环境特征提取在移动机器人导航中的应用

针对移动机器人在未知结构化环境中导航的需要,采用2D激光雷达作为主要传感器,对诸如墙壁、拐角、出口等这些典型的环境特征分别设计了一套有效的特征提取算法,并在该算法的基础上提出了基于特征点的移动机器人导航策略.该策略不需要里程计等其他一些内部传感器的信息,并且也不依赖具体的环境表述模型,从激光雷达扫描一次所得的数据中即可提取出环境特征,从而来指引机器人导航,实现起来快速可靠.应用到移动机器人MORCS-1上进行实验,取得了满意的结果,算法的实时性与鲁棒性得到了验证.

Environment Feature Extraction to Mobile Robot Navigation

Aim at the need that mobile robot navigates in unknown structured environments,based on 2D laser scanner,a set of algorithms of feature extraction are respectively designed for these typical environmental characteristics such as wall,corner and passageway etc..And the strategy of mobile robot navigation is proposed by the characteristic point on the basis of this algorithm.This strategy does not need the information of internal sensors,and not rely on the concrete environment model.Ranging data from the laser scanner in a sampling time is used to obtain the environment characteristics,and then these are adopted in mobile robot navigation,which can be realized fast and simply.With mobile robot MORCS-1 as test platform,the experimental results show that the real-time demand of robot navigation can be met and the effectiveness of the algorithm can be validated.