Shortest Paths for Satellite Mounted Robot Manipulators

Abstract

In ground based robotics the most fundamental form of path planning uses linear interpolation, either in joint space or in cartesian space. It is the purpose of this paper to extend this fundamental path planning method to the space based robotics problem. Space based robotics will often require the manipulation of loads that have a mass that is not negligible compared to that of the satellite on which the robot is mounted. And the manipulation will often be performed with the attitude control system turned off in order to save attitude control fuel. It is nevertheless necessary that the satellite attitude be returned to its original attitude by choice of the robot path. Direct application of linear interpolation fails in this situation, and the analog developed here is to find minimum arc length solutions satisfying the attitude condition. The minimum is in the chosen space, i.e. in joint space, cartesian space for satellite fixed coordinates, or cartesian space for inertial coordinates. This paper presents methods of path planning, and examples of such robot paths, for the second of these alternatives.