Inverse dynamics control of a parallel robot based on RBF neural network

Abstract

For the issue that the parallel robot system is difficult to track the trajectory accurately due to the uncertainty, this paper proposes an inverse dynamics control method of parallel robot based on RBF neural network. The control method makes full use of the prior knowledge of the dynamic model to track and control. At the same time, the RBF neural network is used to approximate the nonlinear function, to realize the fast approximation and compensation of the uncertain part of the system. In addition, through the Lyapunov stability theory, an adaptive algorithm us designed to adjust the weight of RBF neural network. This method eliminates the influence of uncertain factors and other nonlinear factors on the control process, so that the system can obtain accurate dynamic control performance.