ANALYSIS OF THE ACCURACY OF THE SOLUTION TO THE AIRCRAFT GRAVITY GRADIOMETRY PROBLEM

At the present time, more and more attention is paid to the construction of maps of anomalies of the Earthʹs gravitational field by means of aerophotography. Utilization of such maps is promising for gravitational prospecting. The introduction of a gradiometer into the measuring equipment of the aircraft gravimetric system was discussed as early as in 1980s [1]. The gradiometer provides information about the gravity gradient. The advantage of this approach is that a gradiometer is able to separate the gravitational effects from the inertial effects by measurements. When using a gradiometer, it is not necessary to be linked with an a priori chosen model of the Earthʹs gravitational field. However, it was impossible for a long time to measure the gravity gradient tensor with an acceptable accuracy. For this reason the aircraft gravimetry involving gradiometry was considered to have no prospects. At the present time, substantial progress in the improvement of the performance of gravimetric devices has begun to develop. This progress is associated with the development of essentially new technologies and an access to superaccurate navigation information from satellite navigation system (SNS). This makes further investigations in this field promising. When developing algorithms for estimating gravity anomalies, one has to take into account the influence of the instrumental errors on the accuracy of the final result. There are two ways of reducing this influence. One way involves technical improvement of measuring devices. The other way involves thorough analysis of all sources of the errors in order to take into account their influence as efficiently as possible. In this case, the question arises about the choice of the technique to take into account various perturbations. Some of the perturbation parameters, which are most important, can be included in the state vector of the dynamical system. Other parameters can be ill-conditioned, and it is reasonable to treat them as uncontrolled perturbations. Some parameters can be neglected. There are only few works devoted to errors of the gravity gradiometer. In [3], a statistical model is suggested the parameters of which are identified experimentally, the source of specific errors being not considered. Another approach was suggested in [4], where errors of various nature in the output signal of the gravity gradiometer are allowed for by correction coefficients. In the present paper we suggest a model of errors which is traditional for the inertial navigation [5, 6].