Research On Cooperative Operation Quality Of UAV Swarm Based On K-means And Event Algorithm Fusion

When UAV clustering is carried out in multi-task or multi-trajectory environment,due to the limitations of its own endurance ability,communication conditions,response time of main control system,perceived obstacle avoidance ability and external environment,the operational efficiency of multi-UAV in the actual environment is often lower than the expected effect.In particular,with the increase of the number of UAV clusters,the amount of system calculation increases(N-P problem).Easy to cause the operation system response delay or failure.In this thesis,based on the research of the existing UAV clustering algorithm,considering the complexity of the actual operating environment,the diversity of constraints,the superposition of computing and communication structure differences,and combining the cluster architecture,obstacle avoidance control,Simulink platform construction and other theoretical knowledge,an algorithm model of cooperative operation of UAV group based on the fusion of K-means and Event algorithm is proposed.Firstly,the factors affecting the collaborative operation of UAVs were analyzed,and the main factors such as fusion state estimation,dynamic filtering inspection,cluster network construction and perception avoidance technology were analyzed.Secondly,the K-means and Event algorithm models of UAVs cluster operation were studied,and the corresponding characteristics were simulated to prove the rationality of the two algorithms.Then,based on Euclidean distance fusion,PID filter fusion is selected,and on the basis of fusion,Simulink platform and distributed environment construction is carried out.Finally,a model of UAVs was selected as the experimental object in the simulation,and parameters of the fusion algorithm were selected,and the root-mean-square error of position,maneuverability efficiency of cooperative operation,estimation of noise standard deviation and time-varying noise standard deviation(R_k)were used to test.The results showed that the nodes of each UAV were consistent,and the expected fusion effect was achieved.In addition,the probability of cooperative operation of multiple UAVs is calculated,and it can be seen that the fusion algorithm can make cooperative operation of multiple UAVs achieve the expected result.By studying the fusion algorithm of K-means and Event,a new idea is provided for the cooperative operation quality of UAV groups.This algorithm can not only maximize the performance characteristics of UAV groups,but also avoid the impact of operation errors or faults caused by single UAV,so as to improve the fault tolerance and stability of autonomous detection and recovery of the collaborative system.