论文部分内容阅读
针对未知环境下移动机器人自主探索和地图创建问题,在机器人操作系统的框架下,提出一种基于动态精简式混合地图的移动机器人自主探索方法.首先,提出一种基于几何规则的候选目标点生成方法,用于快速提取当前的前沿目标点;然后,从信息收益和路径成本的角度,引入一种改进的效用函数来评价候选目标点;最后,利用缓存增量式的原理优化拓扑节点,进而构建精简式混合地图.实验结果表明,通过拓扑图构建策略的改进,所提出方法具有良好的导航性能.
Aimed at the problems of autonomous exploration and map creation of mobile robot in unknown environment, this paper proposes a method of autonomous exploration of mobile robot based on dynamic concise hybrid map under the framework of robot operating system.Firstly, we propose a method based on geometric rules to generate candidate target points Method, which is used to quickly extract the current leading-edge target point. Then, from the perspective of information gain and path cost, an improved utility function is introduced to evaluate the candidate target point. Finally, the topological node is optimized by using the principle of cache increment, And build a condensed hybrid map.The experimental results show that the proposed method has good navigation performance through the improvement of the topology building strategy.