1P1-T02 Using LRF, input work environment for power distribution line maintenance(Robot Vision)

Abstract

We are developing a semi-autonomous robot system for power distribution line maintenance works. The robot system autonomously performs primary tasks, for example "Insert the bolt" to reduce the load on operators. To perform the tasks autonomously, the robot system has to construct database of work environment which is a set of positions and orientations of the objects in work environments around electric poles. So we applied ICP (Iterative Closest Points) algorithm to the point cloud measured by LRF(Laser Range Finder) to estimate the positions and orientations of the objects. Also, we are using the Kd-tree in order to perform at a high speed processing.