Aplicação de inteligência artificial para prototipação de veículos não tripulados de baixo custo
The ability of machines to learn today is to apply in the market for autonomous vehicles, by large technology companies, either for urban locomotion or for work on factory floors. One of the most used slopes for the development of the indicators is the use of neural networks that convolve depths tha...
| Autor principal: | França, Marcos Tiago Araújo de |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/12502 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The ability of machines to learn today is to apply in the market for autonomous vehicles, by large technology companies, either for urban locomotion or for work on factory floors. One of the most used slopes for the development of the indicators is the use of neural networks that convolve depths that represent an evolution, considering the capacity of expression of the image elements that are used to detect the locomotion or the detection of objects like a pedestrian . One of the greatest difficulties for the production of these equipments is its high computational and financial cost, therefore, the job is a low cost utility, with the feasibility of producing a simple, already functional standalone vehicle. You are using kits such as the Raspberry PI and the Lego Mindstorms Kit for building in conjunction with the image processing libraries, it was possible to install a set of primary exercises and apply this data in an Artificial Neural Network, based on this set. acuracy of 40 %. Based on results, the capacity of the data capture module is greater or greater. |
|---|