The problem of processing measurement information with changing status of measurement results of micromechanical sensors of an intelligent on-board measuring system of a small unmanned aerial vehicle is considered. While vehicle is in the air, the measurement results status changes from confirmed to orienting, for example, due to sensor defects, degradation of the measuring channel, the appearance of false measurement output signals of micromechanical sensors due to vibrations and shocks caused by the movement of air masses. As a result, the probability of stability loss of a small unmanned aerial vehicle increases and it is necessary to raise the accuracy of estimating its orientation in conditions of changing status of measurement results. The measuring procedure of the Kalman structure is considered, the equations of which are determined with accuracy to the parameters of the transition and noise matrices of the state, as well as the perturbation vector, characterizing the measuring process. The parameter values are determined by a mathematical model for converting measuring information based on a dynamic mathematical model, which distinguishes the developed measuring procedure from a measuring procedure with a classical transition matrix. A neural network is used to find unknown parameters. A multilayer perceptron was selected as the basis of a neural network, for which an error back propagation algorithm is used to train. Based on the results of mathematical modeling and measurement experiment, it was found that the accuracy of the synthesized measuring procedure based on a dynamic mathematical model is higher than the accuracy of the measuring procedure of the Kalman structure with a classical transition matrix. The results of the study will be useful in the development of intelligent measurement procedures for on-board measurement systems operating under conditions of changing status of measurement results