Simulation Research On Large-scale UAV Cluster Network Based On Biological Cluster Intelligence

In recent years,with the miniaturization of unmanned aerial vehicle(UAV),intelligent and deflationary force,people in more and more occasions using UAV.In order to deal with the task scenarios that are difficult for a single UAV,the concept of UAV cluster came into being.Using large-scale UAV cluster is largely reduced the difficulty of the task,but then there is a series of cluster control problem: cluster,how to reduce the dependence on central node of the cluster mobile how efficient and try to avoid conflict,how to maintain a relatively stable cluster topology but can quickly under emergency strain,members of the cluster how to balance energy consumption,etc.Aiming at the problems faced by large-scale UAV clusters,we designs a flexible,efficient and adaptable cluster mobile model architecture based on biological cluster technology.The mobility model architecture designed in this paper is established based on the rules of SAC mobile model of bird flock.Through this mobility model,decentralized cluster movement is realized,which reduces the dependence of the cluster on the ground control platform and reduces the possibility of conflicts between UAVs in the cluster.In this paper,the mobility model is integrated into a communication model to verify its effectiveness through simulation.Based on the SAC mobility model,we studies three representative services of large-scale UAV cluster: network aggregation,path planning and clustering algorithm,and improves them.Our algorithms simplify the process of network assembly and reduces the dependence of the cluster on the ground control center.The path planning algorithm can reduce the damage in cluster flight.The clustering algorithm equalizes the cluster energy consumption and prolongs the cluster survival time by dynamically changing cluster heads.In order to verify the above mobility model architecture and algorithm,this paper use the NS-3 network simulator set up and implement the architecture and algorithm,and the artificial fish moving model and comparison between the algorithm and improve the business before the tests show that the presented mobility model architecture and the improved algorithm can enhance the flexibility to adapt to all kinds of test scenarios unmanned UAV cluster efficiency,and reduce the loss in the process of cluster in the task.To sum up,this paper has done useful exploratory work on the large-scale UAV cluster network architecture,and obtained some research results with academic value and application prospect on the network aggregation,path planning and clustering algorithm based on the biological cluster mobile model.In this paper,the research work for UAV cluster management intelligent control to further play a role in promoting.