The study makes an approach to the problem of circumlunar spacecraft navigation using the measurements from the global navigation satellite systems (GNSS) GLONASS, GPS, Galileo and BeiDou. Algorithms have been developed for determining the orbits of low- and high-orbit circumlunar spacecraft, based on the method of dynamic filtering of pseudo-range measurements from "reverse" navigation satellites (NS). The solution to the navigation problem has been simulated by the measurements from four GNSS, and by those from the NS of GLONASS and GPS only. Accuracy and dynamic characteristics of the obtained solutions have been determined and compared to similar solutions for geostationary spacecraft.

The paper considers the methods of high-precision navigation of space geodetic systems. A technology of determining the orbit parameters by kinematic and dynamic methods using GLONASS signals is proposed. It is for the first time that experimental estimates of position errors have been obtained in a study case of Geo-IK-2 spacecraft, with the measurement residuals of the global quantum-optical network being at the level of 0.06 m (RMS).

The paper presents a comparative analysis of the extended Kalman filter (EKF) and the sigma-point Kalman filter (SPKF) applied to solve the problem of SINS/GNSS integration based on a loosely-coupled integration scheme. Complete stochastic measurement models of MEMS inertial sensors are considered. The efficiency of the EKF and the SPKF is evaluated using real experimental data on complex motion from an SINS based on MEMS technology and a GNSS receiver with a double antenna. The estimation accuracy of navigation parameters using the EKF and the SPKF in the presence of the GNSS signal and during the GNSS outages is analyzed. The results of the statistical analysis of the errors in estimating navigation parameters for different periods of GNSS signal outage are considered.

The paper discusses the method that was proposed earlier to provide nonperturbation of dead reckoning (DR) owing to a single-channel inertial vertical, constructed with the use of a triad of accelerometers and a single free gyroscope, as well as compensation for the effect of inertial accelerations directly in computed DR using the data from an external speed meter (a log). The DR method errors specific to this scheme are analyzed, in particular, those conditioned by the fact that, in the general case, positions of the accelerometers and the log for marine underwater and surface vessels do not coincide either with each other or the center of the vessel rolling. Analytical calculations and the simulation results are given to show that the level of DR method errors is insignificant for the class of the objects under consideration.

Ring Laser Gyroscopes (RLGs) are widely used in many airborne and navigation systems for accurate measurement of the true rotation of the body movement. But the RLG’s suffer a serious problem at low frequencies known as Lock–in frequency. To avoid lock-in problem, the RLG is vibrated mechanically to a high frequency which is known as Dithering. In order to get the true rotation of the body the dither signal has to be removed. Single stage, multistage and multirate filters are suggested to remove the dither signal. These filters have the disadvantage that either the FIR filter length is too large or the phase characteristics are nonlinear. In this work, multiresolution Wavelet Transform (WT) techniques are used to remove the dither signal. Five level multiresolution analysis is carried out with various types of wavelets like Discrete Meyer and Daubechies 45 (db45) etc. With none of the standard wavelets, the original and reconstructed signals are matched. A new wavelet is designed to remove the dither signal. The required signal can be reconstructed back using the approximation coefficients at level 5. The dither signal is attenuated by 107.0 dB, and the phase characteristics are found to be linear in the pass band. The computational complexity is also less compared to the three stage combined filter reported earlier.

The paper addresses the practical implementation of a calibration algorithm for a magnetometer integrated in an electronic device, using synchronous measurements of a gyroscope. Recurrent expressions have been derived for accumulating the intermediate matrices, due to which there is no need for accumulating the full set of primary measurements of vector gauges. An algorithm has been formulated for determining the time point when accumulation stops and calculation of calibrated parameters starts.

The Earth-centered, Earth-fixed (ECEF) frame does not have any specific points in the near-Earth space; therefore, it can be used as the basic one for navigation in the polar regions. The paper considers a new algorithm for calculation of ECEF coordinates. The proposed algorithm is complementary to the basic strapdown INS algorithm, which

functions in its normal mode from aircraft takeoff to landing. In the polar regions, the aircraft control law that supports flights on the point-topoint principle is based on calculation of ECEF coordinates.

The paper considers modification of a reaction wheel current control loop based on a fuzzy controller trained by a genetic algorithm. The control logic maintains the motor current which can be represented as a sum of two components, one of which is proportional to the input signal, and the other one corresponds to the error of control moment implementation. It is shown that the system with a fuzzy controller, which implements the variable gain on the error channel, eliminates the torque pulsation and reduces the time of transient processes while adjusting the control actions. The system operation has been simulated by means of MatLab Simulink software to confirm viability of the proposed control loop. The results of the work can be used in developing the advanced systems of spacecraft attitude control.