Vision-Based Tracking and Trajectory Generation for Robotic Capture of Objects in Space

Abstract

In this paper, we describe the development of vision-based algorithms for determining the motion of a moving object and generating trajectory for a robotic manipulator to intercept the object. The ultimate goal of this work is to develop autonomous algorithms for robotic grasping of objects in space. This problem arises in several applications, including on-orbit servicing of satellites and removal of space debris. The proposed methods have been implemented and tested in simulation and are currently being implemented on an experimental facility. The facility is based on a novel concept for experimental evaluation of robotic capture of free-floating objects, in particular, to use a small helium airship to emulate a free-floating object. The paper presents a brief overview of the main components of the facility, describes the vision-based motion estimation, trajectory generation and redundancy resolution algorithms implemented thus far and presents simulation results demonstrating the performance of these algorithms.