Optimal motion planning for flexible space robots

Abstract

This paper is concerned with optimal motion planning of a flexible space robot. The robot is assumed to consist of two flexible links which are attached to a rigid space station floating in space. The optimal motion planning is first formulated as a two-stage functional optimization problem, which is further simplified into an optimal trajectory planning problem using recently developed stable inversion theory. The motion planning is optimal in the sense that the system performance measured by the manoeuvring time together with control and structural vibration energy is minimized. Besides, the controller also keeps the interference from the arm to the space station satisfactorily small. A suboptimal solution to the corresponding trajectory planning problem is obtained via two decoupling on the linearized zero dynamics. One is of the hyperbolic and the nonhyperbolic parts, and another is of the stable and unstable parts. Numerical examples are given to demonstrate the effectiveness of this approach.