宇宙ロボットの姿勢制御の一方法 : ニューラルネッ トによるアプローチ

A free-flying space robot consisting of a satellite main body and manipulator arms is expected to perform lmportant tasks in space. The attitude of the satellite changes when the manipulators are operated. This means the orientation angles would be controlled by an appropriate planning of the trajectory for the manipulators. For good robot performance, trajectory planning methods are required to control the orientation angles of the satellite. In this paper, a neural network is employed to generate the trajectory for the manipulators to control the orientation angles of the satellite and simultaneously to minimize the operational efforts. The proposed algorithm is applied to a free-flying robot with a simple two-link manipulator. By numerical simulation, the optimal trajectories are computed for any change of the orientation angles, and the effectiveness of the proposed algorithm is verified.