Integration of information – the process of combining data to determine or predict the state of an object. The integration provides an increase in the robustness of the robot control and the accuracy of machine perception of information. The paper discusses one of the methods for integrating of the robot onboard data using the example of the mobile tracked robot data. One of the aim is showing how to obtain the Jacobi matrix of the process function and the Jacobi matrix of the observation function of the mobile robot system for subsequent data integration by the state observer, built on the basis of the Kalman filter, that can convert the robot onboard data into the elements of the state vector of the mobile robot system. Each nonzero element of the Jacobi matrix of the observation function of the system is a weight coefficient. It determines the contribution of an element of the output vector of the system corresponding to this weight coefficient to the result of information integration calculated by the Kalman filter – the state observer. The Kalman filter is a sequential recurrent algorithm for filtering information from discrete dynamical systems specified in the state space. A feature of the Kalman filter as an observer of the state of the system is the assumption that the observed system has the effect of white Gaussian noise (characterized by zero mathematical expectation) on its state.