Space manipulator trajectory tracking based on recursive decentralized finite-time control

In this paper, a novel decentralized control strategy with finite-time convergence is developed for the trajectory tracking of a space manipulator. The existing control schemes for a space manipulator can be classified into two main categories: centralized ones and decentralized ones. The former regards the robot as a high dimensional system with strong nonlinearity, thus the controllers are with multiple-input and multiple-output forms. The latter treats the system as a number of decoupled subsystems, each of which corresponds to a link of the manipulator. There are a set of decentralized single-input and single-output controllers for each subsystem. The centralized control has high performance and large calculation cost, whereas the decentralized control has relatively low control precision, but with high computational efficiency. This work combines the above two schemes to propose a compromised control strategy-recursivedecentralized finite-time control. The space manipulator is viewed as a set of interconnected subsystems. The tracking error equations of each subsystem are firstly written based on the recursive dynamic model of the system. Then the decentralized non-singular terminal sliding mode (NTSM) control law is designed to make the tracking error converge to an arbitrarily small neighborhood of the origin within a finite time. Finally, numerical simulations of the trajectory tracking task for a space manipulator are given to verify the effectiveness and efficiency of the control scheme. (C) 2020 Elsevier Masson SAS. All rights reserved.