欢迎访问《兵工学报》官方网站,今天是

兵工学报 ›› 2025, Vol. 46 ›› Issue (8): 240928-.doi: 10.12382/bgxb.2024.0928

• • 上一篇    下一篇

基于可编程门阵列的集群弹药协同决策与同步起爆方法

张传昊, 李豪杰*(), 李长生, 张凌云, 乔诗翔, 于航   

  1. 南京理工大学 智能弹药技术国防重点学科实验室, 江苏 南京 210094
  • 收稿日期:2024-10-05 上线日期:2025-08-28
  • 通讯作者:
  • 基金资助:
    中国博士后面上基金项目(2023M731676); 江苏省卓越博士后计划项目(2023ZB007)

Collaborative Decision-making and Synchronized Detonation Method for Swarm Munitions Based on FPGA

ZHANG Chuanhao, LI Haojie*(), LI Changsheng, ZHANG Lingyun, QIAO Shixiang, YU Hang   

  1. Ministerial Key Laboratory of ZNDY, Nanjing University of Science and Technology, Nanjing 210094, Jiangsu, China
  • Received:2024-10-05 Online:2025-08-28

摘要:

针对巡飞弹集群智能决策与同步打击的需求,为实现理论算法在弹药弱硬件平台中的应用,提高集群弹药协同作战的智能化水平,提出适用于可编程门阵列(Field Programmable Gate Array,FPGA)的威胁评估数学模型与分配方法。基于层次分析法、熵值法与逼近理想解的排序方法,通过基数排序、等效替代等方法建立基于FPGA的威胁评估模型。基于分布式的通信结构与数据多跳传输策略,提出多节点低时延自适应延时起爆方法。基于软件平台与硬件平台的数值对比仿真,验证了FPGA实现威胁评估的快速性与准确性,每次求解仅需要50μs;通过全流程仿真,验证了多节点任务分配、节点安全与收发状态转换以及自适应延时起爆的可行性。最终进行了多节点协同决策动态试验,验证四节点协同决策的可行性与结果的准确性;通过静态全流程试验验证了节点同步起爆的精确性,采集到四节点起爆信号输出总时延不超过200μs。

关键词: 集群弹药, 协同决策, 分布通信, 同步起爆

Abstract:

To address the needs for intelligent decision-making and synchronized strikes in loitering munition swarms and to enable the application of theoretical algorithms on resource-constrained hardware platforms,this study proposes a threat assessment mathematical model and distribution method tailored for Field Programmable Gate Array (FPGA).Utilizing techniques such as Analytic Hierarchy Process (AHP),entropy-based evaluation,and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS),the model incorporates radix sorting and equivalent substitution to construct an FPGA-based threat assessment framework.A multi-node low-latency adaptive delayed detonation method is developed based on a distributed communication architecture and multi-hop data transmission strategy.Comparative simulations on both software and hardware platforms demonstrate the proposed model's efficiency and accuracy,achieving a computation time of 50μs per evaluation.Full-process simulations validate the feasibility of multi-node task allocation,node security,state transitions,and adaptive delayed detonation.Dynamic multi-node experiments further confirm the accuracy and effectiveness of four-node collaborative decision-making.Static end-to-end experiments validate precise synchronized detonation,with a total delay of detonation signal output across four nodes not exceeding 200μs.

Key words: swarm munitions, cooperative decision-making, distributed communication, synchronized detonation

中图分类号: