机器人局部环境特征提取方法的研究

Research on Feature Extraction of Robot Local Environment

考虑到机器人导航过程中的实时性与可靠性要求,充分利用激光雷达信息的快速性与精确性,针对结构化环境中诸如墙壁、拐角、通道等这些典型环境特征分别设计了一套快速有效地特征提取的算法;另外算法还考虑到机器人建立环境地图的需要,在环境特征提取时对某一些密集的障碍物进行了合并,并注意保留了环境特征的一些拓扑信息,为建图作了一些前期准备,拿到移动机器人MORCS上进行实验获得了满意的结果,验证了算法的实时性与可靠性。

Considering real-time character and reliability requirement in robot navigation process, fully utilizing tast and accuracy of laser radar information, a set of algorithms of feature extraction are respectively designed for these typical environmental characteristics in structured environment such as wall, corner and passageway etc. In addition still considered the need of setting up environmental map of the robot, carried out amalgamation to some intensive barriers when the environmental characteristics is drawn in algorithm, paying attention to keeping some topological information of the environmental characteristics, in order to build map in some earlier stages. Took the algorithm to test on the mobile robot MORCS, the satisfactory experimental results has proved the real-time character and validity of the algorithm.