Scholarlink

The perception-planning performance of Unmanned Tracked Vehicle (UTV) serves as the foundation for executing complex and diverse missions. However, the intricate operational environment induces the invalid characterization of measured data. To address these challenges, this study proposes a systematic methodology structured as follows: First, a high-fidelity real-time co-simulation framework is established by integrating Unreal Engine 5 (UE5) virtual modeling with the Vortex dynamics engine. This framework enables closed-loop co-simulation of UTV-environment interactions, ROS-based algorithm-in-the-loop(AIL) control, and dynamic modeling. Second, Distributed interactive simulation communication is achieved via Data Distribution Service (DDS) middleware. Finally, experimental validation confirms the effectiveness of the high-fidelity simulation system in generating realistic data.

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

Co-simulation and Validation Methodology for Unmanned Tracked Vehicle

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Lunfei Liang,
  • Huixin Huang,
  • Geng Chen,
  • Song Liu,
  • Yan Pan

摘要

The perception-planning performance of Unmanned Tracked Vehicle (UTV) serves as the foundation for executing complex and diverse missions. However, the intricate operational environment induces the invalid characterization of measured data. To address these challenges, this study proposes a systematic methodology structured as follows: First, a high-fidelity real-time co-simulation framework is established by integrating Unreal Engine 5 (UE5) virtual modeling with the Vortex dynamics engine. This framework enables closed-loop co-simulation of UTV-environment interactions, ROS-based algorithm-in-the-loop(AIL) control, and dynamic modeling. Second, Distributed interactive simulation communication is achieved via Data Distribution Service (DDS) middleware. Finally, experimental validation confirms the effectiveness of the high-fidelity simulation system in generating realistic data.