Image-based visual servoing for power transmission line inspection robot

This paper describes a vision-based system for a powerline inspection robot. The main emphasis of this paper is on an image-based visual servoing strategy for driving the robot arm toward a desired configuration in order to grasp the line. The method skips the step of transferring image features back to robot pose, and hence makes driving arm motion plans directly in the image plane, providing a 'visual' trajectory in the image plane for the gripper to track and locate the phase line and related obstacles in an effort to make the robot body to align with the phase line. A proportional control law with a dead zone is given to achieve the driving wheel and powerline alignment operation. Experimental results using an inspection robot platform are presented, showing that good servoing can be achieved using the proposed controller of the vision system.