Filter Algorithm Research of "Current" Statistical Mode Based on Multi-models Fusion

Abstract

In order to overcome the problem of what traditional "current" statistical filtering model is difficult to determine the self-related time constant as tracking maneuvering target, this paper advances improved "current" statistical filter model based on multi-models fusion. The algorithm firstly provides some "current" statistical filter model which has different self-related time constant, and then obtains filtering result by probability weighting the result of each model. The simulation result shows that the new algorithm has high tracking accuracy compared with traditional "current" statistical filtering model.