Tracking Maneuvering Trajectory under a Boosted Kalman Filtering

Abstract

In this paper, an adaptive Kalman Filtering method is presented for the state prediction of random systems. It is shown that the adaptive Kalman Filtering method in conjunction with equilibrium optimization solution can estimate the initial accelerations of targets effectively since the equilibrium optimization solution tunes the state prediction vector to diminish the error between measured value and prediction estimation value. We evaluate our model on special and random trajectories. Experimental evidence shows that the proposed method can robustly estimate an initial acceleration from a dynamic model and stably track a trajectory which is moving randomly.