Sistema de controle inteligente para um braço robótico.

Abstract

This dissertation describes the project and the implementation of a System for Intelligent Control of Robotic Arms that can be used for tests of different intelligent control strategies and algorithms. The system is based on some paradigms of the Artificial Intelligence (AI), such as: artificial neural networks and fuzzy logic. The developed system was utilized for the positioning of an experimental robotic manipulator in Neural network's Laboratory (NEUROLAB) of the Department of Computer Systems of UFPB Campus II. The study of the problem of controlling a Robotic Arm to catch a stairway is emphasized with objective of demonstration of the several difficulties found by the robot and the corresponding strategies of intelligent control used to solve them.