Research On The Localization Method Of Radiation Source Based On UAV Swarm

With the development of swarm intelligence sensing technology,unmanned aerial vehicles(UAVs)with low cost and intelligent characteristics are further developing towards the direction of swarming.In the modern complex electromagnetic environment,UAV swarms rely on the advantages of anti-destructive persistence and information sharing to become a new type of collaborative system platform,providing a new solution for highprecision radiation source sensing,but also facing new challenges.In the information society,as the number of swarms increases,the disadvantages such as dense interference in the electromagnetic environment and scarce communication resources become more serious,and it is difficult for swarms to work together.Then it affects the accuracy of the high-precision location information of radiation source.Therefore,this dissertation studies the method of locating radiation source based on UAV swarms.The main work of the paper is as follows:1.In order to solve the problem of high complexity caused by the large number of UAV swarms,a station selection method based on distance rules and angle rules is proposed.Firstly,a three-dimensional time difference of arrival(TDOA)positioning model is established according to the principle of passive TDOA positioning,and the influencing factors of multi-station passive TDOA are studied.By analyzing the influence of base station numbers,baseline length,and angles on the positioning performance under twodimensional and three-dimensional conditions,a station selection strategy based on UAV swarms is proposed.This strategy can select the optimal positioning UAV clusters without constraining the swarms movement parameters,take into account the cost of positioning efficiency and positioning efficiency,and reduce the coordination complexity.2.In the highly dynamic environment,the decrease of the stability,anti-interference ability,and survivability of UAV swarms will cause the loss of the original positioning base station and the decrease of positioning accuracy.For this problem,a corresponding dynamic reconnaissance and station selection strategy is proposed,and the mathematical modeling is performed.A joint optimization model of positioning is proposed from the perspective of space-time fitness and target adaptability.Using the unique role conversion mechanism of artificial bee colony algorithm(ABC)to convert the roles of UAVs to realize self-organization clustering,a dynamic station selection algorithm for UAV swarms based on artificial bee colony algorithm is proposed.The algorithm is used to simulate and analyze single,multiple,stationary and moving targets,and verified the localization performance of UAV loss.Finally,the localization error is analyzed and the station selection process is visualized using the visualization software Satellite Tool Kit(STK).The simulation results show that the strategy still has the ability to quickly achieve high positioning accuracy with low energy cost despite a certain UAV loss rate.3.In the moving target positioning,the swarm ensures localization performance through frequent station selection,but the computational complexity increased.Aiming at this problem,a tracking algorithm module is added to reduce the selection times while ensuring the positioning performance.This dissertation studies the tracking problem under the time difference positioning algorithm,and establishes the corresponding target state model and observation model.By analyzing the existing nonlinear filtering algorithms,the interacting multiple model algorithm(IMM)is introduced to remedy the defects of the single model.On this basis,the IMM transition probability is adaptively updated to improve the model matching and realize the real motion trajectory tracking.Comparative experiments show that the Adaptive Markov Parameter interacting multiple model algorithm Extended Kalman Filter(AMP-IMM-EKF)algorithm can complete the model switching within 5s,which verifies the effectiveness of the algorithm.