A MIMU/Polarized Camera/GNSS Integrated Navigation Algorithm for UAV Application

Abstract

Many insects such as ants, honeybees, butterflies and so on, have taken advantage of the property of sensing polarized light to determinate the heading angle and find the way back home. More and more bio-inspired polarization navigation approaches have been researched by navigation experts because of the new navigation’s advantages of efficiency and reliability. For many small UAVs the navigation system usually consists of MEMS-inertial measurement unit (MIMU) and GNSS (Global Navigation Satellite System) receivers. And magnetometers are used to measure the yaw angle. But magnetometers are easily disturbed or interfered by power supply and/or iron, steel material. This paper discusses a MIMU/polarized camera/GNSS integrated navigation algorithm for UAV application. A polarized camera is designed to sense the polarization angle and polarization degree of the natural sun light. We have developed a single-chip polarization imaging sensor. Polarization imaging sensor utilizes pixelated linear polarization filters deposited on an array of silicon-based vertically CCD photodetectors. There are four pixelated polarization filters oriented nominally at 0°, 45°, 90°, and 135°. The 2 by 2 pattern is used to calculate the polarization information. Based on the time and position provided by GNSS receiver and the horizontal angles measured by MIMU, yaw angle can be obtained by polarized camera. The accuracy of yaw angle is dependent on the sky and weather status. This paper proposed a MIMU/polarized camera/GNSS integrated algorithm based on Kalman filter. The yaw angle is observed because polarized camera is used. And some land vehicle field experiments were carried out to demonstrate the feasibility of the new algorithm. The accuracy of yaw angle is better than 1° and can be applied to UAV navigation.