Setup Windows

To set up the camera processor for the first time, proceed with the following.

Install dependencies

1. pip install -r requirements.txt

Set up a virtual environment

1. python -m venv venv
2. venv\scripts\activate.bat
3. pip install -r requirements.txt again

Setup Mac

To set up the camera processor for the first time, proceed with the following.

Install dependencies

1. pip install -r requirements.txt

Set up a virtual environment

1. python3 -m venv {chosen name of venv}
2. source {chosen name of venv}/bin/activate
3. pip install -r requirements.txt again

Running Camera Processor

1. Ensure you're in the virtual environment
2. Ensure you have docker installed and running on your machine
3. docker compose up --build
4. Ensure that the simulation is running
5. Open a new terminal and navigate to hivemind directory. Enter hivemind venv.
6. In the hivemind venv python3 main.py (Mac) OR
7. py main.py (Windows)
8. In camera processor venv python3 cam_processor.py {camera number, starts at 0} -d (Mac) OR
9. py cam_processor.py {camera number, starts at 0} -d (Windows)
   • Examples: python3 cam_processor.py 0 -d, py cam_processor.py 0 -d
   • Note: The command may be python rather than py

Running Hivemind

1. Ensure that the simulation is running
2. Ensure Camera Processor is running (See above)
3. Ensure you're in a seperate hivemind virtual environment
4. python3 hivemind.py (Mac) OR
5. py hivemind.py (Windows)
   • Note: The command may be python rather than py

Interfaces

API

• POST /world/init: one-time request from Unity. Responds with cop and camera IDs?

{
    'map': ??,
    'cops': {},
    'cameras': {}
}

• POST /camera/<camera_id>: request processing camera image with YOLO.

{
    'position': [f, f, f],
    'rotation': [f, f, f, f],
    'fov': f,
    'image': str
}

• POST /world/robber_report: report initial robber sigthing.

{
    'location': ??,
    'timestamp': f,
    'prefab_id': ??
}

• GET /world/dispatch/<cop_id>: receive dispatch directions.