| Abstract: | In this paper a square root covariance form of the Kalman filter is presented which is in many ways the covariance analog of the Paige and Saunders information filter and shares many of the advantages of their implementation. For example it is a true square root implementation and does not just evaluate the basic filter equations making use of recurred square root factors. Also, it provides a compact, convenient, and effective square root implementation of the interpolation smoother. The algorithm is based on the Duncan and Home generalised least squares form of the filter equations, but uses these in a recursive manner. The same ideas can be applied to the full system, and this is also discussed. |
|---|