Segmented Angular Rate Joint Estimation of Inertial Sensor Arrays for UAV Navigation

A segmented angular rate joint estimator for fusing the measurements in an inertial sensor array is presented with higher precision and lower calculation burden, which offers better performance to unmanned aerial vehicle (UAV) navigation. In recent years, microaerial vehicles are increasingly indispensable in various industries. Attitude control is crucial for normal flight; therefore, the precision of angular rate measurement is the determinant for attitude calculation based on inertial measurement unit (IMU). The rapid development of MEMS industry, in addition to heterogeneous sensors, has given birth to multiple methods for improving the precision and reliability of angular rate measurement using homogeneous sensors called inertial sensor array. Computing angular motion based on inertial sensor array, including gyroscope and accelerometer triads, is a joint optimal estimation process. Due to the scale effect of accelerometer array, it is also sensitive to the angular motion. Both types of sensors can form a complementary filter at sensor level. In this article, we have proposed extended Kalman filter (EKF) and Levenberg-Marquardt (LM) framework for angular velocity estimate of inertial array system and fused them in the form of subsection measurement to give full play to the characteristics of the two methods and two types of sensors. Monte-Carlo simulation shows that the precision of angular velocity estimate is improved by up to 4.57%, and the calculation amount of the algorithm is greatly reduced.