[MAPF Tracker] Call for Submissions

[Daniel Harabor]

Dear MAPF researchers,

This is a call for results/submissions to the MAPF Tracker (https://tracker.pathfinding.ai), a community database dedicated to advancing research on the topic of Multi-Agent Path Finding (MAPF).

The MAPF Tracker is a centralised, openly accessible platform that collects best-known results on the standard set of 1.46M+ classical MAPF benchmark problems, used by many researchers in our community. Its mission is to:

• Track the state of the art, by recording best known solutions and best known bounds for each problem instance
• Highlight notable algorithms that advance the research frontier
• Compare methods, understand trends, and build on each other’s work
• Identify opportunities for further improvement
• Lower barriers for entry into this research area

The platform comes packed with visualisations, comparative graphs, performance statistics, and additional tools to support deeper exploration of benchmark results. 

Community contributions are essential to making this system work. That means we need your help. If you have recently published a paper on classical MAPF we encourage you to upload your results to the database:

• Computed plans
• Achieved costs 
• Lower-bound claims
• Links to the paper and source code, which others can use to read about and cite your work.

Submit your results to MAPF Tracker: https://tracker.pathfinding.ai/submit. Drag-and-drop json/csv/yaml files directly into your browser or upload data programmatically via the web API. The system accepts results for single problem instances or upload-in-bulk.

By submitting your work, you boost its visibility, make it easier for others to discover and compare against, and contribute to a more transparent and reproducible research ecosystem.

Find out more:

• Explore existing submissions:

  https://tracker.pathfinding.ai/submissions
• Analyse a best known solution for a very challenging problem:

  https://tinyurl.com/mrydeak6
• Watch a demo video from ICAPS 2023 (NB: MAPF Tracker v.1.0)

  https://tracker.pathfinding.ai/docs/system-demo

Thank you for supporting this community initiative!

Kind regards,

Kevin Zheng and Daniel Harabor

On behalf of the MAPF Tracker team