FireCommander - The MARL Package

A detailed description of the heterogeneous multi-agent domain that we created for our experiments in the original HetNet paper (see here: https://ifaamas.org/Proceedings/aamas2022/pdfs/p1173.pdf), the FireCommander environment,can be accessed through the following links:

FireCommander arXiv Paper: FireCommander: An Interactive, Probabilistic Multi-agent Environment for Heterogeneous Robot Teams

FireCommander Paper Link: https://arxiv.org/abs/2011.00165

FireCommander Codebase: https://github.com/EsiSeraj/FireCommander2020

HetNet AAMAS'22 Paper: https://ifaamas.org/Proceedings/aamas2022/pdfs/p1173.pdf

HetNet AAMAS'22 Codebase: https://github.com/CORE-Robotics-Lab/HetNet

NOTE: For utility, please refer to the full documentation and tutorial provided in folder Documentation

• Documentation: Tutorials and documentations.
• LfD-HRI Package: FireCommander Python Code and GUI ready for data collection and user studies
• MARL Package: FireCommander Python Code and environments ready for RL and MARL algorithms to be implemented on.

License

Code is available under MIT license.

Citation

If you use this work and/or this codebase in your research, we ask you to please cite the following:

@article{seraj2020firecommander,
  title={FireCommander: An Interactive, Probabilistic Multi-agent Environment for Heterogeneous Robot Teams},
  author={Seraj, Esmaeil and Wu, Xiyang and Gombolay, Matthew},
  journal={arXiv preprint arXiv:2011.00165},
  year={2020}
}