A fast gravimetric estimation technique.

An algorithm for the rapid computation of minimum variance estimates from large amounts of gravimetric data has been developed as a means of improving the accuracy of inertial navigation systems. Based on an extension of generalized Wiener filtering with finite data, the algorithm increases the accuracy over most of the track while maintaining efficiency through use of the fast Fourier transform. The algorithm, when used with a large-scale computer, permits processing of 50,000 points in about 15 min. An application of the algorithm to simulated gravity data from a point-mass model is demonstrated. Application to area data in two dimensions is under study.