Eye-in-Hand Tracking Control of a Free-Floating Space Manipulator

This paper studies the visual-tracking (visual-servo) control problem for a free-floating space manipulator using an eye-in-hand camera for the capturing process, where the motion of the target spacecraft is unknown and the dynamics of the system are uncertain. A depth-independent matrix is developed to help design an observer for estimating the unknown positions of target points on the spacecraft in the inertial frame. The target points are used to estimate depth information for visual servo control. An adaptive controller is used to force the feature points on the target to converge onto the desired positions on the image plane as well as estimate the uncertain dynamic parameters. The Lyapunov approach is used to show asymptotic stability of the system and convergence of image errors. Finally, simulation results are presented to validate the performance of the proposed control algorithm.