摘要
为实现无人机蜂群效能的快速评估,提出一种基于ADC(availability dependability capability)系统效能评估和BP神经网络预测的混合模型,以应对无人机蜂群配置和状态的多样性以及效能计算的复杂性。在分析蜂群效能构成要素的基础上,建立包含无人机通用平台能力,系统级能力,以及任务执行能力的能力指标体系。利用ADC法生成蜂群作战效能样本集合,运用BP神经网络构建关于无人机参数和能力指标的综合作战效能评估模型。利用该评估模型实现异构无人机蜂群实例的综合作战效能评估。结果表明:该模型评估误差可达5%以下,基于样本的评估时间可达3h以内,验证了该模型在异构无人机蜂群效能评估中的有效性及高效性。同时,通过分析数量、配置对无人机蜂群综合效能的影响,获得了异构无人机蜂群配置的可行建议。
This paper presents a hybrid model based on availability dependability capability(ADC)system performance evaluation and back propagation(BP)neural network prediction to realize a rapid performance evaluation of UAV swarms and cope with the diversity of UAV swarm configuration and state and the complexity of performance calculation.By analyzing the components of swarm performance,a capability index system including the general platform capability,system-level capability,and task execution capability of UAVs is established.By using the ADC method,a swarm combat performance sample set is generated,and the BP neural network is used to construct a comprehensive combat performance evaluation model of UAV parameters and capability indexes.The evaluation model is used to evaluate the comprehensive combat performance of heterogeneous UAV swarms.The results show that the evaluation error of this model can reach less than 5%,and the evaluation time based on samples is less than three hours,which verifies the effectiveness and high efficiency of this model in the evaluation of heterogeneous UAV swarm performance.At the same time,by analyzing the influence of quantity and configuration on the comprehensive performance of UAV swarms,feasible suggestions on the configuration of heterogeneous UAV swarms are obtained.
作者
卢元杰
龙珊珊
赵航
冯国旭
赵晓葭
Lu Yuanjie;Long Shanshan;Zhao Hang;Feng Guoxu;Zhao Xiaojia(Shenyang Aircraft Design Institute,AVIC,Shenyang 110035,China;Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《系统仿真学报》
CAS
CSCD
北大核心
2024年第3期700-712,共13页
Journal of System Simulation
基金
国家自然科学基金(51805440)
航空科学基金(201913053001)。
关键词
异构无人机
蜂群系统
效能评估
ADC-BP神经网络
混合模型
heterogeneous UAV
swarm system
performance evaluation
availability dependability capability-back propagation(ADC-BP)neural network
hybrid model
