Path Planning For Satellite Mounted Robots

Abstract

One of the most important problems in space based robotics is the disturbance to the satellite attitude and to the satellite microgravity environment caused by satellite mounted robot operation. This paper reports on the development of algorithms for optimal path planning that minimizes these disturbances, and solve the inverse kinetics problem, i.e. with the satellite attitude control system off. Specific optimality criteria are studied, including minimum induced angular velocity of the satellite and minimizing the maximum acceleration of the satellite center of mass. In addition, the space based analog is generated for the common ground based linear interpolation in joint or Cartesian space, i.e. shortest distance paths. Some properties of the various types of optimal paths are developed analytically, and understanding of the typical nature of the optimal paths is obtained in numerical examples. The shortest paths in principal planes, seen in inertial space, are shown to be arcs of circles. The computation required to produce such optimized trajectories is 1 or 2 minutes on a workstation, and methods can be used to substantially decrease this number if necessary. Thus, it can be practical to make use of these optimized path plans in space.