: 2011-03-19 11:53:41
Description: The thesis focuses on the extended Kalman filter (EKF), unscented Kalman filter (UKF) and the improvement of unscented Kalman filter (MAUKF) algorithm research, researched the basic principles of three algorithms and their characteristics. Extended Kalman filter (EKF) is the local linear Kalman filter, the algorithm and computation are simple in weakly nonlinear and Gaussian environment. Unscented Kalman filter (UKF) is determined using a series of samples to approximate the state posterior probability density. Improved UKF algorithm (MAUKF) is introduced attenuation factor based on the UKF.