A method of improved RBF neural network for robot trajectory planning

In order to solve the problems of slow convergence rate and low learning accuracy in general RBF neural network learning algorithm for robot trajectory planning,a hybrid learning algorithm is designed in this paper.It uses nearest neighbor clustering algorithm to determine the structure and parameter of network.The network parameter is adjusted by adding adaptive learning factor in learning rate to motivate the convergence speed.MATLAB tool is...