هادی مکارم

Star tracker is one of the most important devices used on satellites for attitude determination. Since its output is discontinuous, it needs to be aided to complement its discontinuity. Using gyroscope unit is the most suitable choice for aiding the star tracker. However, using these two kinds of sensor simultaneously has some challenges. In other words, not only biases lead to low accuracy in the attitude determination, but also the installation error has a significant effect on the accuracy. In this paper after presenting the important role of installation errors between star tracker and gyroscope in the accuracy of attitude determination, an effective method is proposed to determine the misalignment error between these two sensors which is based only on their measurements, and the mathematical formulation is presented in detail. Then, to validate the performance of the proposed method, it is implemented to calculate the instantiation error of an experimental dataset gathered in the Mount Pooladkaf, for which the results are reported.

Navigation errors in synthetic aperture radar (SAR) applications lead to phase errors in SAR echo signal and image quality degradation. Among different phase errors, zero and first order errors have no effect on image quality, while second and higher order terms are very important. The majority of second and higher order phase errors are due to initial alignment errors, and bias and random errors of inertial sensors. Integrated INS/GPS navigation systems are used for solving this problem. However, discontinuities in integrated navigation data lead to severe SAR image quality degradation. In this paper, the effect of different navigation approaches on SAR azimuth impulse response was studied and the importance of specially-designed algorithms to integrate INS and GPS was illustrated. Then, the proposed algorithm for navigation in SAR application was presented according to navigation errors behavior, and its benefit in azimuth impulse response, and imaging quality was illustrated. ‌‌

Antenna positioning is very important in synthetic aperture radar systems and it is necessary in the design of a navigation subsystem to account for different sources of error and their effects on the accuracy of navigation results. According to the error behavior of inertial and satellite navigation systems, integration of inertial sensors with satellite positioning systems is a common method to achieve high accuracy navigation results. However, synthetic aperture radar considerations, lead to some problems in the utilization of integrated navigation results in the imaging period. Therefore, only the inertial navigation results are used in the imaging period, and integrated navigation results are used just as the initial conditions for the algorithm. This paper studies the estimation of these initial conditions and predicts the navigation error growth caused by them. Since the extended Kalman filter is the most common tool for the integration of inertial sensors and satellite data, the elements of the corresponding state covariance matrix represent the accuracy of integrated navigation results. In this study for a synthetic aperture radar flight scenario, in which the nominal path is a straight path with a constant velocity, the state covariance matrix is calculated analytically both in the steady-state conditions and after the GNSS data outage. These analytical results are verified with simulations. Specifically, the estimation accuracy of antenna position, velocity and attitude are calculated with respect to the noise level of inertial sensors and GNSS data. Results can be used in the design and/or selection of a proper navigation system in synthetic aperture radar applications.

Motion control of a planar nonholonomic system with four DoF is addressed in this paper. Three actuators are responsible for shape control of this system. Furthermore، assuming no external forces and zero angular momentum، imposes a nonholonomic constraint to the problem. First it is shown that although the simplified equations of motion for this system، could be converted to Heisenberg and chained-form systems، the conventional control methods for these systems، may not be applied to the considered problem. Then، using sliding modes and online path planning، two different closed-loop control laws are designed for bringing the system to and stabilizing around any desired equilibrium state started from any initial condition. Simulation results، show the efficiency of the proposed methods.