Control of Free-floating Space Robotic Manipulators base on Neural Network

Abstract

For problems of trajectory tracking of the free-floating space robot model with uncertainties in task space, neural networks adaptive control strategy is put forward by this paper. Because the non-linear system model can not be obtained accurately, neural network is used to directly identify all parts of the system parameters through GL matrix and its multiplication operator “· ”. Robust controller is designed to eliminate the approximation errors of neural network and external disturbances. The control strategy neither requires an estimate of inverse dynamic model, nor calculates the inverse Jacobin matrix. Global asymptotic stability based on Lyapunov theory is proved by the paper. Simulation results show that higher control precision is achieved. The control strategy has great value in engineering applications.