Research On The Algorithm Of Strapdwon AHRs Based On MIMU

Recent years, micro-inertial instrument has developed quickly. At the same time, the research of micro inertial measure unit has attracted the attention of scholars all over the world. Attitude and Heading Reference System based on micro-inertial instruments has much smaller size, and cuts development cost substantially. Relying on such notable and unique advantages, micro Attitude and Heading Reference System can be applied to aircrafts, spaceships, cars, biomedical instruments, robots, and so on.The micro Attitude and Heading Reference System is composed mainly by micro-inertial sensors and digital signal processing system. The angular rate outputs of micro-gyroscopes are used to update attitude quartern with the Runge-Kuta method. Micro-accelerometers are used to observe the earth gravity field, and magnetometers are used to observe the earth magnetic field. The outputs of accelerometers and magnetometers can be used to compensate for the gyro bias. Then the attitude quartern error estimates is given by the Kalman filter, and used to compensate for attitude quartern updated.Firstly the importantance of micro-AHRS is introduced, and the development of MIMU and micro-AHRS is given briefly. The principle of strapdown inertial navigation and traditional attitude algorithms are elaborated in detail. Comparison among these algorithms is conducted in the way of tables. Secondly, the scheme of Micro-AHRS is designed elaborately. The system is composed mainly by Micro-inertial sensors and digital signal processing system. Then the principles of sensors are introduced briefly.Thirdly, based on traditional rotation vector method, dynamic error model is constructed. Then corrected rotation vector method is deducted using the information of former update period. Also enhanced rotation vector method is introduced.Finally, the information integrating method is deducted, basing on Kalman Filter. The method can integrate all sensors'output, in order to get attitude quartern error estimates and compensate for attitude quartern updated immediately. The MTi-G producted by Xsens, a company in Holland, has quite high precision while observing the attitude of carriers. So the simulation of information integrating method is done by using the corrected sensors'data of MTi-G.