Modelación de un sistema inteligente de tráfico vehicular por medio de una simulación basada en agentes
In recent years, people's quality of life has been affected by the constant increase in population. This is especially prevalent in cities, where vehicle traffic is increasing every day, causing time losses to their inhabitants. In order to analyze the behavior of vehicular traffic, a distribut...
- Autores:
-
Cárdenas González, Laura
Soto Lozano, Camila
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2020
- Institución:
- Pontificia Universidad Javeriana
- Repositorio:
- Repositorio Universidad Javeriana
- Idioma:
- spa
- OAI Identifier:
- oai:repository.javeriana.edu.co:10554/53029
- Acceso en línea:
- http://hdl.handle.net/10554/53029
- Palabra clave:
- Multy-agent system
Distributed control architecture
Agent-based simulation
Netlogo
Ingeniería industrial - Tesis y disertaciones académicas
Sistemas multiagentes
Métodos de simulación
- Rights
- openAccess
- License
- Atribución-NoComercial-SinDerivadas 4.0 Internacional
| Summary: | In recent years, people's quality of life has been affected by the constant increase in population. This is especially prevalent in cities, where vehicle traffic is increasing every day, causing time losses to their inhabitants. In order to analyze the behavior of vehicular traffic, a distributed control architecture is designed to represent different traffic systems by means of an agent-based simulation. The simulation will be done in Netlogo's multi-agent modeling software. First, the parameters, variables and indicators of influence on the operation and behavior of vehicular tra ffic in Bogotá were identified. From the information collected, a zone of the city was chosen for the construction of the model, together with the system's agents and their attributes. Secondly, the structure of the control architecture was built. In this work, only the executing level was analyzed, which is given by the heterarchic communication among the agents. Thirdly, four different behaviors of intelligent traffic lights were identified, so they were all evaluated together to validate the efficiency o f each one. Finally, statistical validation and sensitivity analysis were carried out to identify the best policy together with the variables of greatest impact and thus improve vehicle flow. The designed control architecture resulted in a flexible multi - agent model capable of describing vehicle behavior. |
|---|