Realization of Maneuvering Target Prediction in Image Tracking System

摘要

A method of maneuvering target prediction is presented for image tracking system on vehicle. The target motion is divided into two parts, namely global motion and local motion, which are separately predicted according to their own dynamic behavior. This paper puts its emphasis on local motion prediction. The maneuvering target tracking (MTT) technique is studied in order to build a good tracker, which is then extended into a predictor to estimate the future moving state of target, thereby putting forward "interacting multiple-model(IMM) prediction algorithm". As the simulation shows, this algorithm has high prediction precision and good self-adaptive capability. Besides, the prediction gate computation is given.