jaxmapp - GitLite

KO | EN

gitlite — search

#javascript #react #ai #typescript #llm #golang #android #machine-learning #deep-learning #awesome #security #mcp #ios #openai

jaxmapp

(★ 54)

JAX-based implementation for multi-agent path planning (MAPP) in continuous spaces.

File Explorer

Download Latest Version (.zip)

- benchmark.png
- gallery.png
- roadmap.png
- - 0000.txt
  - 0001.txt
  - 0002.txt
  - 0003.txt
- conf.py
- index.rst
- jaxmapp.env.instance.rst
- jaxmapp.env.obstacle.rst
- jaxmapp.env.rst
- jaxmapp.model.ctrmnet.rst
- jaxmapp.model.cvae.rst
- jaxmapp.model.preprocessor.rst
- jaxmapp.model.rst
- jaxmapp.model.utils.encoder.rst
- jaxmapp.model.utils.features.rst
- jaxmapp.model.utils.rst
- jaxmapp.planner.planner.rst
- jaxmapp.planner.prioritized_planning.rst
- jaxmapp.planner.result.rst
- jaxmapp.planner.rst
- jaxmapp.planner.utils.rst
- jaxmapp.roadmap.grid_sampler.rst
- jaxmapp.roadmap.learned_sampler.rst
- jaxmapp.roadmap.random_sampler.rst
- jaxmapp.roadmap.rst
- jaxmapp.roadmap.sampler.rst
- jaxmapp.roadmap.timed_roadmap.rst
- jaxmapp.roadmap.utils.rst
- jaxmapp.rst
- jaxmapp.utils.data.rst
- jaxmapp.utils.rst
- jaxmapp.utils.sampler.rst
- jaxmapp.utils.viz.rst
- make.bat
- Makefile
- modules.rst
- - checkpoint_894
- - create_eval_data.yaml
  - create_training_data.yaml
  - eval.yaml
  - train.yaml
- .gitignore
- create_all_eval_data.sh
- create_eval_data.py
- create_tfrecords.py
- create_training_data.py
- download_all_data.sh
- eval.py
- eval_all.sh
- eval_example.sh
- train.py
- - check_continuous_collision.pyx
- - __init__.py
- data_test.py
- env_test.py
- planning_test.py
- roadmap_test.py
- - benchmark.png
  - gallery.png
  - roadmap.png
- 1. Quickstart.ipynb
- 2. Develop roadmap construction methods.ipynb
- 3. Benchmarking roadmap construction methods.ipynb
- 4. Training sampler.ipynb
- 5. Sparsify solution paths.ipynb
- create_map.ipynb
- Gallery.ipynb
.dockerignore
.gitignore
docker-compose.yml
Dockerfile
LICENSE
pyproject.toml
README.md
setup.py

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# JAXMAPP: JAX-based Library for Multi-Agent Path Planning in Continuous Spaces
![gallery](https://raw.githubusercontent.com/omron-sinicx/jaxmapp/main/assets/gallery.png) 
![roadmap](https://raw.githubusercontent.com/omron-sinicx/jaxmapp/main/assets/roadmap.png)
![metrics](https://raw.githubusercontent.com/omron-sinicx/jaxmapp/main/assets/benchmark.png)

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](https://github.com/yonetaniryo)), Keisuke Okumura ([@Kei18](https://github.com/Kei18)), Mai Nishimura ([@denkiwakame](https://github.com/denkiwakame))

## Installation

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.

### venv

```console
$ python -m venv .venv
$ source .venv/bin/activate
(.venv) $ pip install -e .[dev]
```

### Docker container
```console
$ docker-compose build
$ docker-compose up -d dev
$ docker-compose exec dev bash
```

### Docker container with CUDA enabled 
```console
$ docker-compose up -d dev-gpu
$ docker-compose exec dev-gpu bash
```

and update JAX modules in the container...

```console
# pip install --upgrade "jax[cuda]==0.3.16" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
```

## Tutorials
- [1. Quickstart: Create MAPP problems and solve them using default roadmap construction methods](https://github.com/omron-sinicx/jaxmapp/blob/main/tutorials/1.%20Quickstart.ipynb)
- [2. Develop roadmap construction methods](https://github.com/omron-sinicx/jaxmapp/blob/main/tutorials/2.%20Develop%20roadmap%20construction%20methods.ipynb)
- [3. Benchmarking roadmap construction methods](https://github.com/omron-sinicx/jaxmapp/blob/main/tutorials/3.%20Benchmarking%20roadmap%20construction%20methods.ipynb)
- [4. Training sampler](https://github.com/omron-sinicx/jaxmapp/blob/main/tutorials/4.%20Training%20sampler.ipynb)

## Citation

```
@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}
}
```