Vector gravimetry using a precise inertial navigation system continual
ly updated with external position data, for example using GPS, is stud
ied with respect to two problems. The first concerns the attitude accu
racy requirement for horizontal gravity component estimation. With cov
ariance analyses in the space and frequency domains it is argued that
with relatively stable uncompensated gyro drift, the short-wavelength
gravity vector can be estimated without the aid of external attitude u
pdates. The second problem concerns the state-space estimation of the
gravity signal where considerable approximations must be assumed in th
e gravity model in order to take advantage of the ensemble error estim
ation afforded by the Kalman filter technique. Gauss-Markov models for
the gravity field are specially designed to reflect the attenuation o
f the signal at a specific altitude and the omission of the long-wavel
ength components from the estimation. With medium accuracy INS/GPS sys
tems, the horizontal components of gravity with wavelengths shorter th
an 250 km should be estimable to an accuracy of 4-6 mgal (mu g); while
high accuracy systems should yield an improvement to 1-2 mgal.