Key Technology Research On Lightweight Position&Attitude Measurement System Based MEMS Sensor

The Position and Attitude system(PAS)is a device that acquires precise position and attitude information of the target.It has been widely applied in Precision Guide Weapon and other military areas,Unmanned Aerial Vehicle(UAV)s,Robots,Wearable smart devices and other Civilian application domains.Widely utilized of PAS have placed greater demand on PAS in lightweight,miniaturized and precision.The concept of lightweight has two level of meanings:the lighter of the system heft and the system core algorithm;the latter means that pruning the system structure and optimizing the system algorithm performance when the system meets the usage requirem'ents.Laser Gyro,Fiber Optic Gyroscope and Maglev mechanical Gyroscope are high-precision inertial devices,which price is up to tens of thousands,beyond that,the large size and the high weight,the complex solution algorithm of them and need higher platform computing ability.All of those characters make it be restricted in utilized;The emergence of the MEMS inertial navigation devices have made the lightweight and miniaturized of PAS come ture and become the R&D direction of position and attitude measurement systems,gradually.However,MEMS inertial devices have the character of low sensitivity and poor consistency,which seriously restricts their application in higher precision applications.With the aim to improve the whole precision and stability of the PAS,the paper mainly concerned on the keys technologies of the lightweight PAS with the core of MEMS sensor,focusing on the causes compensation and calibration methods of error of MEMS inertial devices,and fusion methods.The main efforts done in the thesis are as follows:1.In view of the complex and diverse kinds of inertial device errors,the MEMS-IMU error compensation theory of strapdown inertial navigation system is studied.The random error source of MEMS gyroscope is analyzed by Allan variance.The analysis shows that the main impact factor of MEMS gyro is bias instability and angular random walk,and then the gyroscope error model was established.After analyzing the design of the multi-position calibration accelerometer of the swivel table,the calibration scheme of plumb line and level aiding accelerometer was proposed to simplify the process of accelerometer calibration,the calibrated data shows that the scheme is feasible and can improve the precision.Finally,the least-squares ellipsoid fitting method are used to calibrate the magnetometer.for MEMS gyro,An algorithm for on-line calibration of magnetometer-assisted gyro is proposed in this paper.The Kalman filter system equation and observation equation based on the conversion relationship between the magnetic coordinate system and the body coordinate system are established.2.Studying the error transfer principle and attitude update algorithm of SINS system,comparing the pros and cons of the existing attitude update algorithm combined with the characteristics of MEMS inertial device,the angular increment of MEMS inertial device output was selected to obtain the attitude quaternion,thus reducing the computing load of algorithm.The accelerometer temperature drift compensation is completed by multivariate finite fitting method,the random drift of it reduce more than 30%and less than 4mg.The pure SINS position calculation based on MEMS inertial device is verified,the results shows that the SINS height channel does not converge.3.Finally,the system equation and observation equation of the Kalman filter based on the navigation parameters of the strapdown inertial navigation system are established by GNSS/SINS loose combination.The position and attitude calculation based on MEMS inertial device is realized on STM32 series platform.The results show that the system can effectively suppress position error and attitude error.The error of static horizontal positioning by CEP mode is 4 meters in 50%confidence probability,and the attitude measurement accuracy is better than 0.2°.In the condition of dynamic maneuvering process,the certainty of attitude is better than 1°.