This project provides a microservices-based platform for crowdsourced navigation and data aggregation. It comprises the following services:
Navigation Service: Routes users to their destinations using OSRM.
Crowdsourcing Service: Collects real-time user feedback and location data.
Aggregation Service: Processes and clusters the collected crowdsourced data using the Louvain community detection algorithm.
OSRM Service: Provides high-performance routing via the hamirho/osrm:trafficless Docker image.
Authentication Service: Secures API endpoints and manages user identities (ensure this is running or reachable).
Real-time route calculation with OSRM.
Crowdsourced data collection for traffic, POIs, and user feedback.
Graph-based clustering of crowdsourced events using Louvain.
Modular microservices architecture for scalability and maintainability.
Out-of-the-box Docker support using Docker Compose.
Docker & Docker Compose for containerization.
Go / Python (service-specific stacks).
OSRM for routing engine.
Louvain algorithm for community detection.
RESTful APIs for inter-service communication.
Docker Engine (>= 20.10)
Docker Compose (>= 1.29)
(Optional) A running Authentication Service exposing a REST API
- Clone the Repository
git clone https://github.com/ECTLab/Crowdsourcing-Platform.git
cd Crowdsourcing-Platform
- Download the OSM Map of Iran Before running the services, download the required OpenStreetMap data by executing the following script:
sh download_map.sh
This will download and place the Iran OSM map on your system.
- Build and Start All Services
docker-compose up --build
This command will:
Build navigation, crowdsourcing, and aggregation from local Dockerfiles
Pull and run the OSRM image (hamirho/osrm:trafficless)
Start all containers in detached mode
- Verify the Services
docker-compose ps
You should see:
navigation listening on port 8080
crowdsourcing listening on port 8081 (container port 8080)
osrm listening on port 5000
aggregation restarting on exit to process new data
redis running on port 6379
- Access the Interface Once all services are up, open your browser and navigate to:
http://localhost:3000
- Submit a Report Zoom in to a specific road or street on the map.
Click on two points along the road to define the route.
3. Import Reports
After submitting a report, click the Import button.
Red pins will appear for reports that are not yet aggregated.
Wait up to 30 seconds, then click Import again to see the aggregated event.
Routes with reported police presence will now include police icons.
4. Confirm or Decline Police Presence
Click the Confirm button to select a police pin on the map.
A dialog will appear to confirm or decline the police presence.
Once confirmed/declined, click Import again to see the updated confidence level.
The Aggregation Service:
Reads crowdsourced event data from the database.
Constructs a graph where nodes represent events/locations and edges represent proximity or similarity.
Applies the Louvain community detection algorithm to identify clusters (communities) of related events.
Outputs cluster assignments or summary statistics for downstream consumption.
Running Aggregation Manually If you need to trigger aggregation on-demand:
docker-compose run --rm aggregation
docker-compose logs -f
docker logs -f navigation