Application of the Three State Kalman Filtering for Moving Vehicle Tracking

Abstract

The three-state Kalman filter (KF) is applied in the optimal estimation of three state (position, velocity and acceleration) in a moving vehicle; the problem is modeled like linear time invariant (LTI) system in presence of additive white Gaussian noise (AWGN). The steady-state filter parameters have been simulated and analyzed for different process acceleration noise (covariance). We show that KF estimation produce minimum mean square error (MSE) if acceleration noise and measurement noise are lower.