面向地图构建的移动机器人局部路径自主规划
Active local path planning of mobile robots for map construction

移动机器人在未知场景中规划路径以自主完成定位与地图构建是机器人领域的一个重要研究课题. 本文阐述了一种利用实时构建的信息熵地图动态生成机器人的局部探索路径, 并综合转向约束和避障约束设计了一种基于模糊评价方法的方向选择策略跟踪生成的局部路径并进行环境构图. 与现有方法相比, 本文方法能够根据环境动态地生成平滑连续的局部探索路径, 并能引导机器人进行障碍物躲避和完成自主构图. 实验结果表明相较对比方法, 本文方法的探索路程最短,观测覆盖度最高, 同时整个自主构图过程所需的时间也更短.
The path planning problem for active simultaneous localization and mapping (SLAM) in unknown environment is a very important research subject in robotics. In this paper, we propose a novel method to generate local path for robotic exploration based on the information entropy map constructed in real time, and our method also designs a direction selection strategy based on fuzzy evaluation method combined with steering constraint and obstacle avoidance constraint to guide the robot to track the generated local path and mapping. Compared with the existing methods, the algorithm can dynamically generate smooth and continuous local exploration path according to the environment, and can guide robot obstacle avoidance and autonomous map construction. The experimental results show that our method has the shortest path and the highest environmental coverage, and time consumption for the whole active SLAM process is shorter than the comparison method.