Vector-optimized request used to MAG-μ IMU aiming at influence depression of linear acceleration

The widely-used attitude algorithms for geomagnetic and inertial measurement unit (MAG-μIMU) are extended Kalman filter and complementary filter, which aim to fuse the outputs of the sensors in MAG-μIMU for attitude determination of the carrier. How to determine whether the carrier is moving or not so as to depress the influences of linear acceleration to attitude determination is the key and however not-well-settled problem to those attitude algorithms. In this paper, taking REQUEST as attitude algorithm, a movement decision method is proposed, which regulates the weight of gravity acceleration vector of REQUEST at each sampling time according to its accuracy, modeled by the amplitude of the carrier's linear acceleration vector plus the angle between it and geomagnetic field vector. Our experiments show that, our method performs well, and make REQUEST have both good static and good dynamic performance.