Path planning method for multi-robots using a cellular neural network

Abstract

This paper proposes a new method of path planning for multiple autonomous robots. A cellular neural network is used. This comprises a resistive grid network based on analog dynamics, a competitive network for detecting a maximum and a digital network for performing path searches. A computer simulation result is presented. The algorithm for the proposed method is essentially an extension of the local current comparison method. The local current comparison method is closely related to node-neighborhood analysis. It is a method that is well suited to dedicated hardware implementation using a hybrid analog-digital chip. In other words, the proposed method is considered to be suited to realization of a dedicated hardware engine for path planning. The basic operation of the hardware engine is founded on analog dynamics and very fast path planning can be executed. It is expected that a system can be constructed that will be able to plan the paths for multiple robots moving with high speed. © 1998 Scripta Technica. Electron Comm Jpn Pt 3, 81(3): 10–21, 1998