Mathematical Problems Of Attitude Measurement With Magnetometer And Gyroscope

Inertial navigation system is to mount the inertial element on the carrier. Random drifterror of gyro and designed attitude of the algorithm not only play a key role in the inertialnavigation, but also have affection significantly on the measurement of the inertial navigationsystemaccuracy. So handling the gyro random drift error and the purpose of improvingaccuracy of the choice of the inertial sensors and attitude algorithms in the system is the focusof the study.This paper studies the research of inertial navigation and MEMS gyro random drift error,and gives a brief introduction of the inertial navigation system. This paper analysis detailedseveral algorithms of attitude solution and compares them.Gyro random drift error is established by ARIMA model. In the last part this paperintroduces the structure and the characteristic functions of the time series models; then thispaper measures drift data using time series to modeling pretreatment, test and analyze itsstatistical properties in order to identify model order initially; and then this paper gives themethod of model parameters estimation and several common criteria in order to determine themodel order further based on the principle of minimum criteria; in the last part of this paperanalyzes the applicability of the model test.Simultaneously, this paper introduces the basic theory of Kalman filter and analyse thestability characteristics of the Kalman filter briefly. First this paper introduces the Kalmanfilter theory and gives the basic equation of the discrete system and the solver steps.Finally,this paper analyses the stability of the system and proposes concept of discrimination condition of stability.Finally, this paper designs the model of the Kalman filter and simulation experiments.Firstly this paper designs error model of MEMS gyro random drift with the Kalman filter andanalyzes its stability of filter system. The results of analysis show that the filter is uniformlyasymptotically stable; secondly this paper designs kalman filter based on a combination ofMEMS gyroscope and magnetometer the attitude determination system; finally this paper hasthe experimen of parameters on the simulation and gives experimental results.