Dynamics and Control of Spacecraft Rendezvous By Nonlinear Model Predictive Control

This doctoral research investigates the fundamental problems in the dynamics and control of spacecraft rendezvous with a non-cooperative tumbling target. New control schemes based on nonlinear model predictive control method have been developed and validated experimentally by ground-based air-bearing satellite simulators. It is focused on the autonomous rendezvous for a chaser spacecraft to approach the target in the final rendezvous stage. Two challenges have been identified and investigated in this stage: the mathematical modeling of the targets tumbling motion and the constrained control scheme that is solvable in an on-line manner. First, the mathematical description of the tumbling motion of the target spacecraft is proposed for the chaser spacecraft to rendezvous with the target. In the meantime, the practical constraints are formulated to ensure the safety and avoid collision during the final approaching stage. This set of constraints are integrated into the trajectory planning problem as a constrained optimization problem. Second, the nonlinear model predictive control is proposed to generate the feedback control commands by iteratively solving an open-loop discrete-time nonlinear optimal control problem at each sampling instant. The proposed control scheme is validated both theoretically and experimentally by a custom-built spacecraft simulator floating on a high-accuracy granite table. Computer software for electronic hardware for the spacecraft simulator and for the controller is designed and developed in house. The experimental results demonstrate the effectiveness and advantages of the proposed nonlinear model predictive control scheme in a hardware-in-the-loop environment. Furthermore, a preliminary outlook is given for future extension of the spacecraft simulator with consideration of the robotic arms.