Implementação e análise de mecanismos de negociação em um sistema multiagente aplicado a alocação de vagas em um estacionamento inteligente
Searching for solutions to problems related to slow traffic, innovative projects develop smart parkings willing to save drivers time and free the traffic in busy streets. MAPS project looks for solutions to these problems by the approach of multi-agent systems to create a smart parking that automate...
| Autor principal: | Marini Filho, Angelo Bittencourt |
|---|---|
| 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/1/15964 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Searching for solutions to problems related to slow traffic, innovative projects develop smart parkings willing to save drivers time and free the traffic in busy streets. MAPS project looks for solutions to these problems by the approach of multi-agent systems to create a smart parking that automates the allocation of car spots to drivers. To date, MAPS has solved the problem of spots allocation by modeling agents that represents drivers and the system manager. The development of this work implemented multi-agent system negotiation mechanisms to find solutions related to car spots allocation for smart parking. Three negotiation mechanisms (English auction, Dutch auction and Contract Net Protocol) were implemented using the JaCaMo framework, and test scenarios were constructed for the analysis of time, message exchange and the cost to the drivers that each negotiation mechanism presented as results. With the analysis of these approaches, it was observed that each mechanism presents better results in different scenarios, so it is necessary to implement a hybrid negotiation mechanism which optimizes the process of resource allocation. |
|---|