摘 要
轮式移动机器人在未知环境中的自主探索与运动规划是智能机器人领域的重要研究方向。随着机器人技术的发展，如何使轮式移动机器人在复杂环境中实现高效、安全的自主探索成为亟待解决的问题。为此，本文针对轮式移动机器人自主探索与运动规划展开研究，旨在提高其在未知环境中的适应能力。研究中提出了一种融合多源传感器信息的环境感知方法，通过激光雷达和视觉传感器获取环境数据，并采用改进的SLAM算法构建环境地图，解决了传统方法中单一传感器信息不足的问题。同时，设计了基于深度强化学习的自主探索策略，该策略能够根据环境动态变化调整探索路径，提高了探索效率。此外，针对运动规划问题，提出了一种结合A*算法与人工势场法的混合运动规划算法，在保证路径最优性的同时避免了局部极小值问题。实验结果表明，所提出的环境感知方法能准确构建环境地图，自主探索策略可有效提高探索效率，混合运动规划算法能在复杂环境中规划出安全可行的路径。本研究为轮式移动机器人在未知环境中的自主作业提供了新的思路与方法，具有重要的理论意义和应用价值。
关键词：轮式移动机器人 自主探索 运动规划
Abstract
Autonomous exploration and motion planning of wheeled mobile robots in unknown environments represent a critical research direction in the field of intelligent robotics. As robotic technology advances, achieving efficient and safe autonomous exploration in complex environments has become an urgent challenge. This study focuses on enhancing the adaptability of wheeled mobile robots in unknown environments through improved autonomous exploration and motion planning. A multi-sensor information fusion approach for environmental perception is proposed, integrating data from LiDAR and visual sensors, and employing an enhanced SLAM algorithm to construct environmental maps, thereby addressing the limitations of single-sensor information in traditional methods. Additionally, a deep reinforcement learning-based autonomous exploration strategy is designed, which dynamically adjusts exploration paths according to environmental changes, thus improving exploration efficiency. For motion planning issues, a hybrid motion planning algorithm combining the A* algorithm with artificial potential field method is introduced, ensuring optimal path planning while avoiding local minima problems. Experimental results demonstrate that the proposed environmental perception method accurately constructs environmental maps, the autonomous exploration strategy effectively enhances exploration efficiency, and the hybrid motion planning algorithm successfully plans safe and feasible paths in complex environments. This research provides new insights and methodologies for autonomous operations of wheeled mobile robots in unknown environments, offering significant theoretical implications and practical value.
Keyword:Wheeled Mobile Robot Autonomous Exploration Motion Planning
目 录
1绪论 1
1.1研究背景与意义 1
1.2国内外研究现状 1
1.3本文研究方法 2
2轮式移动机器人环境感知技术 2
2.1传感器配置与融合 2
2.2环境建模方法 3
2.3障碍物检测与识别 3
3自主探索策略研究 4
3.1探索算法设计 4
3.2未知区域评估 4
3.3动态环境适应 5
4运动规划方法分析 6
4.1路径规划算法 6
4.2实时避障策略 6
4.3规划效率优化 7
结论 7
参考文献 9
致谢 10
