Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

It Takes Two to Tango: A Holistic Simulator for Joint Order Scheduling and Multi-Agent Path Finding in Robotic Warehouses

Haozheng Xu, Wenhao Li, Zifan Wei, Bo Jin, Hongxing Bai, Ben Yang, Xiangfeng Wang

TL;DR

The paper addresses the problem that decoupled order scheduling and MAPF in RMFS lead to brittle, congested operations under dynamic orders and execution failures. WareRover presents a closed-loop platform that unifies scheduling and MAPF with a high-fidelity, topology-agnostic warehouse model, online replanning, and resilience features. Key contributions include a closed-loop joint optimization interface, a configurable warehouse environment with heterogeneous fleets and multi-stage tasks, and native failure simulation with safety corridors for recovery. Results on realistic scenarios demonstrate that state-of-the-art MAPF methods struggle under coupled constraints, highlighting WareRover's value as a challenging testbed for robust warehouse coordination and sim-to-real research.

Abstract

The prevailing paradigm in Robotic Mobile Fulfillment Systems (RMFS) typically treats order scheduling and multi-agent pathfinding as isolated sub-problems. We argue that this decoupling is a fundamental bottleneck, masking the critical dependencies between high-level dispatching and low-level congestion. Existing simulators fail to bridge this gap, often abstracting away heterogeneous kinematics and stochastic execution failures. We propose WareRover, a holistic simulation platform that enforces a tight coupling between OS and MAPF via a unified, closed-loop optimization interface. Unlike standard benchmarks, WareRover integrates dynamic order streams, physics-aware motion constraints, and non-nominal recovery mechanisms into a single evaluation loop. Experiments reveal that SOTA algorithms often falter under these realistic coupled constraints, demonstrating that WareRover provides a necessary and challenging testbed for robust, next-generation warehouse coordination. The project and video is available at https://hhh-x.github.io/WareRover/.

It Takes Two to Tango: A Holistic Simulator for Joint Order Scheduling and Multi-Agent Path Finding in Robotic Warehouses

TL;DR

The paper addresses the problem that decoupled order scheduling and MAPF in RMFS lead to brittle, congested operations under dynamic orders and execution failures. WareRover presents a closed-loop platform that unifies scheduling and MAPF with a high-fidelity, topology-agnostic warehouse model, online replanning, and resilience features. Key contributions include a closed-loop joint optimization interface, a configurable warehouse environment with heterogeneous fleets and multi-stage tasks, and native failure simulation with safety corridors for recovery. Results on realistic scenarios demonstrate that state-of-the-art MAPF methods struggle under coupled constraints, highlighting WareRover's value as a challenging testbed for robust warehouse coordination and sim-to-real research.

Abstract

The prevailing paradigm in Robotic Mobile Fulfillment Systems (RMFS) typically treats order scheduling and multi-agent pathfinding as isolated sub-problems. We argue that this decoupling is a fundamental bottleneck, masking the critical dependencies between high-level dispatching and low-level congestion. Existing simulators fail to bridge this gap, often abstracting away heterogeneous kinematics and stochastic execution failures. We propose WareRover, a holistic simulation platform that enforces a tight coupling between OS and MAPF via a unified, closed-loop optimization interface. Unlike standard benchmarks, WareRover integrates dynamic order streams, physics-aware motion constraints, and non-nominal recovery mechanisms into a single evaluation loop. Experiments reveal that SOTA algorithms often falter under these realistic coupled constraints, demonstrating that WareRover provides a necessary and challenging testbed for robust, next-generation warehouse coordination. The project and video is available at https://hhh-x.github.io/WareRover/.
Paper Structure (4 sections, 2 figures, 1 table)

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

Table of Contents

  1. Introduction
  2. Related Work
  3. The WareRover Platform
  4. Experiments

Figures (2)

  • Figure 1: The WareRover integrates (i) a warehouse environment builder for structured configurations and high-fidelity maps, (ii) a task & order module for realistic order streams and task decomposition, (iii) a step-level joint optimization loop that couples the scheduler and MAPF planner via global state feedback and online replanning, (iv) a physical motion executor that enforces kinematic constraints with continuous collision checking, and (v) failure simulation & recovery that triggers maintenance corridors and rerouting under non-nominal execution.
  • Figure 2: System demonstration of WareRover. From left to right: (a) real-time warehouse simulation with AGV motion and task execution; (b) failure injection and safety corridor generation under AGV breakdown; (c) interactive control panel for experiment management and parameter configuration; (d) warehouse environment editor for structured map construction and layout configuration.