Running this code

To run this code, ensure that the desired hyperparameters are set. These can be found in settings and multi_agent_settings. To use the optimal parameters, set USE_OPTIMAL_PARAMETERS = True. Ensure that you are running a valid AGENT_TYPE, from ["DDQN", "VDN", "QMIX", "IPPO", "MAPPO"].

For where to save results, update LOCAL_DIR and SUB_DIR.

To render the environment, ensure that RENDER_ENVIRONMENT = True, and update the relevant screen parameters in graphics_settings

To run a multi-agent algorithm, run multi_agent_run. This will render the environment, assuming you set the appropriate value to True, and results will be displayed in the console.

For any issues with packages, ensure that all dependencies in requirements.txt are installed, and the correct version.

Alternatively, you can run:

• pip install -r requirements.txt if running locally
• pip install -r requirements_colab.txt if running on Google Colab

---

The foundation environment extended in this work is taken from Farama-Foundation/HighwayEnv. The citation can be seen below:

@misc{highway-env,
  author = {Leurent, Edouard},
  title = {An Environment for Autonomous Driving Decision-Making},
  year = {2018},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/eleurent/highway-env}},
}