Optimal control of under-actuated redundant space-robot system
欠驱动冗余度空间机器人优化控制

The underactuated control is an important aspect of the fault tolerance technology in space field. In this paper the optimal motion planning and control method are studied for the underactuated redundant space robot system with passive joints equipped with lockers. The optimizing ability and control method for such a system are analyzed on the basis of dynamic equations. A coupling index between the actuated and passive joints is introduced and an optimizing control algorithm is proposed, which is called the virtual model lead control method (VMLCM). By this method, a fully actuated manipulator that structurally is equivalent to the underactuated one is used as the virtual model to plan the motion, and the underactuated manipulator approximates the planned trajectory in the space of joint. Therefore, the algorithm can drive the manipulator to track a desired trace and improve the tracking accuracy. A three DOFs planar robot with passive end joint is used for simulation, and the simulation results justify the proposed algorithm.