Neural-Network-Based Terminal Sliding Mode Control of Space Robot Actuated by Control Moment Gyros

This paper studies the trajectory tracking control of a space robot system (SRS) in presence lumped uncertainties with no prior knowledge their upper bound. Although some related methods have been proposed, most them either not applied to SRSs or lack rigorous stability proof. Therefore, it is still challenge achieve high accuracy and theoretical proof for SRSs. proposes new integrated neural network- based scheme SRS actuated by moment gyros (CMGs). A adaptive non-singular terminal sliding mode (ANTSM) method developed on an improved radial basis function network (RBFNN). In method, weight update law proposed learn bound uncertainties. With advantages RBFNN ANTSM, controller has accuracy, fast learning speed finite-time convergence. Different from on-ground robotic manipulator controllers, kinematic position attitude laws also designed satellite platform remain stable. The closed-loop proved Lyapunov mathematical standard. Comparative simulation results demonstrate effectiveness preferable performance robustness.