Rapid IMU Alignment Method for Small Unmanned Systems Under Large Manoeuvring Conditions

With the increasing use of low-cost and high-reliability unmanned aerial vehicles (UAVs), navigation systems are becoming more sophisticated. These systems typically consist of a micro inertial measurement unit (IMU) and global satellite navigation (GNSS). However, IMUs face limitations in performing initial alignment due to their accuracy constraints. Existing low-cost unmanned system IMU initial alignment solutions commonly utilize magnetometers, but they are prone to magnetic field interference. Moreover, satellite-assisted alignment generally requires the UAV to maintain linear motion during alignment, which can be challenging for small unmanned systems that are unable to maintain regular motion for extended periods and may perform large maneuvers. In this paper, we propose a new method for IMU alignment that combines coarse alignment of the inertial system with fine alignment using dynamic GNSS feedback. We have developed an algorithmic model that leverages GNSS measurements of velocity and position to establish a Kalman filter, This filter is used to estimate misalignment angles, as well as gyro and accelerometer zero biases. Our simulation experiments, conduct-ed under conditions involving large maneuvers, demonstrate that this method can achieve convergence of pitch and roll errors within 0.1° in just 20 seconds of coarse alignment. Furthermore, it achieves convergence of heading angle errors within 0.5° after 100 seconds of fine alignment, while accurately estimating gyro and accelerometer zero biases. Our proposed method holds significant research value for addressing positioning challenges encountered by small unmanned systems.