Inertial Sensors Error Compensation Aided By Actuation System

MEMS-IMU has become the preferred choice for inertial sensors in the field of low to medium precision strapdown inertial navigation due to its advantages such as low cost and small size.However,MEMS sensors have significant errors,making it difficult to ensure the accuracy of the strapdown inertial navigation system during the navigation process without compensation.Gyro bias is the largest proportion of MEMS sensor errors and also one of the core indicators for measuring MEMS-IMU,Therefore,estimating and compensating for gyro bias is an important means to improve the performance of MEMS strapdown inertial navigation systems.This error is generally estimated and compensated for in initial alignment and integrated navigation using external information such as satellite signals.However,when external information input is rejected or there is no external information input,it will revert to the situation of relying solely on the strapdown inertial navigation system for navigation,making it difficult to maintain navigation accuracy.This thesis proposes a gyroscope bias estimation compensation method assisted by an actuator system that does not require external information input to address this technical issue.The main tasks and features are as follows:Firstly,an analysis is conducted on the working principle and error composition of the strapdown inertial navigation system and inertial sensors.Secondly,a gyro bias estimation and compensation method is designed,based on the concept of navigation and control integrative system.The angle values of the actuator collected by the actuation system in the attitude control system are fed back to the navigation system as measurement signals,and a state space model is constructed.The Kalman filtering algorithm is used to obtain the optimal estimation of the gyro bias,and simulation results demonstrate that the proposed estimation method can effectively and accurately estimate the bias stability of the gyro after a single power-up without initial alignment.After compensation,the navigation accuracy of the strapdown inertial navigation system can be improved.Subsequently,a method called "Estimating while filtering "was proposed to address the problem of gyro zero bias variation over time after a single power-on.This method uses statistical analysis to regress and linearly fit the bias instability.A suitable sample size was selected through experimental statistics,and simulation was used to verify that the designed regression analysis method can effectively fit the bias instability of the gyro.Additionally,to account for the special case where there are exceptional data acquisition problems with the actuator during the actual process,the χ2 test method was used to detect,diagnose,and repair actuator failures,thereby improving the stability and accuracy of the gyro bias estimation results.The experimental results show that the designed method for estimating and compensating for gyroscope bias can effectively and accurately estimate the gyroscope zero bias and its unstable changes.The proposed method can be widely applied as a low-cost,easy-to-use navigation method for MEMS and other low-cost navigation platforms,as well as an emergency navigation method when GNSS signals are blocked and the combination navigation fails.