论文部分内容阅读
护理机器人在室内三维结构化环境下进行导航时,面临着三维建图计算量大且地图中缺乏语义信息的缺点。提出了基于点和平面特征的混合地图构建方法,结合点和平面在地图构建中的优势,并基于该混合地图搭建室内导航系统。首先快速地提取特征点和特征平面,使用解释树的方法进行数据关联,并使用平滑建图工具构建因子图,进行机器人位姿和路标的联合优化,改进并更新混合地图。然后,搭建室内导航系统,实现了三维障碍物检测、路径规划与运动控制。最后,在走廊环境下进行了室内导航实验,并以由激光雷达构建的二维栅格地图为参考,分析了地图构建效果和机器人定位精度,证明了基于混合地图的室内导航系统在室内结构化环境下的优势。
Nursing robots face the disadvantage of large computational load of 3D map and lack of semantic information in the map when navigating 3D indoor environment. This paper proposes a hybrid map construction method based on point and plane features, combines the advantages of point and plane in map construction, and builds an indoor navigation system based on the hybrid map. Firstly, the feature points and feature planes are extracted rapidly, the data is related by the method of interpretation tree, and the factor diagrams are constructed by smoothing drawing tools. The joint pose of robots and road signs are optimized and the hybrid maps are improved and updated. Then, build an indoor navigation system to achieve three-dimensional obstacle detection, path planning and motion control. Finally, the indoor navigation experiment was carried out in the corridor environment. Based on the two-dimensional grid map constructed by lidar, the effect of the map construction and the robot positioning accuracy were analyzed. It was proved that the indoor navigation system based on the hybrid map was structured indoors Environment advantage.