Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimal Task Assignment and Path Planning using Conflict-Based Search with Precedence and Temporal Constraints

Yu Quan Chong, Jiaoyang Li, Katia Sycara

TL;DR

The paper addresses TAPF-PTC, a problem that unifies task assignment, path planning, and user-defined objective under precedence and temporal constraints. It proposes CBS-TA-PTC, a Conflict-Based Search extension that on-demand assigns tasks and computes collision-free paths, guided by an RL-based oracle to maximize return. The approach is theoretically complete and optimal under a single-subtask assumption and introduces a bounded-suboptimal variant for scalability; empirically, it outperforms MAPPO-based MARL and adapted TAPF methods on a complex bomb-defusing domain. The work demonstrates practical gains in solving tightly constrained multi-agent tasks, highlighting potential for integration with learning-based methods to improve efficiency and applicability in real-world scenarios.

Abstract

The Multi-Agent Path Finding (MAPF) problem entails finding collision-free paths for a set of agents, guiding them from their start to goal locations. However, MAPF does not account for several practical task-related constraints. For example, agents may need to perform actions at goal locations with specific execution times, adhering to predetermined orders and timeframes. Moreover, goal assignments may not be predefined for agents, and the optimization objective may lack an explicit definition. To incorporate task assignment, path planning, and a user-defined objective into a coherent framework, this paper examines the Task Assignment and Path Finding with Precedence and Temporal Constraints (TAPF-PTC) problem. We augment Conflict-Based Search (CBS) to simultaneously generate task assignments and collision-free paths that adhere to precedence and temporal constraints, maximizing an objective quantified by the return from a user-defined reward function in reinforcement learning (RL). Experimentally, we demonstrate that our algorithm, CBS-TA-PTC, can solve highly challenging bomb-defusing tasks with precedence and temporal constraints efficiently relative to MARL and adapted Target Assignment and Path Finding (TAPF) methods.

Optimal Task Assignment and Path Planning using Conflict-Based Search with Precedence and Temporal Constraints

TL;DR

The paper addresses TAPF-PTC, a problem that unifies task assignment, path planning, and user-defined objective under precedence and temporal constraints. It proposes CBS-TA-PTC, a Conflict-Based Search extension that on-demand assigns tasks and computes collision-free paths, guided by an RL-based oracle to maximize return. The approach is theoretically complete and optimal under a single-subtask assumption and introduces a bounded-suboptimal variant for scalability; empirically, it outperforms MAPPO-based MARL and adapted TAPF methods on a complex bomb-defusing domain. The work demonstrates practical gains in solving tightly constrained multi-agent tasks, highlighting potential for integration with learning-based methods to improve efficiency and applicability in real-world scenarios.

Abstract

The Multi-Agent Path Finding (MAPF) problem entails finding collision-free paths for a set of agents, guiding them from their start to goal locations. However, MAPF does not account for several practical task-related constraints. For example, agents may need to perform actions at goal locations with specific execution times, adhering to predetermined orders and timeframes. Moreover, goal assignments may not be predefined for agents, and the optimization objective may lack an explicit definition. To incorporate task assignment, path planning, and a user-defined objective into a coherent framework, this paper examines the Task Assignment and Path Finding with Precedence and Temporal Constraints (TAPF-PTC) problem. We augment Conflict-Based Search (CBS) to simultaneously generate task assignments and collision-free paths that adhere to precedence and temporal constraints, maximizing an objective quantified by the return from a user-defined reward function in reinforcement learning (RL). Experimentally, we demonstrate that our algorithm, CBS-TA-PTC, can solve highly challenging bomb-defusing tasks with precedence and temporal constraints efficiently relative to MARL and adapted Target Assignment and Path Finding (TAPF) methods.
Paper Structure (30 sections, 4 figures, 1 table, 2 algorithms)

This paper contains 30 sections, 4 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Background
  3. Multi-Agent Path Finding (MAPF)
  4. Conflict-Based Search (CBS)
  5. Multi-Agent Reinforcement Learning (MARL)
  6. Target Assignment and Path Finding (TAPF)
  7. Related Work
  8. Problem Definition
  9. Absolute Temporal Range Constraints
  10. Inter-Goal Temporal Constraints
  11. Precedence Constraints
  12. Theoretical Overview
  13. CBS-TA-PTC
  14. High-Level of CBS-TA-PTC
  15. Low-Level of CBS-TA-PTC
  16. ...and 15 more sections

Figures (4)

  • Figure 1: An instance of the bomb-defusing task environment
  • Figure 2: Evaluated mean return of MAPPO from 16 rollouts
  • Figure 3: Evaluated results for CBS-based algorithms, where the intervals highlight the standard deviation of averaged results across 10 trials.
  • Figure :