCDC)
ResearchGate: View full publication
Abstract
This research addresses the challenging problem of robot navigation in indoor uncertain discrete environments by proposing a reinforcement learning-driven algorithm for rapid path replanning. The algorithm integrates adaptive learning mechanisms that enable autonomous robots to efficiently respond to dynamic environmental changes, obstacles, and uncertainties commonly encountered in indoor settings.
Key Contributions
- Development of a novel reinforcement learning framework for dynamic path replanning
- Enhanced adaptability to uncertain and changing indoor environments
- Improved computational efficiency for real-time navigation applications
- Comprehensive evaluation in discrete indoor environment scenarios
Click the Cite button above to import publication metadata into your reference management software.
