AGADE Traffic Simulation

AGADE Traffic is an agent based traffic simulator written in Java. Main objective of this project is to provide a tool that allows traffic planners to simulate and analyse behavioural structures in road traffic. The model assumes that personal preferences and decisions made by individual travellers have major impact on overall traffic flow. For this reason, AGADE Traffic focuses on modelling individuals as intelligent software agents. The modelling component uses semantic technologies (OWL and SWRL) to express agent knowledge and preferences that utimately has an effect on travel behaviour. The simulator uses the JADEX framework for implementing BDI agents. For representation of geographic network infrastucture, publicly available map data from OpenStreetMap is stored in a graph database (Neo4J). For more information visit our website www.agade.de and get access to simulation examples as described in our publications. We also provide an online live demo of AGADE Traffic frontend.

Getting Started

Requirements

Initial Setup

• Create a local repository using git clone https://github.com/KITE-Cloud/AGADE-TRAFFIC.git
• Open Frontend, Backend and RoutingKit as separate Maven projects in your IDE.
• Start RoutingKit/osm-cleaner/src/main/java/application/SetupManagerApplication.main(). On first startup SetupManagerApplication creates important environmental variables that are required for connecting service modules. You will need to restart Java to detect changes.

Data Preprocessing and Import

• Manually select a map area at openstreetmap and use overpass API to download necessary osm data for your simulation
• Rename the downloaded file and add .osm file ending e.g. london.osm
• Place osm data in {ProjectDir}/RoutingKit/osm-cleaner/src/main/resources/OSM_Files/
• Launch SetupManagerApplication to select your osm file and clean your osm data. This produces a new osm file e.g. london_cleaned.osm in the same directory.
• Then use SetupManagerApplication to select the cleaned osm file and import data into database.

Launch Service Modules

Make sure you have Docker running!

For End-Users

• Open a terminal and navigate to your project directory
• Startup AGADE Traffic Services (Neo4j, RoutingEngine and Backend) by typing docker-compose up
• On first startup project build must be completed. Project build will be cached which makes startup of subsequent sessions faster.
• You will have to start AGADE Traffic Frontend separately. In the Frontend project, launch src/main/java/de/thm/agade/frontend/FrontendApplication.main()
• Open localhost:8081 in your browser

For Developers

For developers working with AGADE Traffic we recommend starting up services separately.

• Open a terminal and navigate to {ProjectDir}/RoutingKit
• Type in docker-compose build to build a prepared docker image of Neo4j database. If you make changes to the database (e.g. import new data), you will need to rebuild.
• Once build process is completed, type docker-compose up to start up the database
• In the RoutingKit project, launch routing-engine/src/main/rmi/RMIServer.main()
• In the Backend project, launch the src/main/java/BackendApplication.main() in the AGADE project
• In the Frontend project, launch src/main/java/de/thm/agade/frontend/FrontendApplication.main()
• Open localhost:8081 in your browser

Create a simulation

AGADE.Traffic.-.Tutorial.-.Overview.mp4

Simulation results will be output in the Frontend project under the following path: src/main/resources/output

Licensing

AGADE Traffic is an open source product licensed under MIT