Abstract: Aiming at the characteristics of complex roadway environment and narrow road in mining area,a path planning optimization method of roadway inspection robot based on SLAM (Simultaneous Localization and Mapping) technique is proposed to realize the autonomous positioning and map construction of the robot. Firstly,a variety of sensors such as LiDAR and RGB-D cameras are fused to obtain 3D point cloud data of mining area,and SLAM algorithm is used to construct 3D map of mining area in real time. At the same time,through the registration of the currently obtained point cloud data and the built map,the robot can realize the autonomous positioning in the mining area. Secondly,an incremental A∗ optimization algorithm is proposed for path planning in view of the narrow,bending,branching and other complex environments in mines. Based on the traditional A∗ algorithm,the algorithm introduces optimization strategies such as path smoothing and corridor width constraints, which can generate smooth feasible paths that meet the constraints of complex mining environment. The algorithm is updated incrementally, and only needs to research the changed local area,which greatly reduces the calculation time of the overall path planning. The experimental results show that compared with the traditional route planning method,the algorithm can complete the inspection task more quickly and accurately,which provides a reference for the popularization and application of mine roadway inspection robot.

Key words: mine roadway inspection robot,SLAM technique,incremental A? optimization algorithm,path optimization