Maneuver Target Tracking Based on Grey System Theory

Abstract

The result of maneuvering target tracking based on theory of kalman filter depends on the state model of target and the measurement model. When the state model of target is imprecise or is not described by liner space model of state, the result will be volatilization. However, the establishment of mathematical models is complex and difficult. Sometimes, it is impossible to establish accurate mathematical models. If the established model is adaptive to more situation, the filter will not be volatilization, even if the target is maneuvering. This paper provides a new algorithm by introducing grey model into kalman filter. In course of kalman filter, the forecasting value depends on the state model of target no longer. It forecasts next value by using a few forward estimated values with grey differential equation. It gets the current estimated value by using the forecasting value of previous moment and the observed value of current moment. Not only its accuracy is higher, but also its performance is better. Especially, during target maneuvering, the result of filter is better than the traditional method. Experiment results indicate that the way of maneuver target tracking is feasible.