Research On The Model Of Human-machine Fusion Intelligent Decision-making Based On Typical Helicopter Missions

With the trend of informatization and intelligence,great changes have taken place in both the civilian and military fields.Artificial intelligence has landed in various application scenarios in the civilian field,such as face recognition,license plate recognition,speech recognition,etc.However,in the military field,in the face of complex,dynamic and deceptive military confrontation scenarios,ordinary artificial intelligence is still unable to cope.Ordinary artificial intelligence relies too much on data and historical knowledge,and cannot gain insight into the value and factual judgments that people have.The computing process,computing methods and data processing capabilities of machines still cannot meet the needs of military decision-making systems in complex environments.In the decisionmaking process,the processing and decision-making methods of subjective and objective information cannot meet the information and knowledge needs of human-machine fusion intelligence.The representation of information uncertainty,the accuracy of decision-making algorithms,and the credibility of decision-making processes cannot be fully represented by simple mathematical logic models.Therefore,this paper proposes a new decision-making model--human-machine fusion intelligent decisionmaking model based on fuzzy decision tree.Investigated decision tree algorithms such as ID3,C4.5 and CART,compared the advantages and disadvantages of traditional decision tree and fuzzy decision tree,and decided to use fuzzy decision tree to build the model of human-machine fusion intelligent decision-making.Based on typical helicopter missions,the inputs and outputs of the system are determined.Objective data input includes:scene,our helicopter attributes,enemy radar detection distance,air defense weapon attack distance,jamming capability,environmental information,etc.;Subjective data input includes mission determination,time requirements,mission value mission expectations,etc.According to the subjective and objective input data,the parameters are simplified through the corresponding normalization formula,and further input into the situation calculation formula to complete our attack situation,our defense situation,enemy threat situation,environmental situation,mission determination,time requirements,The situational calculation of task value and task expectation is corrected according to the calculation results and further subjective results of people,and the fuzzification is completed.Using the fuzzy decision tree,input the subjective and objective parameters after the situation assessment,and get decision suggestions and decision paths.This system is mainly built by the Flask framework,SQLite database to store data,HTML5 to build a visual interface,and combine real-time subjective and objective input data to complete normalization,fuzzification and fuzzy decision-making.Demonstrate decision-making results and decision-making process.The effectiveness of the humanmachine fusion intelligent decision-making system is verified by comparing the satisfaction of the decision results between the humanmachine fusion intelligent decision-making system and the ordinary fuzzy decision tree.