Research On Construction Method Of Information Interaction Topology For UAV Swarm Formation Keeping

High stability of information interaction topology(network topology)is an important foundation for multi-node joint detection,high dynamic network transmission,distributed cooperative control and intelligent decision making of the UAV(Unmanned Aerial Vehicle)swarm system.In three-dimensional dynamic application scenarios,fast and stable network topology construction is of great significance for the reliable applications of UAV swarm system.However,current topology construction methods mainly take the topology stability as an evaluation indicator after the topology construction rather than a criterion in the process of topology construction.To address this issue,based on the criterion of topology stability,this dissertation investigates the topology construction model of UAV swarm network and typical distributed and clustered UAV network topology construction methods.The main work and innovative contributions of this dissertation are listed as follows:1.A topology construction model for UAV swarm network based on the criterion of topology stability is proposed.Due to the time variability of the task objectives and the inherent highly dynamic characteristics of the UAV swarm system,the relative positions and mobility relationship between UAV nodes changes frequently.A stable network topology is important for network maintenance,network overhead and node energy consumption,and for efficiency and safety of the collaborative of the UAV swarm system.Therefore,a topology construction model for UAV swarm system based on the criterion of topology stability is proposed.The proposed model can maximize the UAV swarm system topology duration while satisfying the system-specific network performance requirements including network connectivity,end-to-end delay,network throughput and node energy consumption.2.A cluster-based topology construction method for UAV swarm system based on the similarity of group trend is proposed.The cluster-based topology construction method has the advantages of simple routing strategy,high interaction efficiency and low maintenance cost,and is of great research significance in large-scale swarm application scenarios.Based on the proposed model,a Group Trend Similarity-based Clustering(GTSC)method with dual-head is proposed for the topology construction of UAV swarm.The proposed GTSC method jointly considers the mobility characteristics of swarm system and the operation mechanism of clustering network.In the process of clustering,based on the group mobility trend similarity of swarm system,nodes with close locations and similar mobility trends are grouped into the same cluster to maximize the network topology duration.In the process of cluster head election,factors related to network communication performance including mobility similarity,energy consumption,inter-cluster distance and inter-cluster transmission hops are fully considered.Simulation results show that the proposed GTSC method exhibits better performance than existing clustering topology construction algorithms in terms of topology duration,network throughput,end-to-end delay and energy balance in large-scale UAV swarm scenario.3.A rapid PSO-based distributed topology construction method is proposed for UAV swarm system.The distributed network topology is decentralized and can perform the information interaction without relying on control and forwarding of specific nodes,and therefore has the advantage of high invulnerability,but tend to be slow to converge.Therefore,based on the proposed topology construction model,a rapid PSO-based distributed topology construction method is proposed.In particular,in order to achieve fast convergence of the distributed topology construction method,the initial value of the PSO algorithm is optimized by jointly considering the static characteristics and dynamic trends of the swarm nodes,while the update direction and step size of each iteration are optimized based on the feature similarity function.Simulation results show that the PSO method requires an average of 5.5 iterations to obtain the optimal solution for a swarm size of 100 nodes,while the proposed algorithm converges to the optimal solution within an average of 2 iterations while obtaining the same end-to-end delay and network throughput performance.This dissertation focused on the topology construction criterion and corresponding construction methods for UAV swarm network.The proposed method is of strong practical significance and has important theoretical and engineering value for UAV swarm system applications.