Grupos neurais e sistemas nebulosos : aplicação a navegação autonoma

AUTOR(ES)
DATA DE PUBLICAÇÃO

1996

RESUMO

In this work a seIf organizing, neurofuzzy control architecture for a class of autonomous vehicles navigation is presented. The aim is to find target positions, without colliding with obstacles of an unknown environment. The architecture uses neural networks and fuzzy systems together with the theory of neuronal group seIection to learn navigation skills. Neurofuzzy sensor information builds up adaptive fieIds whose intensity triggers fuzzy control actions in response to the environment characteristics. That is, the control system deveIops adaptive behavior from the interactions between the vehicle, the environment,and Iearning strategies. Simulation results show that the control system is able to efficiently learn navigation strategies that perform better when compared with altemative schemes

ASSUNTO(S)

conjuntos difusos inteligencia artificial redes neurais (computação)

Documentos Relacionados