Position control of a two-link satellite tracker manipulator using optical flow

Abstract

In this work, the objective was to implement visual servo control based on the optical flow method to move a robotic manipulator end-effector in order to provide on-orbit servicing of space satellites. The visual servo system is made up of a single monocular camera mounted on the end-effector, and is to be used by the robot for autonomous location of the satellite. Firstly, using the pyramidal Lucas-Kanade method, optical flow field is extracted from image sequences captured by the camera. The pixel coordinates corresponding to the targeted location in the image frame is marked and registered into the system. The marked pixel coordinates is then converted into the robotic manipulator workspace coordinates. These coordinates are then used to compute the manipulator joint angles. Using the camera to send visual feedback information in closed loop control, the task is repeated until the target is reached