Research On Algorithm Of Nonlinear Filtering In Submarine Tracking And Targeting

Due to the complex variability of the marine environment and the increasingly serious situation of the seas in China,submarines are playing increasingly important roles in the exploration of the unknown oceans and the protection of the integrity of our territory and the rights and interests of the surrounding oceans.In the strike of the submarine against enemy ship targets,quick and accurate tracking and positioning is the most important method to improve the operational efficiency of the submarines.Therefore,how to realize the quick and accurate tracking and localization of the submarines against enemy ship targets has become a hot issue for many experts at home and abroad.Firstly,the observability of tracking and positioning system of submarines against enemy ship targets is studied in this paper.The main purpose of researching the observability of the system is to deepen the cognition and comprehension of the observability of the system,and to provide a theoretical foundation and basis for the study of the target positioning and tracking algorithms of submarines.Then based on the Kalman filter,the nonlinear filtering algorithm of tracking and positioning of the submarines against enemy ship targets is studied.This paper firstly introduces the principle of Kalman filtering,which is the optimal filtering method when the system is linear.Actually,however,the positioning and tracking of the submarines against the enemy ship targets is a nonlinear system.Therefore,the two most commonly used nonlinear filters,the Extended Kalman Filtering Algorithm(EKF)and the Untracked Kalman Filtering Algorithm(UKF)are studied in this paper.The performance of EKF and UKF tracking and positioning of enemy ship targets is compared through simulation.From simulation experiments,compared with EKF,UKF improves the accuracy of submarine tracking and positioning of enemy ship targets,and also solves the problem of divergence.In the simulation experiment,after 12 seconds of tracking and positioning of submarines against the enemy ship targets,use the situation that EKF appeared divergence,the accurate positioning and tracking of the enemy ship target could not be achieved,which can reduce the operational efficiency of the submarine in the modern warfare.At this time,UKF can achieve better on tracking and localization against enemy ship targets.For its complex variability of environment of the submarine facing underwater and the issue that the noise it receives may not be Gaussian noise,this paper also studied the filtering algorithm for non-linear,non-Gaussian systems,namely Particle Filter(PF).However,the biggest problem faced when using PF is the problem of particle degradation.In order to solve the influence of particle degradation on accuracy of tracking and positioning of the submarines against enemy ship targets,this paper also studies the improved algorithm of particle filtering,namely the extended Kalman particle filter(EPF)and untracked Kalman particle filter(UPF).Under a same initial condition,simulations of three filtering algorithms are operated.The simulation results show that the UPF algorithm is the best filtering effect in the tracking and positioning of enemy ships.Finally,in order to adapt to the maneuvering of enemy ship targets,this paper also discusses the interactive multi-model filter algorithm(IMM).Through simulation experiments,it can be seen that when the enemy ship targets are maneuvering,the IMM algorithm can quickly and better achieve the positioning and tracking of enemy ship targets.This is of great significance in modern warfare.