A new path-planning algorithm for mobile robot based on neural network

Abstract

A new path-planning algorithm based on neural networks is proposed for mobile robots. The neural network is used in the algorithm to model the environment and calculate the collision energy function (CEF) which is the dominating term in the cost function. To implement the path-planning procedure, rather than calculating the minimum value of the cost function directly, a discrete method is used to approximate the minus gradient direction of the cost function in order to determine the motion tendency of the point set along the path. Finally, the performance and efficiency of the algorithm are estimated through computer simulation. The algorithm is very efficient in situations where real-time operation is required.