We present the implementation of an efficient multi-robot task and path planning (MRTPP) method for multi-robot coordination in large-scale cluttered environments. The source code of our method, along with the compared state-of-the-art (SOTA) solvers, is implemented in Python and publicly available here.

The main contributions are summarized as follows: 1) A fast path planner suitable for large-scale and cluttered workspaces that efficiently constructs the cost matrix of collision-free paths between tasks and robots for solving the MRTPP problem. 2) An efficient auction-based method for solving the MRTPP problem by incorporating a novel memory-aware strategy, aiming to minimize the maximum travel cost for robots to visit tasks.

Paper: Efficient Multi-robot Task and Path Planning in Large-Scale Cluttered Environments

Authors: Gang Xu, Yuchen Wu, Sheng Tao, Yifan Yang, Tao Liu, Tao Huang, Huifeng Wu, and Yong Liu

Accepted to: IEEE Robotics and Automation Letters (RA-L), 2025

Code: The source code will be released soon.

@article{xu2025efficient,
  author={Xu, Gang and Wu, Yuchen and Tao, Sheng and Yang, Yifan and Liu, Tao and Huang, Tao and Wu, Huifeng and Liu, Yong},
  journal={IEEE Robotics and Automation Letters}, 
  title={Efficient Multi-Robot Task and Path Planning in Large-Scale Cluttered Environments}, 
  year={2025},
  volume={},
  number={},
  pages={1-8},
  doi={10.1109/LRA.2025.3592146}
}