Research On UAV Countermeasure Algorithm Based On Fuzzy Logic

With the widespread application of UAVs in the battlefield environment,higher requirements are placed on UAV countermeasures algorithms.The battlefield environment has complex constraints.The scale of drone confrontation is larger,and the confrontation methods are more intense and diversified.This leads to problems such as a decline in coordination between drones and poor decision-making effects.It is necessary to study drone confrontation algorithms.Ensure that the UAV reasonably analyzes the battlefield situation and its own attributes to complete the strike and annihilation of enemy targets.For this reason,this paper studies the UAV countermeasure algorithm based on fuzzy logic control theory,which mainly includes the following three points:（1）Aiming at the problem of UAV confrontation decision-making in complex air combat environment,this paper designs a UAV cluster intelligent attack system based on the cascade fuzzy idea to realize the control of the UAV individual and cluster attack behavior.The system includes situational threat assessment fuzzy inference system,capability threat assessment fuzzy inference system,target threat assessment fuzzy inference system,battle participation decision fuzzy inference system,pursuit decision fuzzy inference system,shell delay fuzzy inference system and formation target selection fuzzy inference system.The simple structure of the small fuzzy inference system forms the UAV cluster intelligent attack system through cascade.The rule set is the core of fuzzy inference.The fuzzy inference system uses the rule set to calculate the relationship matrix to complete the conversion from input to output to realize the decision-making control of the UAV.（2）Aiming at the optimization problem of the UAV cluster intelligent attack system,this paper adopts the differential evolution algorithm to train the UAV intelligent cluster attack control system.By proposing a fuzzy inference system encoding rule,the rule set and membership function in the cascaded fuzzy inference system are encoded into character arrays,and finally a parameter adaptive improved differential evolution algorithm is proposed to represent the UAV with character arrays.The cluster intelligent attack control system is optimized for training.（3）Aiming at the problem of low training efficiency of the UAV cluster intelligent attack system,this paper uses fuzzy Q learning to train the fuzzy system.Through the learning of the Q value,the rule library of the attack decision module of the UAV cluster intelligent attack system is explored.Compared with the differential evolution algorithm through the fitness function at the end of each round of the test,the fuzzy Q learning Training is more efficient and has the advantages of online evaluation.The experimental results show that the UAV attack algorithm based on fuzzy logic control theory can control the UAV to better complete the combat mission.At the same time,the design of the UAV control system based on fuzzy logic is simple,easy to understand,and has high application value and research significance.