Avaliação de desempenho de assistente de velocidade aplicado em AGV utilizando controle fuzzy

• Main Author: Silva, Debora Fernandes da
• Publication Date: 2022
• Other Authors: Santos, Marcio Luis Borges dos
• Format: Bachelor thesis
• Language: por
• Source: Repositório Institucional da UTFPR (da Universidade Tecnológica Federal do Paraná (RIUT))
• Download full: http://repositorio.utfpr.edu.br/jspui/handle/1/32485
• topic: Robôs; Controle eletrônico; Sistemas difusos; Velocidade - Controle automático; Robots; Electronic control; Fuzzy Systems; Speed - Automatic control; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::ELETRONICA INDUSTRIAL, SISTEMAS E CONTROLES ELETRONICOS

The proposed work seeks to evaluate the performance of a speed assistant developed in fuzzy logic and applied to autonomous line follower robots. The main objective is to avoid a collision between the robots, the speed assistant must be able to control the speed decreasing or increasing it according to the distance to the robot in front of it. The project was initially thought só that both the robot in front and the robot behind would have the same code and both with the speed assistant só that if they changed places the behavior would be the same. However, during the development of the fuzzy code, it required more processing, só it was necessary to exchange the Arduino Uno for the Arduino Mega, as the cost of the Arduino Mega is much higher than the Arduino Uno decided to implement the speed assistant in only one of the robots. In this way, the line follower robot that remains in the front remained with the Arduino Uno and the control is always done manually by an application developed in MIT Inventor, however the robot that is behind received the Arduino Mega and was adjusted to work with all hardware components and their speed is controlled by the speed assistant or manually by another application developed in MIT Inventor. There is also a third application that was used to collect data from the robot with a speed assistant that stores the distance and speed data in a google spreadsheet in the cloud. The work in general was successful, however, during the development several problems were found that had to be worked around during the project, from adaptations of the hardware that had already been designed, it hasn’t been tested until the exchange of Arduino for the lack of processing power. We then concluded that fuzzy met the premises as a speed assistant, but that perhaps using another intelligence the performance is better and that there is a need to improve the hardware, inserting new sensors that facilitate speed control or improving the arrangement of sensors existing. Another need is to review the issue of the battery, which at the beginning of the project was tried with a 9 V battery and then was changed to 3 rechargeable batteries 18650 3800 mAh.