JAXMAPP is a JAX-based library for multi-agent path planning (MAPP) in continuous spaces, with a particular emphasis on roadmap construction and evaluation. With JAXMAPP, You can:
- Create MAPP problem instances with homogeneous/heterogeneous agents
- Construct roadmaps and perform prioritized planning to solve MAPP
- Develop and evaluate your own roadmap construction methods
Main contributors: Ryo Yonetani (@yonetaniryo), Keisuke Okumura (@Kei18), Mai Nishimura (@denkiwakame)
The code has been tested on Ubuntu >=16.04 as well as WSL2 (Ubuntu 20.04) on Windows 11, with python3 (>=3.8). Planning can be performed only on the CPU, and the use of GPUs is supported for training/evaluating machine-learning models. We also provide Dockerfile to replicate our setup.
$ python -m venv .venv
$ source .venv/bin/activate
(.venv) $ pip install -e .[dev]$ docker-compose build
$ docker-compose up -d dev
$ docker-compose exec dev bash$ docker-compose up -d dev-gpu
$ docker-compose exec dev-gpu bashand update JAX modules in the container...
# pip install --upgrade "jax[cuda]==0.3.16" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html- 1. Quickstart: Create MAPP problems and solve them using default roadmap construction methods
- 2. Develop roadmap construction methods
- 3. Benchmarking roadmap construction methods
- 4. Training sampler
@misc{jaxmapp_2022,
author = {Yonetani, Ryo and Okumura, Keisuke},
month = {2},
title = {JAXMAPP: JAX-based Library for Multi-Agent Path Planning in Continuous Spaces},
url = {https://github.com/omron-sinicx/jaxmapp},
year = {2022}
}
@inproceedings{okumura2022ctrm,
title={CTRMs: Learning to Construct Cooperative Timed Roadmaps for Multi-agent Path Planning in Continuous Spaces},
author={Okumura, Keisuke and Yonetani, Ryo and Nishimura, Mai and Kanezaki, Asako},
booktitle={Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS)},
year={2022}
}