A new prediction algorithm to improve training the neural networks and its application in mobile robot control system

This paper proposes a new prediction model for stable control of mobile robot based on chaotic neural networks. Programming mobile robots can be long and difficult task. In this study, we intend to demonstrate the chaotic learning algorithm to improve neural networks' learning efficiency and obtain better prediction. In order to validate the prediction performance of recurrent neural networks, a novel stimulation study and analysis paradigm has been done on the practical data. Finally, through computer simulations, we demonstrate the effectiveness and stability of the proposed controller according to changed working conditions.

1. introduction

In recent years, artificial Neural Networks (ANNs) have shown great promise in modeling time-series, nonlinear prediction, and the analysis of underlying features of data. They are typically designed to learn complicated functions [1]. The back propagation learning algorithm can effectively revise the weights and thresholds of hidden nodes [2], [3].