회전 표적에 대한 모델 예측 제어 기반 랑데부 및 도킹

초록

Spacecraft proximate operations including rendezvous and docking requires several essential technologies such as space robotics, inter-satellite communication, autonomous on-board operations, and guidance and control. Especially, for close-range rendezvous, it requires optimal guidance accounting for safety and approaching velocity constraints as well as autonomous operations independent of ground stations. This paper proposes a rendezvous guidance and control method based on Model Predictive Control(MPC) successfully implemented in several platforms including cars, robots, and aircraft. This work assumes a non-cooperative target rotating at a constant speed, and designs an optimal problem considering several constraints like docking port approaching direction. Three dynamics models with different sets of state variables are introduced, and then the selected model expands to a quadratic programming problem by adding a quadratic cost function in terms of the squared impulse sum as well as linearized constraints. In addition, a well-known glideslope guidance is implemented as a feedback control logic beyond the control window in the MPC framework. Numerical results demonstrate the performance of the proposed method for a tumbling target rendezvous case.

키워드