acforvs/dhc-robust-mapf

Learnable MAPF. “Distributed Heuristic Multi-Agent Path Finding with Communication” (DHC) algorithm from ICRA 2021 is implemented and benchmarked in out-of-distribution (OOD) scenarios. A new robust training loop to handle communication failures is introduced.

22
/ 100
Experimental

This project helps operations engineers or robotics researchers coordinate multiple autonomous agents, like robots in a warehouse or drones for delivery, to move efficiently without collisions. It takes in grid-based environment maps and agent starting/ending positions, then outputs optimized paths for each agent. The main users are researchers and practitioners working on multi-agent systems and robotics.

No commits in the last 6 months.

Use this if you need to find robust, decentralized paths for multiple agents in environments where communication between agents might be unreliable or experience packet loss.

Not ideal if your multi-agent system has perfect communication channels, or if you are looking for centralized pathfinding solutions.

robotics multi-agent-systems pathfinding logistics-automation drone-delivery
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

24

Forks

Language

Python

License

MIT

Last pushed

Nov 09, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/acforvs/dhc-robust-mapf"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.