The application of Li-function activated RNN to acceleration-level robots' kinematic control via time-varying matrix inversion

In this paper, the application of recurrent neural network (RNN) activated by Li function to acceleration-level robots' kinematic control via time-varying matrix inversion is presented and investigated. Specifically, by exploiting Li activation function and by computing the time-varying inverse of the related nonsingular matrix, the resultant RNN model is applied to kinematic control of redundant robot manipulators at the joint-acceleration level. Note that such a Li-function activated RNN (LFARNN) model can achieve the purpose of finite-time convergence, and thus is feasible to acceleration-level kinematic control of redundant robot manipulators. Simulation results based on a four-link planar robot manipulator and a PA10 robot manipulator further substantiate the effectiveness of the presented LFARNN model, as well as show the LFARNN application prospect.