Scholarlink

<p>This paper presents a distributed task allocation framework for multiple unmanned vehicles on the robot operating system 2 (ROS2) platform. The framework incorporates a generalized multi-waypoint task model, in which each task is defined as a sequence of waypoints generated from parameterized operational modes, enabling flexible representation of complex missions. A mission planning computer is developed to automatically generate such tasks and manage them in real time via a graphical user interface. Based on the task model, two representative task allocation algorithms, the asynchronous consensus-based bundle algorithm (ACBBA) and the distributed auction algorithm (DAA), are implemented within the distributed framework as modular ROS2 nodes. The algorithms are evaluated in hardware-in-the-loop (HITL) simulation with three mission scenarios and benchmarked against the global optimal solution. Simulation results empirically validate that DAA achieves near-optimal performance in a wide area search (WAS) scenario, while ACBBA provides robust performance in multi view observation (MVO) and maneuvering by altitude (MA) scenarios. To facilitate better understanding of the HITL simulation results, we provide demonstration videos at <a href="https://www.youtube.com/playlist?list=PLUIBjqIFRraEkQdp1zQHSRwhXDAeJjZ2D">https://www.youtube.com/playlist?list=PLUIBjqIFRraEkQdp1zQHSRwhXDAeJjZ2D</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A ROS2-based Distributed Task Allocation Framework for Unmanned Vehicles: From Multi-waypoint Task Modeling to HITL Evaluation

  • 期刊论文

  • 发表日期：

    2025年12月5日

查看全文
  • Hwan-Yong Park,
  • Yeong-Ung Kim,
  • Jun-Oh Park,
  • Yoo-Bin Bae

摘要

This paper presents a distributed task allocation framework for multiple unmanned vehicles on the robot operating system 2 (ROS2) platform. The framework incorporates a generalized multi-waypoint task model, in which each task is defined as a sequence of waypoints generated from parameterized operational modes, enabling flexible representation of complex missions. A mission planning computer is developed to automatically generate such tasks and manage them in real time via a graphical user interface. Based on the task model, two representative task allocation algorithms, the asynchronous consensus-based bundle algorithm (ACBBA) and the distributed auction algorithm (DAA), are implemented within the distributed framework as modular ROS2 nodes. The algorithms are evaluated in hardware-in-the-loop (HITL) simulation with three mission scenarios and benchmarked against the global optimal solution. Simulation results empirically validate that DAA achieves near-optimal performance in a wide area search (WAS) scenario, while ACBBA provides robust performance in multi view observation (MVO) and maneuvering by altitude (MA) scenarios. To facilitate better understanding of the HITL simulation results, we provide demonstration videos at https://www.youtube.com/playlist?list=PLUIBjqIFRraEkQdp1zQHSRwhXDAeJjZ2D.