Estudo e implementação de controladores fuzzy e pid para controle de direção e velocidade de um agv com visão computacional

Abstract

The decrease in manufacturing costs of electronics components, the emergence of new technologies and research areas, and the need for cheaper labor have been driving the construction of mobile robots and favoring their spread in several sectors, particularly in the industrial sector. Among mobile robots, Automated Guided Vehicles (AGV) are gaining ground in the industrial sector by lowering costs and increasing productivity. AGVs are robots used to move equipment and parts to the factory floor and do not require human supervision. The development of a Fuzzy controller in conjunction with Proportional-Integral-Derivative (PID) controllers, the use of a digital camera and the application of computer vision techniques can be used to assist AGVs in the displacements, avoiding collisions and divergences in the waypoints. Based on the facts presented, the objective of this work is to study and implement a control and speed control system based on Fuzzy and PID controllers for an experimental mobile robot. In this robot will be used a digital camera and computer vision techniques to detect paths to be followed. The methodology consists in developing an algorithm that captures the images and, based on image processing techniques, delivers an image segmented by binarization that allows finding the center of the track to be followed. This information is provided in a Fuzzy controller that determines the direction of the robot by varying the speed of each wheel. PID controllers are used to ensure the individual speed of the motors and thus the wheels. Finally, basic notions of odometry are used to recreate the path traveled by the robot and thus compare with the real path. In this way, the Fuzzy controller in conjunction with the PID controllers were able to control the robot during the entire trajectory, favoring its alignment with the center of the track