Fuzzy logic combined with neural algorithm to control industrial robot

Abstract

The problem of finding the optimal path for the robot arm is one of the most important problems of industrial robot. Problem consists when robot looking for specific routes that require the lowest power consumption. Path that established between any two end points, can follow many paths. All these paths require different amounts of energy depending on the distance, velocity and acceleration. Path planning for robotic arms have a several degrees of freedom. This problem is solved by using neuro-fuzzy techniques. Using analytical and numerical techniques is very difficult to find a good solution. Mathematically is more difficulty to move a robotic arm in the presence of obstacles, but child instinctively moving his hand in the presence of obstacles. A way that allows us to progress is a neuro-fuzzy fusion systems. Neural networks make the ability to learn, while Fuzzy logic is based on the emulation of thinking of an expert. In addition, as hardware technology advances, more and more value will be placed on solutions that can be used in parallel processing, such as neural networks and fuzzy logic with neural networks.