Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Collaborative Task Assignment, Sequencing and Multi-agent Path-finding

Yifan Bai, Shruti Kotpalliwar, Christoforos Kanellakis, George Nikolakopoulos

TL;DR

This work tackles the joint problem of task assignment, sequencing, and multi-agent pathfinding (TSPF) under agent-task compatibility, aiming to minimize the total flowtime. It introduces CBS-TS, a two-level framework that uses MILP to determine next-best task sequences and CBS with Multi-Label A* to compute collision-free paths, with proofs of completeness and optimality. The approach yields higher success rates and exact optimal solutions in many settings compared to the CBSS baseline, and is validated by both simulated experiments and real robot trials. The results demonstrate practical applicability for scalable, collision-free coordination in multi-robot systems.

Abstract

In this article, we address the problem of collaborative task assignment, sequencing, and multi-agent pathfinding (TSPF), where a team of agents must visit a set of task locations without collisions while minimizing flowtime. TSPF incorporates agent-task compatibility constraints and ensures that all tasks are completed. We propose a Conflict-Based Search with Task Sequencing (CBS-TS), an optimal and complete algorithm that alternates between finding new task sequences and resolving conflicts in the paths of current sequences. CBS-TS uses a mixed-integer linear program (MILP) to optimize task sequencing and employs Conflict-Based Search (CBS) with Multi-Label A* (MLA*) for collision-free path planning within a search forest. By invoking MILP for the next-best sequence only when needed, CBS-TS efficiently limits the search space, enhancing computational efficiency while maintaining optimality. We compare the performance of our CBS-TS against Conflict-based Steiner Search (CBSS), a baseline method that, with minor modifications, can address the TSPF problem. Experimental results demonstrate that CBS-TS outperforms CBSS in most testing scenarios, achieving higher success rates and consistently optimal solutions, whereas CBSS achieves near-optimal solutions in some cases. The supplementary video is available at https://youtu.be/QT8BYgvefmU.

Collaborative Task Assignment, Sequencing and Multi-agent Path-finding

TL;DR

This work tackles the joint problem of task assignment, sequencing, and multi-agent pathfinding (TSPF) under agent-task compatibility, aiming to minimize the total flowtime. It introduces CBS-TS, a two-level framework that uses MILP to determine next-best task sequences and CBS with Multi-Label A* to compute collision-free paths, with proofs of completeness and optimality. The approach yields higher success rates and exact optimal solutions in many settings compared to the CBSS baseline, and is validated by both simulated experiments and real robot trials. The results demonstrate practical applicability for scalable, collision-free coordination in multi-robot systems.

Abstract

In this article, we address the problem of collaborative task assignment, sequencing, and multi-agent pathfinding (TSPF), where a team of agents must visit a set of task locations without collisions while minimizing flowtime. TSPF incorporates agent-task compatibility constraints and ensures that all tasks are completed. We propose a Conflict-Based Search with Task Sequencing (CBS-TS), an optimal and complete algorithm that alternates between finding new task sequences and resolving conflicts in the paths of current sequences. CBS-TS uses a mixed-integer linear program (MILP) to optimize task sequencing and employs Conflict-Based Search (CBS) with Multi-Label A* (MLA*) for collision-free path planning within a search forest. By invoking MILP for the next-best sequence only when needed, CBS-TS efficiently limits the search space, enhancing computational efficiency while maintaining optimality. We compare the performance of our CBS-TS against Conflict-based Steiner Search (CBSS), a baseline method that, with minor modifications, can address the TSPF problem. Experimental results demonstrate that CBS-TS outperforms CBSS in most testing scenarios, achieving higher success rates and consistently optimal solutions, whereas CBSS achieves near-optimal solutions in some cases. The supplementary video is available at https://youtu.be/QT8BYgvefmU.
Paper Structure (14 sections, 5 equations, 6 figures, 1 table, 1 algorithm)

This paper contains 14 sections, 5 equations, 6 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Contribution
  4. Problem Description
  5. Methodology
  6. High Level of CBS-TS
  7. Sequential Next Best Task Sequencing (MILP Formulation)
  8. Low level of CBS-TS: Multi-label A*
  9. Example
  10. Properties of CBS-TS
  11. Experiments
  12. Simulation
  13. Physical Experiment
  14. Conclusion and Future Work

Figures (6)

  • Figure 1: An example demonstrating how different task assignments result in different conflict-free paths for two agents (starting at $s_1$, $s_2$) and three tasks (at $g_1$, $g_2$, $g_3$), with dashed lines representing the planned conflict-free paths. The circular segment in the blue path represents a wait action required to resolve conflicts. The black cells denote obstacles. The color of the starts and goals indicates the agent-task compatibility constraints (i.e. $g_3$ can be visited by either the green or blue agent).
  • Figure 2: An example of conducting the best-first search on the conflict forest to find the optimal solution in CBS-TS.
  • Figure 3: Success rate with 60$s$ time limit.
  • Figure 4: Suboptimality ratio for CBSS and ECBS-TS. Only successful cases are included.
  • Figure 5: Boxplot of runtime. Only successful cases are included.
  • ...and 1 more figures