Research On On-line Calibration And Compensation Technology Of MEMS Inertial Navigation System

MEMS inertial navigation systems have a wide range of applications in military,intelligent driving,robot control and other fields,and occupy a prominent position in the navigation field.The advantages of MEMS inertial sensors are low cost,small size,low power consumption,good measurement autonomy and stability.However,due to its low accuracy,the accumulation error of the inertial navigation system will increase with time.Therefore,how to improve the navigation performance of the MEMS inertial navigation system needs to be solved.In this paper,the online calibration and compensation technology of MEMS inertial navigation system is studied.With a view to achieving highprecision,autonomous and reliable navigation of the MEMS vehicle navigation system,It provides theoretical basis and method reference for high-precision navigation of MEMS inertial navigation system.First,this paper studies the online calibration technology of MEMS inertial navigation system,the equation of state is established according to the online calibration method of zero offset and scale factor in MEMS inertial navigation system,observation equations were established by GPS external measurement system,five different paths were designed,observable analysis of the error variables in the designed path,and using traditional Kalman filtering to calibrate the value of zero offset and scale factor.Secondly,in view of the fact that the external environment interference is relatively large when collecting observation information,this paper proposes a filtering algorithm that combines fuzzy logic and conventional Kalman filtering,In this algorithm,the actual and theoretical values of the residuals are defined,use the ratio of their traces as a fuzzy input,the output is a correction factor to correct external observation noise.Through simulation,it is verified that the algorithm can effectively improve the navigation accuracy of the system,and compared with the calibration results of conventional Kalman filter algorithm,the algorithm can better suppress the divergence of filtering,to a certain extent,it eliminates the influence of external harsh environment on the calibration process.Finally,on the basis of the above two filtering algorithms,a semi-physical simulation platform was designed to evaluate the calibration and compensation effects through the vehicle-mounted MEMS navigation system and high-precision GPS system.Use the collected measured data for inertial navigation solution,the time synchronization method is adopted to solve the problem of time synchronization between two sensor systems.Then compensate the calibration result to the input,successfully improved the accuracy of the system.When the environmental impact is large,the simulation result of fuzzy adaptive Kalman filter is about 20 times lower than that of Kalman filter.Therefor,the algorithm proposed in this paper improves the accuracy of the system more effectively.