= PkHk (HkPkHk + V kRkV k ) (2.16)Р - -Р xˆ k = xˆ k + Kk(zk – h(xˆ k, 0)) (2.17)Р -Р Pk = (I – KkHk)Pk (2.18)РAs with the basic discrete Kalman filter, the measurement update equations in Table 2-2 correct Рthe state and covariance estimates with the measurement zk . Again h in (2.17) comes from (2.4), Р Hk and V are the measurement Jacobians at step k, and Rk is the measurement noise covariance Р(1.4) at step k. (Note we now subscript R allowing it to change with each measurement.)РThe basic operation of the EKF is the same as the linear discrete Kalman filter as shown in РFigure 1-1. Figure 2-1 below offers plete picture of the operation of the EKF, combining the Рhigh-level diagram of Figure 1-1 with the equations from Table 2-1 and Table 2-2.Р UNC-Chapel Hill, TR 95-041, July 24, 2006
全文预览
