Skip to main navigation Skip to search Skip to main content

面向多点动态集结的异构无人系统空地协同任务规划

Translated title of the contribution: Air-ground coordination task planning for heterogeneous unmanned systems in multi-point dynamic aggregation
  • Yu Yang Wang
  • , Bin Xin*
  • , Yu Long Ding*
  • , Bin He
  • , Jie Chen
  • *Corresponding author for this work
  • Tongji University
  • Ministry of Education in China
  • Beijing Institute of Technology
  • Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

Abstract

A class of large-scale tasks with dispersed locations and changing states exists in real-world scenarios, such as disaster response and forest firefighting. These tasks can be modeled as multi-point dynamic aggregation (MPDA) tasks. To minimize casualties and improve task execution efficiency, deploying unmanned ground vehicles (UGVs) for such tasks is an ideal solution. However, in large-scale MPDA tasks, direct communication among UGVs is often difficult. An unmanned aerial vehicle (UAV) can serve as a messenger, periodically visiting each UGV for data transmission. UGVs must leverage the UAV-transmitted information to make distributed decisions and collaborate in completing the MPDA task. The planning problem of UAV-UGV coordination for MPDA tasks can be modeled as a multi-task, multi-stage dynamic scheduling problem for heterogeneous robots. In this problem, the messenger UAV and UGVs have limited communication frequency, and the information transmitted by the UAV to the UGVs involves communication delays and uncertainties. As a result, existing task planning methods cannot resolve conflicts effectively and produce reasonable task allocations. To address this problem, we propose a UGV sequential decision-making mechanism tailored for messenger-based air-ground coordination. As the MPDA task dynamically evolves, each UGV sequentially updates its task plan from a global perspective based on the visiting sequence of the UAV to UGVs. Building upon this mechanism, we introduce a critical path-based UAV sequencing method. By prioritizing the UAV to visit the UGVs that appear most frequently on the critical path, this method effectively reduces the completion time of MPDA tasks. Furthermore, we propose a UGV distributed coordination strategy under the sequential decision-making framework. Each UGV updates its predicted speed using a particle filter-based probabilistic model and extrapolates the task plans of unvisited UGVs through a heuristic approach to mitigate potential global task conflicts caused by prediction errors. Experimental results demonstrate that the proposed algorithm generates better task plans and achieves shorter MPDA task completion times than other state-of-the-art algorithms.

Translated title of the contributionAir-ground coordination task planning for heterogeneous unmanned systems in multi-point dynamic aggregation
Original languageChinese (Traditional)
Pages (from-to)1447-1466
Number of pages20
JournalZhongguo Kexue Jishu Kexue/Scientia Sinica Technologica
Volume55
Issue number9
DOIs
Publication statusPublished - 1 Sept 2025
Externally publishedYes

Fingerprint

Dive into the research topics of 'Air-ground coordination task planning for heterogeneous unmanned systems in multi-point dynamic aggregation'. Together they form a unique fingerprint.

Cite this