Autonomous unmanned aerial vehicles (UAVs) require highly reliable navigation information. Generally, navigation systems with the inertial navigation system (INS) and global navigation satellite system (GNSS) have been widely used. However, the GNSS is vulnerable to jamming and spoofing. The terrain referenced navigation (TRN) technique can be used to solve this problem. In this study, to obtain reliable navigation information even if a GNSS is not available or the degree of terrain roughness is not determined, we propose a federated filter based INS/GNSS/TRN integrated navigation system. We also introduce a TRN system that combines batch processing and an auxiliary particle filter to ensure stable flight of UAVs even in a long-term GNSS-denied environment. As an altimeter sensor for the TRN system, an interferometric radar altimeter (IRA) is used to obtain reliable navigation accuracy in high altitude flight. In addition, a parallel computing technique with general purpose computing on graphics processing units (GPGPU) is applied to process a high resolution terrain database and a nonlinear filter in real-time on board. Finally, the performance of the proposed system is verified through software-in-the-loop (SIL) tests and captive flight tests in a GNSS unavailable environment.

In this paper, we present recent work on the g-sensitivity error of the MEMS vibratory gyroscope. Generally, the g-sensitivity error has been ignored in the use of commercial MEMS vibratory gyroscopes, but it deserves our attention if we are to achieve tactical grade performance for military applications. First, we mathematically show the reason the g-sensitivity error occurs as an additional scale-factor error during the use of MEMS vibratory gyroscopes. Then, we estimate the g-sensitivity error using FEM simulation and verify it by experiment using a centrifugal machine. Consequently, we propose a compensation model to accommodate the g-sensitivity error of a gyroscope and confirm the theoretical prediction with experimental results.

A calibration method for a strap-down inertial navigation system (SDINS) is presented here. The body-frame drift induced by the variation of the accelerometer sensing axe due to temperature changes is an important source of navigation error not considered in conventional calibration methods. In this paper, we propose an advanced calibration method that can compensate for SDINS error sources including not only bias, scale factor, and misalignment but also the body-frame drift by utilizing a 2-axis turntable. The body-frame drift is estimated using horizontal accelerometer measurements during a calibration procedure. Simulation results show that the body-frame drift behaves like an additional gyro bias error and that the proposed calibration technique can significantly improve pure navigation performance by removing the effect of body-frame drift.y x C = Direction cosine matrix relating a vector's components in x frame to y frame C a = Acceleration input into the C frame C w = Angular rate input into the C frame C i a δ = i-axis acceleration error in the C frame C i w δ = i-axis angular rate error in the C frame C i a = i-axis true specific force in the C frame C i w = i-axis true angular rate in the C frame i α = i-axis accelerometer bias ii α = i-axis accelerometer scale factor error ij α = i-axis accelerometer misalignment toward jaxis (i ≠ j) i β = i-axis gyro bias ii β = i-axis gyro scale factor error ij β = i-axis gyro misalignment toward j-axis (i ≠ j) R′ = Pseudo-reference frame determined by the offset angular position of a turntable T B =Table body frame M B = Body frame of mount fixture installed between the SDINS and the turntable ' T T B R B w = Angular rate of the turn table R IR w = Earth rate Γ = Initial angular position error matrix of the turntable M = Misalignment matrix between the IMU body frame and the turntable body frame a A = Non-orthogonal sensing axes vector of accelerometer triad g A = Non-orthogonal sensing axes vector of gyro triad a A R ε = Misalignment matrix between A a and R g A R ε = Misalignment matrix between A g and R X = Calibration coefficient vector Z = Measurement vector H = Observation matrix

Three stage in-flight alignment method which estimates error state vector of a strapdown inertial navigation system(SDINS) based on the GPS is newly presented. The three stage in-flight alignment consists of a coarse horizontal alignment, coarse azimuth alignment, and adaptive Kalman filter which shapes error covariances in real time. In the coarse horizontal alignment stage, the horizontal attitude is estimated using the acceleration differences between a SDINS and GPS as damping coefficients. In the coarse azimuth alignment stage, the filter composed of heading increments of a SDINS and GPS tracking angle solves the SDINS heading. At the last stage, the adaptive Kalman filter estimates the optimal error covariances and compensates the error state variables. With this three stage in-flight alignment, it is proposed that the performance enhancement of in-flight alignment and stability increment of filter can be achieved. The error covariance of an adaptive Kalman filter is divided into two parts. The one is the periodically updated covariance starting with the fixed initial statistical values, and the other is statistical weighed gain which is calculated to minimize the cost function of an error covariance derived from measurement error residuals. Errors from the statistical process and measurement noise affect directly the error covariance. So they cause an estimation accuracy and stability degradation. This paper also shows that the estimation precision degree and stability of the filter can be improved by compensating the displacement of an error covariance.

No abstract