A multi-agent autonomous driving simulation built on MetaDrive, enabling Vehicle-to-Vehicle (V2V) communication, cooperative decision-making, and secure message exchange.
The project evaluates how shared situational awareness improves safety, coordination, and responsiveness among autonomous vehicles.
- Multi-agent autonomous driving simulation
- Secure V2V communication (encryption enabled)
- Hybrid A* based path planning
- Intelligent Driver Model (IDM) integration
- Config-driven simulation (YAML)
- Detailed logging and analysis support
.
├── communication/ # V2V communication & security
│ ├── broadcaster.py
│ ├── receiver.py
│ ├── encryption_utils.py
│ └── logging_config.py
│
├── decision_engine/ # Autonomous decision making
│ ├── planner.py
│ ├── hybrid_astar_planner.py
│ └── response_planner.py
│
├── metadrive_env/ # MetaDrive environment handling
│ ├── env_manager.py
│ └── env_manager_with_idm.py
│
├── config/ # Configuration files
│ └── config.yaml
│
├── logs/ # Runtime logs
│ ├── communication.log
│ └── sim_log.txt
│
├── Video Demo/ # Simulation demo videos
├── Research Paper/ # Reference research papers
│
├── main.py # Main entry point
├── main_autonomous_complex.py
├── main_autonomous_idm.py
├── drive_in_real_env.py
│
├── requirements.txt # Dependencies
└── secret.key # Encryption key (not for public repos)
[MetaDrive Environment]
↓
[Autonomous Agents]
↓
[Decision Engine]
↓
[V2V Communication Layer]
↓
[Cooperative Actions]
git clone https://github.com/Pushpak1203/V2
cd V2
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txtRun the default multi-agent simulation:
python main.pyRun IDM-based simulation:
python main_autonomous_idm.pyRun complex cooperative scenario:
python main_autonomous_complex.pysecret.key is used for encrypted V2V communication.
Do NOT commit this file to public repositories. Add it to .gitignore.
- Communication events →
logs/communication.log - Simulation behavior →
logs/sim_log.txt
These logs are used for latency analysis, collision evaluation, and coordination assessment.
- V2I / V2X communication support
- Reinforcement learning-based agents
- Latency and packet-loss modeling
- Real-world traffic dataset integration
- Project Report (PDF)
- Research Papers
- Demo Videos
This project is intended for academic and research purposes.
Link: https://drive.google.com/drive/folders/1YL4FTn4tIRJDHd-Yo_GZrP4uuiuF-Qa5?usp=sharing