Research On The Technology Of Maneuvering Target Tracking

Nowadays, with the development of technology, the circumstance of modern war has changed a lot. The circumstance of war is becoming more and more complicated, and there are more and more new weapons come forth, too. All above bring maneuvering target tracking a big challenge. Therefore, many new techniques continuously are introduced to the target tracking for more complicated situation. While, one core part in target tracking is filtering algorithm. Here, the algorithms of Kalman filter are studied.Firstly, the principle of target tracking is introduced in this paper, analyzing the classical target tracking method based on Kalman filter. Through a series of simulations, on the one hand it is known that the Kalman filtering precision is higher than other algorithm filter precision obviously; On the other hand may know the"current"statistical model tracking accuracy is decided by the acceleration bound which determined beforehand and the maneuvering frequency, and the Kalman filtering algorithm inherent flaw has also affected the filter precision. In this paper, Kalman filtering algorithm based on the BP artificial neural network (ANNKF algorithm) is proposed.Secondly, has carried on the detailed introduction to the BP neural network, and aims at its insufficiency, introduced the improvement method. Through the simulation it is known that:①T he BP neural network has stronger self organization, self study and nonlinearity approximation ability;②After the training achieved the allowable error precision the neural network not necessarily has general ability;③The astringency of improvement BP neural network is better than traditional BP neural networks obviously.Thirdly, to the insufficiency of the Kalman filtering algorithm, introduced the BP neural network adjustment principle, and has designed based Kalman filter on BP the neural network, and proposed several measures to enhances general ability.Finally, applies ANNKF in the actual combat, tracks the flight trajectory of airplane in the actual combat with it. The simulation result indicated the ANNKF algorithm is better than KF (Kalman filter) algorithm.