Aplicação de técnicas de controle e visão computacional em robótica móvel com controlador FUZZY E PID

Abstract

This undergraduate thesis presents the development of a line-following robot using digital image processing and the implementation of Fuzzy and PID controllers. The project’s objective is to build a system that enables the robot to autonomously follow a predefined trajectory using a camera sensor to capture images of the path. The implementation process involved the use of digital image processing techniques to identify the path and determine the robot’s position relative to it. Additionally, a Fuzzy controller and a proportional integral-derivative (PID) controller were incorporated to optimize the robot’s performance in trajectory correction. The project results demonstrate the effectiveness of the developed system, showcasing the robot’s ability to accurately follow the trajectory even under challenging lighting conditions and path variations. The comparative evaluation of Fuzzy and PID controllers, using Odometry as a metric, revealed superior performance by the Fuzzy controller. The Fuzzy controller achieved an 80% accuracy rate in trajectory following and the number of completed paths, while the PID controller achieved a 60% accuracy rate. The integration of digital image processing with Fuzzy and PID controllers provided a quick and efficient response of the robot to changes in the environment. The resulting robot from this work exhibited coherence with the initial proposal of applying Digital Image Processing (PDI) with Fuzzy and PID control, highlighting the efficiency of the Fuzzy controller compared to PID in this project. These successfully applied techniques can serve as a reference for future thesis projects.