•  
  •  
 

Journal of System Simulation

Abstract

Abstract: To improve the inadequacy of traditional test and identification systems, this paper proposed an overall architecture for dynamic testing across the entire lifecycle based on the concept of parallel battlefield (integration of physical, virtual, and cognitive battlefields), meeting the new requirements for the testing of intelligent unmanned systems. This architecture included high-fidelity mapping between virtual and physical battlefields, red-blue adversarial deductions and model optimization, simulation to reality (Sim2Real), human-machine collaboration, and cloud-end integrated control, as well as multidimensional assessment and confidence analysis. Centered on the principles of "mutual driving between virtual and physical battlefields, dynamic closed-loop, human-machine collaboration, and integrated testing and warfare, " this architecture effectively met the spiral evaluation and upgrade requirements of intelligent unmanned systems. Key supporting technologies for implementing the architecture were demonstrated. A feasibility analysis was conducted, and measures and recommendations were proposed to promote the implementation of the architecture. This architecture could provide a reference for establishing a testing system for intelligent unmanned systems.

First Page

1933

Last Page

1950

CLC

E917；TP391

Recommended Citation

Liu Dayong, Dong Zhiming, Guo Qisheng, et al. Dynamic Testing Architecture of Intelligent Unmanned Systems Based on Parallel Battlefields[J]. Journal of System Simulation, 2025, 37(8): 1933-1950.

Corresponding Author

Dong Zhiming

DOI

10.16182/j.issn1004731x.joss.25-0345

Download

Share

COinS