Research on Target Tracking Based on Unscented Kalman Filter

Aiming at the multi-source heterogeneous of target tracking system information. On the basis of "current" statistical model, this paper researches the unscented Kalman filter information fusion method and analyzes its mathematical model. According to the mathematic model researching of optimal estimation real time tracking algorithm, it be able to describe and process the sensor information of uncertainty characteristics by using fuzzy theory, making an adaptive adjustment of the measurement parameters of the Kalman filter, so as to achieve the system error calibration and measurement error adaptive function. In line with the model of functional parameters, carrying on the simulation of the single sensor filtering error and multiple sensor filter error, by comparing the trajectory and actual motion analysis shows, this method can effectively improve tracking precision and stability. It can avoid external interference; accelerate the convergence speed of response and adapting active period of target tracking measurement requirements. Copyright © 2013 IFSA.