Using GPS, IMU and Laser Ranger Technology to Obtain an Optimal Global Gravity Model Based on Intersatellite Range-Rate Change Data

David Gleason

Abstract:	The AF is examining how i; gravitational intersatellite range-rate changes can help build an optimal global gravity model to replace the WGS84 geopotential model. Two satellite to satellite tracking (SST) modes are being pursued: (1) a “high-low” mode wherein a single “low” satellite would be tracked by the multiple “high” GPS satellites and (2) a “low-low” mode wherein an intersatellite, continuous wave laser ranger would also reside on two co-orbiting “low” satellites. The non- gravitational forces acting on the “low” satellite(s) would be measured by advanced cryogenic IMU while those acting on the GPS satellites can be modeled. Using the laws of potential theory on the ‘i> data can yield an Earth surface set of gravity disturbances. The accuracy of the disturbance set obtained from each SST mode is the same provided the i; noise and sampling rate, mission duration and “low” altitudes are the same. The ‘b accuracy spec being sought is 0.1 mGa1 (=10w7g) at 1 Hz. Although new, all digital GPS receivers should reduce the averaging time required to smooth out the noise in the 5 values, it is still questionable whether a GPS/IMU payload can meet the spec. If it is supplemented with the intersatellite laser ranger interferometry the spec should be met. Signal to noise spectral error analyses and a fully fledged simulation both revealed that (o=O.lmGal, 1 Hz, alt=300km, duration=60 days) p data can produce 1 deg mean disturbance values on the Earth’s surface to an accuracy of 5 mGals. Instability problems related to the downward continuation process were solved with singular value decomposition techniques. The global, uniformly accurate set of disturbances (currently nonexistent) can produce a geopotential coefficient set out to degree & order n=m=180 using the principles of Least Squares Collocation (LSC). LSC minimizes the coefficient quadratic errors due to 1) the noise in the input mean disturbance values and 2) the discretization of sampling errors arising from the finite size of each 1 deg cell over which a mean value is assigned. The current, arguably restrictive LSC assumption that the disturbance errors are uncorrelated (needed to make LSC tractable) could be dropped with new computer architecture. This assumption is one reason the lower degree coefficient error estimates are better if obtained from ground-based satellite tracking observations. Ground-based DGPS tracking observations of the “low” and TOPEX satellites could be part of a final adjustment that could also include the disturbance data and/or SST-LSC model. If the final “combined” model recognizes as many correlated noise sources as possible it would truly be optimal. Estimated error spectra plots of the proposed and existing models illustrate the improvements the former offers.
Published in:	Proceedings of the 49th Annual Meeting of The Institute of Navigation (1993) June 21 - 23, 1993 Royal Sonesta Hotel Cambridge, MA
Pages:	251 - 260
Cite this article:	Gleason, David, "Using GPS, IMU and Laser Ranger Technology to Obtain an Optimal Global Gravity Model Based on Intersatellite Range-Rate Change Data," Proceedings of the 49th Annual Meeting of The Institute of Navigation (1993), Cambridge, MA, June 1993, pp. 251-260.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In