Kei18 / pibt

Priority Inheritance with Backtracking for Iterative Multi-agent Path Finding (IJCAI-19)

Home Page:https://kei18.github.io/pibt/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

PIBT

Let multiple agents move smoothly. See the project page.

IMPORTANT NOTES

  • In v1.0-v1.2, the PPS implementation includes bugs. Use the latest (≥v1.3). You can check the PPS results here. The results are not so different from the PIBT paper.
  • A new version (pibt2), substantially updated for a journal paper, is out. I strongly recommend using the new one instead of this repo.

Demo

Multi-agent Path Finding

MAPF

Multi-agent Pickup and Delivery (sushi-mode)

MAPD

Requirement

The visualization relies on openFrameworks. You need to install openFrameworks beforehand and export OF_ROOT of your environment.

export OF_ROOT={your openFrameworks directory}

The latest implementation relies on boost.

Usage

At first, you must prepare param file. To confirm details, see a sample file. Then you can execute the simulator as following.

  • implementation with openFrameworks
make of
make ofrun param=sample-param.txt
  • without visualization (for experiment)
make c
make crun param=sample-param.txt

Licence

This software is released under the MIT License, see LICENSE.txt.

Others

Author

Keisuke Okumura is currently a Ph.D. candidate at Tokyo Institute of Technology, working on multiple moving agents.

About

Priority Inheritance with Backtracking for Iterative Multi-agent Path Finding (IJCAI-19)

https://kei18.github.io/pibt/

License:MIT License


Languages

Language:C++ 90.6%Language:Smalltalk 9.0%Language:Makefile 0.5%