Geno-fuzzy control in autonomous servicing of a space station

The provision of spacecraft servicing to a space station comprises the tasks of assembly, resupply, repair and maintenance of manufactured space parts in-orbit. Servicing tasks use rendezvous and docking operations to accomplish their objectives. Autonomous servicing relies on intelligent fly control systems to perform fast, soft, and precise docking operations. This paper proposes the use of a fuzzy logic control system, mounted on an active chaser vehicle which wants to dock with a big passive space station in orbit around the Earth. The gene-fuzzy controller is a knowledge-based controller that performs the closed-loop operations autonomously. It produces smooth control actions in the proximity of the target and during the docking, to avoid disturbance torques in the final assembly orbit. The use of a genetic algorithm tool to optimise the controller so as to reduce docking time and fuel consumption is analyzed. It performs the optimization by finding the best fuzzy sets of the controller membership functions, and the most suitable rule data base. (C) 1998 Elsevier Science Ltd. All rights reserved.