Fixed-Wing UAV Integrated Navigation with Low-Cost IMU/GPS

14 januari 2011 | 14:00
plaats: Collegezaal B
door B. A. Hummelink

For the graduation research one particular UAV platform is chosen for its fast dynamics, a small fixed-wing UAV. To further increase the use of fixed-wing UAV’s, their navigation filters must be robust and stable.

Afstudeer presentatie

 

Abstract - Today, there is an increase in the use of Unmanned Aerial Vehicles (UAV’s), for applications that can be considered dull, dirty or dangerous when compared to applications of conventional aircraft or helicopters. For the graduation research one particular UAV platform is chosen for its fast dynamics, a small fixed-wing UAV. To further increase the use of fixed-wing UAV’s, their navigation filters must be robust and stable. The problem with navigation filters it that when properly designed with fully theoretical observable states a diverging and unstable solution can still be the case in practice. Based on this the research goal was defined as, define the minimal requirements of sensors and other UAV hardware for an Inertial Navigation System (INS) such that the navigation filter requirements imposed by small fixed-wing UAV dynamics and size can be satisfied, under the constraint of using low-cost IMU/GPS with MEMS sensor technology.

Based on derived fixed-wing aircraft kinematics together with the coupling of an IMU and GPS receivers, a set of Angle Correction (AC) equations is derived that serves as angle observers. With this set of AC equations with it now possible to bound the attitude/heading angles estimation errors, which has been validated by a developed observability analysis tool that can directly link singular values to system states. An alternative state identification algorithm was explored to the conventional (Extended) Kalman filtering by utilizing the coordinate transformation RBE(f,q,y) between a North-East-Down (NED) and body-fixed frame of reference. The antisymmetric transformation matrix fits in the category of Special Orthogonal Lie groups with a dimension of three, called SO(3). Based on the geometry of the SO(3) Lie group, a non-linear Passive Complementary Filter (PCF) has been designed that doesn’t require linearization and matrix inversions.

Based on a time synchronization expression for an IMU/GPS integration the overall navigation performance is dependent on the combination of filter innovation magnitude and updating frequency, IMU accuracy, time synchronization between IMU/GPS and platform dynamics. The results show that with the set of AC equations, the attitude/heading angles become instantaneously observable resulting in an accurate and stable IMU/GPS + AC navigation filter.

 

© 2014 TU Delft

Metamenu