Mobile robot guidance using cellular neural networks and fuzzy logic

Abstract

We show how a Cellular Neural Networks based im­age processing system together with a Tuzzy Logic controller are capable of providing the necessary signal processing to guide an autonomous mobile robot in a maze drawn on the fl.oor. In this way, a non-trivial navigation task is obtained by very sim­ple hardware, making real autonomous operation feasible. An autonomous line-following robot has been first simulated and is currently being imple­roented by simulating the CNN with a DSP, while the fuzzy algorithms run on a 386-microprocessor ­based microcontroller.