Study of strong tracking augmented unscented kalman filter in integrated navigation system

In order to solve the problem of inaccurate state estimation and divergent outputs of the filter of the low-cost integrated navigation system, a strong tracking augmented unscented kalman filter is proposed in this paper. This method extends the strong tracking filter principle into the augmented unscented kalman filter, which improves the strong tracking ability of the system states mutation. Using the state switching technology to reduce the dimension during the Sigma points´ sampling and this improves the real-time property of the filter. Applying this method into the low-cost integrated navigation system, the experiments results prove that this method can track the state mutation quickly and inhibit divergent outputs of the filter.