| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.88 MB | Adobe PDF |
Abstract(s)
O setor agrícola é reconhecido pela sua elevada exigência em termos de recursos e mão-de-obra. Assim, muitos agricultores têm vindo a adotar cada vez mais tecnologia e a automação como solução para este desafio. No entanto, os robôs agrícolas existentes são frequentemente complexos, lentos e dispendiosos, o que dificulta a sua adoção generalizada. Como resultado, o setor agrícola tem ficado para trás na integração de tecnologias mais avançadas, existindo algum atraso na transformação digital possível no sector. É importante observar que, apesar dos avanços nas pesquisas e na implementação de robôs agrícolas, a maioria das investigações e robôs existentes está focada na cultura de vinhas de uvas para produção de vinho. Poucos estudos abordaram adequadamente a colheita de uvas de mesa. Como tal, este trabalho de investigação estudou e desenvolveu o conceito/arquitetura de um robô agrícola projetado para a colheita de uvas de mesa em campos agrícolas, bem como para a monitorização geral das culturas. Neste trabalho, o robô proposto foi construído num software de simulação robótica e foi criado um modelo de simulação semelhante a uma exploração agrícola. A simulação robótica foi escolhida devido à sua rapidez, facilidade de implementação e capacidade de testar diferentes cenários e hipóteses de operação de forma simultânea, sem necessidade de construção de protótipos. Além disso, esta oferece liberdade e criatividade, sem preocupações com danos no hardware e custos reduzidos de desenvolvimento. No caso de estudo da colheita de uvas, e dos vários cenários estudados, o cenário 2 foi o que representou maiores dificuldades pois os objetos eram diferentes, tanto a nível de peso como de cores. A maior complicação foi na recolha dos objetos maiores, tendo sido necessário aumentar ligeiramente o ângulo de abertura das pinças da garra. No entanto, os resultados de simulação permitiram comprovar a funcionalidade do conceito desenvolvido.
The agricultural sector is renowned for its high demands on resources and labour. As a result, many farmers are increasingly adopting technology and automation as a solution to this challenge. However, existing agricultural robots are often complex, slow and expensive, which hinders their widespread adoption. As a result, the agricultural sector has lagged in integrating more advanced technologies, and there is some delay in the digital transformation possible in the sector. It is important to note that, despite advances in research and the implementation of agricultural robots, most of the existing research and robots are focused on the cultivation of grape vines for wine production. Few studies have adequately addressed the harvesting of table grapes. As such, this research work studied and developed the concept/architecture of an agricultural robot designed for harvesting table grapes in agricultural fields, as well as for general crop monitoring. In this work, the proposed robot was built using robotic simulation software and a simulation model similar to a farm was created. Robotic simulation was chosen because of its speed, ease of implementation and ability to test different scenarios and operating hypotheses simultaneously, without the need to build prototypes. It also offers freedom and creativity, without worrying about hardware damage and reduced development costs. In the case study of the grape harvest, and of the various scenarios studied, scenario 2 was the most difficult because the objects were different, both in terms of weight and color. The biggest complication was picking up the larger objects, which required slightly increasing the opening angle of the grapple's grippers. However, the simulation results proved the functionality of the concept developed.
The agricultural sector is renowned for its high demands on resources and labour. As a result, many farmers are increasingly adopting technology and automation as a solution to this challenge. However, existing agricultural robots are often complex, slow and expensive, which hinders their widespread adoption. As a result, the agricultural sector has lagged in integrating more advanced technologies, and there is some delay in the digital transformation possible in the sector. It is important to note that, despite advances in research and the implementation of agricultural robots, most of the existing research and robots are focused on the cultivation of grape vines for wine production. Few studies have adequately addressed the harvesting of table grapes. As such, this research work studied and developed the concept/architecture of an agricultural robot designed for harvesting table grapes in agricultural fields, as well as for general crop monitoring. In this work, the proposed robot was built using robotic simulation software and a simulation model similar to a farm was created. Robotic simulation was chosen because of its speed, ease of implementation and ability to test different scenarios and operating hypotheses simultaneously, without the need to build prototypes. It also offers freedom and creativity, without worrying about hardware damage and reduced development costs. In the case study of the grape harvest, and of the various scenarios studied, scenario 2 was the most difficult because the objects were different, both in terms of weight and color. The biggest complication was picking up the larger objects, which required slightly increasing the opening angle of the grapple's grippers. However, the simulation results proved the functionality of the concept developed.
Description
Trabalho de projeto para obtenção do grau de Mestre em Engenharia Mecânica
Keywords
Robótica Agricultura de precisão Colheita de uvas Simulação Robotics Precision agriculture Grape harvesting
Citation
Aires, Filipa Gomes Leite Sempiterno- Projeto de um robô móvel para tarefas de apoio à vindima. Lisboa: Instituto Superior de Engenharia de Lisboa, 2023. Dissertação de Mestrado