Location Tracking And Trajectory Prediction For Underwater Incoming Target

With the rapid development of underwater weapons,torpedo and other incoming targets are characterized by high speed,strong maneuverability and good concealment.The threat to the warships is becoming more and more serious in modern naval warfare.For successfully defending against the attack of the underwater incoming target,we must improve the comprehensive ability of early warning,location,tracking and prediction for the incoming targets.Therefore,it is imminent to build a underwater incoming target defense system with good performance of early warning recognition,location tracking and prediction.This paper focuses on the location,tracking and prediction technology of the underwater incoming target.Using Two vector hydrophone to locate the target,the baseline length and azimuth accuracy is the key factor affecting the accuracy of positioning.Increasing the baseline length or improve the precision of azimuth estimation,can effectively improve the positioning accuracy.In order to improve the azimuth estimation precision of the target at high speed and low SNR,the method of instantaneous correlation integral is adopted in this paper.The effective tracking can largely improve the positioning accuracy of incoming targets,and whether the maneuvering models match with the actual motion state of the target will decide the effect of tracking.Different maneuver models have their own application,this paper simulates the application of CA,CV,Singer,CS model for different target maneuvering.It is found that a single model is usually unable to accurately describe the complex maneuver of a target.Therefore,the IMM algorithm is used in this paper to improve the tracking performance of complex maneuvering target.It is verified through simulation and lake test data that the IMM algorithm's ability to track complex maneuvering targets is better than that of a single model.This paper studied the prediction effect of Kalman prediction and grey model prediction,the results show that this two methods can effectively forecast.This paper also presents a grey Kalman prediction algorithm which combines grey model with Kalman prediction.The validity of the proposed method is verified by simulation and lake test data.The results show that the forecasting effect of the gray Kalman prediction method is better than the former two algorithms.