Implementação de uma rede neural artificial em ambiente simulado para o controle de trajetória de um robô móvel com a arquitetura Ackermann

Abstract

Advances in the field of artificial intelligence in the last decade have provided one of the most interesting times so far in the field of robotics, especially in mobile robotics. One of the factors that contributed to this phenomenon is the complexity involved in the control of the mobile robot during the autonomous navigation of more complex situations, which requires a controller that can generalize and have a good success rate in unprecedented situations. The present work aims to implement, in a simulated environment, an artificial neural network as the controller of the mobile robot used in the NXP Cup. From the data gathered from the simulated experiments it can be concluded that the trained artificial neural network was able to control the mobile robot in an unknown track without letting the mobile robot leave the route, requiring few adjustments of its synaptic weights. Implementação De Uma Rede Neural Artificial Em Ambiente Simulado Para O Controle De Trajetória De Um Robô Móvel Com A Arquitetura Ackermann.