A novel stable and safe model predictive control framework for autonomous rendezvous and docking with a tumbling target. (November 2022)

• Record Type: Journal Article
• Authors: Dong, Kaikai; Luo, Jianjun; Limon, Daniel
• Is Part Of: Acta astronautica. Volume 200(2022)
• Page Start: 176
• Page End: 187
• Publication Date: 2022-11
• Subjects: Safe and stable-by-design -- Autonomous rendezvous and docking -- Tumbling target -- Time variant LOS region -- Deterministic convergent -- Novel MPC framework; Astronautics -- Periodicals; Outer space -- Exploration -- Periodicals; Astronautics; Periodicals; 629.405
• DOI: 10.1016/j.actaastro.2022.08.012
• Languages: English
• ISSNs: 0094-5765

Abstract

Abstract: This study proposes a novel, safe, and stable-by-design model predictive control (MPC) framework for multistage autonomous rendezvous and docking (AR&D) with a tumbling target, considering several practical challenges (e.g., control saturation, velocity constraints, and collision avoidance) in dock-enabling conditions. In the first stage, global near-optimal and deterministic convergent strategies are designed to drive the chaser to a time-varying line-of-sight (LOS) region within a few steps. The proposed controller includes a terminal constraint to ensure recursive feasibility and stability. In the second stage, the proposed controller is a periodic MPC for tracking under the novel framework, whose reference to be followed is the trajectory of the docking port of the tumbling target. This controller combines trajectory planning and control in a single layer, thereby improving the real-time performance of the algorithm. Moreover, the novel MPC framework incorporates a terminal constraint that guarantees that the closed-loop system enjoys recursive feasibility, safe evolution, and asymptotic convergence to the optimal admissible periodic trajectory with respect to the trajectory of the target docking port. The simulation results verified the efficiency of the proposed control strategy. Highlights: A novel, safe, and stable MPC framework for multistage AR&D with a tumbling target. In the first stage, a global near-optimal and deterministic convergent strategy is proposed. In the second stage, the controller unites trajectory planning and control in a single layer. The closed-loop system enjoys recursive feasibility, safe evolution, and asymptotic convergence.