In this paper,a hybrid deep learning network-based model is proposed and implemented for maneuver decision-making in an air combat environment.The model consists of stacked sparse auto-encoder net-work for dimensionality reduction of high-dimensional,dynamic time series combat-related data and long short-term memory network for capturing the quantitative rela-tionship between maneuver control variables and the time series combat-related data after dimensionality reduction.This model features:using time series data as the basis of decision-making,which is more in line with the actual de-cision-making process;using stacked sparse auto-encoder network to reduce the dimension of time series data to predict the result more accurately;in addition,taking the maneuver control variables as the output to control the maneuver,which makes the maneuver process more flex-ible.The relevant experiments have demonstrated that the proposed model can effectively improve the predic-tion accuracy and convergence rate in the prediction of maneuver control variables.

School of Electronics and Information,Northwestern Polytechnical University,Xi'an 710072,China;School of Engineering,London South Bank University,London SE1 0AA,UK

[1]LI Bo;LIANG Shiyang;CHEN Daqing;LI Xitong-.A Decision-Making Method for Air Combat Maneuver Based on Hybrid Deep Learning Network)[J].电子学报（英文）,2022(01):107-115

Decision,Making,Method,Air,Combat,Maneuver,Based,Hybrid,Deep,Learning,Network,In,this,paper,hybrid,deep,learning,network,model,proposed,implemented,maneuver,decision,making,air,combat,environment,consists,stacked,sparse,auto,encoder,dimensionality,reduction,high,dynamic,series,related,data,long,short,term,memory,capturing,quantitative,tionship,between,control,variables,after,This,features,using,basis,which,more,line,actual,process,reduce,result,accurately,addition,taking,output,makes,flex,ible,relevant,experiments,have,demonstrated,that,can,effectively,improve,accuracy,convergence,prediction

0.508234