A SLAM approach to generation a 3D map for a robot that performs electrical power distribution maintenance work using a laser scanner

摘要

Up until recently, power distribution maintenance work at very high altitude, and high-voltage has been carried out manually. These tasks are usually very dangerous, as they threaten the lives of the workers performing them. Therefore, we started from this premise to conceive a more secure way to perform these power distribution works: The development of a robot which can perform autonomous distribution operations. The purpose of this study is to construct a three-dimensional model of the work environment around an electric utility pole, save it so that it can be processed and used by the robot, as the robot can perform autonomously maintenance work. In this paper, we introduce a method to build a 3-D model of the robot's working environment by superimposing a group of many 3-D point clouds to the group that was obtained using the laser range sensor in the Kinect2. The aggregation of these point clouds is done by the ICP (Iterative Closest Point) algorithm.