A coordinated multi-agent system that simulates security patrolling in a warehouse environment using a drone, static surveillance cameras, and a security personnel. The system implements computer vision using YOLO and agent coordination to detect and respond to potential security threats.
This project implements a surveillance system where multiple agents (drones, cameras, and security personnel) work together to monitor and secure facilities. The system features:
- Autonomous drone patrolling
- Computer vision-based threat detection
- Multi-agent coordination
- Real-time response protocols
- YOLO model integration for object detection
- Threats or suspicious activities
- Unity Hub
- Python 3.x
- Git
- Required Python packages (specified in requirements.txt)
# Clone the repository
git clone https://github.com/v3gaaa/multiagentes
# Navigate to project directory
cd multiagentes- Open Unity Hub
- Select "Add project from disk"
- Navigate to and select the
multiagentesfolder - Choose "UnityEvidencia2"
- Open the project
# Navigate to Unity project
cd multiagentes
# Install Python dependencies
pip install -r requirements.txt
# Start the server
cd Server
python server.py- In Unity, navigate to:
- Assets → Scenes
- Select "WarehouseDon"
- Click the Play button to start the simulation after running the server on the terminal
- Primary Detection: Either through:
- Fixed cameras detecting suspicious activity
- Drone's onboard camera direct detection (using YOLO model with confidence threshold >0.7)
-
Initial Alert:
- Camera detection: Alerts the drone
- Drone detection: Alerts the security guard
-
Drone Response:
- Navigates to detection coordinates
- Performs detailed area inspection
- Confirms threat presence
-
Security Guard Protocol:
- Goes to the control station upon alert
- Assumes drone control once it has reached the station
- Maintains position at control station during investigation
- Red warning lights activation
- Drone returns to landing position
- Simulation termination sequence