SPACECRAFT RENDEZVOUS AND DOCKING. AN AUTONOMY ASSISTED HUMAN OPERATOR APPROACH

摘要

The phenomena and problems encountered when a rendezvous maneuver, and possible docking, of two spacecrafts has to be performed, have been the topic for numerous studies and details of a variety of scenarios has been analyzed. So far, all solutions that have been brought into realization have been based entirely on direct human supervision and control. This paper describes a vision based system and methodology, that autonomously generates accurate guidance information that may assist a human operator in performing the tasks associated with both the rendezvous and docking navigation procedures. The method described generates, based on a single camera and a priory information about the target vehicle and orbit data, all necessary guidance information for closed loop autonomous navigation, from first detection at far distance to a close up hold point. Furthermore, the system provides relative pose information to assist the human operator during the docking phase. The closed loop and operator assistance performance of the system have been assessed using a test bench including human operator, navigation module and high fidelity visualization module. The tests performed verified the general accuracy, real timeliness and robustness of the system. Especially the pose information proved to be valuable for the human operator, and allows for implementation of automatic safety features such as proximity fuse and target rotation rate synchronization.