OPTIMAL KINEMATIC CONTROL OF REDUNDANT SPACE ROBOTIC SYSTEMS FOR ORBITAL MAINTENANCE: SIMULATED MICROGRAVITY TESTS

This paper presents the theoretical formulation and the experimental validation of an innovative algorithm for the kinematic inversion of redundant space robotic systems, which causes minimum attitude disturbances on the spacecrafts on which they are mounted. Different authors face the problem by exploiting the minimization of spacecraft attitude disturbances by means of the null space of the system Jacobian matrix. In this paper an original approach is proposed: closed loop differential kinematics has been used at acceleration level, and an additional constraint has been imposed in order to minimize the torque transferred to the spacecraft center of mass due to the robotic arm movement. This problem results to be a constrained linear least squares problem and a closed form solution has been proposed for non singular trajectories. This algorithm leads to the minimum of torque transmitted and results to have better performances with respect to the other algorithms in literature. An extension of this algorithm has been presented for singular trajectories, in order to limit the joints accelerations under acceptable values. In this case the problem results to be a constrained linear least squares problem with inequality constraints, and need iterative calculations to be solved. The proposed algorithm has been experimentally tested using a 3D free-flying robot previously tested in an ESA Parabolic Flight Campaign. In this test campaign the 3D robot has been converted in a 2D robot taking advantage of its modular structure, and it has been suspended by means of air-bearings on a granite plane. In this way it is possible to perform simulated microgravity tests without time constraints. The base of the robot has been fixed on ground by means of a custom designed dynamometer, which measures the torque transferred to the ground to be minimized. The experimental results validated the proposed algorithm and confirmed its good performance, and motivate the ongoing work for the extension of the algorithm to a 3D free-flying robot.