DYNAMIC COORDINATION OF A MULTI-DOF MANIPULATOR-PLATFORM SYSTEM

In this paper the performance of a novel reaction control method recently introduced by some of the authors is analyzed in the case of multi-degrees-of-freedom manipulator-platform systems. The proposed method locally minimizes the base reactions transferred by redundant manipulators to the base spacecraft by using simple least-squares real-time routines. The interesting performances that can be achieved with a high level of redundancy are analyzed in this paper in case of minimizing the base reaction torques, forces, or weighted combinations of them. The Zero Reaction Workspace in which a null reaction torque is possible has been computed for different space manipulators with increasing degrees of freedom, showing that it can be significantly increased by increasing the available degree of redundancy. This study has been carried out for free-floating and fixed-based manipulators, which are used as a reference, and both considering the robot physical joint limits or not. In addition, a study on the peak joint angles, velocities, and accelerations required to perform an end-effector trajectory inside the Zero Reaction Workspace by robot with increasing degrees of freedom has been carried out, using the classical pseudoinverse solution as a reference. Then, the Zero Reaction Workspaces in case of minimizing the base reaction torques, forces, or weighted combinations of them are compared for different robot degrees of freedom, with particular attention to their shape and area, both in the case of free-floating base and fixed-based robots. Finally, the study of the Zero Reaction Workspace for increasing number of desired null reaction components has been carried out for a 5-degrees-of-freedom manipulator.