Skip to content

This is the code for ICAPS-24 paper: Map Connectivity and Empirical Hardness of Grid-based Multi-Agent Pathfinding Problem

Notifications You must be signed in to change notification settings

USC-ACTLab/MAPF-Lambda2

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MAPF-Lambda2

This is the code repo for ICAPS-24 paper: "Map Connectivity and Empirical Hardness of Grid-based Multi-Agent Pathfinding Problem"

Dataset

The following figure shows the detailed map information for Experiment 1 in the paper. $\lambda_2$ is annotate on x-axis and the smallest $\lambda_2$ map is at the top left corner. Maps from the MovingAI benchmark dataset are marked in red. All other maps are generated using our fractal and QD instance generator. The map files can be found in dataset/ICAPS-24.

map info

Usage

You can find the QD map generator in qd-generator/.

The mapf-lambda2.py is a sample script to calculate the $\lambda_2$ of a map.

More updates will come very soon including several MAPF map generators and a well documented code base.

Please cite the following paper if you find the results in this paper is helpful:

[1.] Ren, J., Ewing, E., Kumar, T. S., Koenig, S., & Ayanian, N. (2024). Map Connectivity and Empirical Hardness of Grid-based Multi-Agent Pathfinding Problem. In 34th International Conference on Automated Planning and Scheduling.

About

This is the code for ICAPS-24 paper: Map Connectivity and Empirical Hardness of Grid-based Multi-Agent Pathfinding Problem

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages