Estimation And Correction Of Geomagnetic Iterference Based On Multisensor Fusion

In recent years,due to the rich geomagnetic information of geomagnetic field,carrier navigation and target detection based on geomagnetic information have become a hot research topic at home and abroad.In practical applications,three-axis magnetometers are commonly used to acquire geomagnetic information.Due to the characteristics of the threeaxis magnetometer,errors and external disturbances have a great influence on it.The research on the correction technology of the three-axis magnetometer is of great significance for improving the accuracy of navigation and target detection.With the rapid development of sensor integration technology,multi-sensor fusion can solve the problem of low accuracy and poor reliability of single sensor in the measurement process.How to use the multi-sensor fusion technology to estimate and correct the three-axis magnetometer interference error becomes an important technical difficulty in the field of navigation and target detection research.In this thesis,the following aspects are studied in response to the above problems.(1)The source and properties of geomagnetic interference error are analyzed,and they are classified into additive interference error and multiplicative interference error according to its properties.Then a three-axis magnetometer interference error model is established.The ellipsoidal fitting method is used to correct the zero offset error,hard magnetic interference,non-orthogonal error and coefficient error of a single triaxial magnetometer.However,the ellipsoidal fitting method needs the accurate value of the geomagnetic intensity,and needs to collect a large number of magnetic vector data under different attitude in the early stage.(2)In the background of three-axis magnetometers integrated with three-axis gyroscopes and three-axis accelerometers,this thesis proposes two geomagnetic interference estimation and correction techniques based on heterogeneous sensors.The vector point product invariant method is proposed based on the three-axis accelerometer.Using the point product invariance principle of gravity vector and geomagnetic vector,the gravity and geomagnetic data are collected in the zero speed interval to estimate the interference and error of the three-axis magnetometer.Then the three-axis magnetometer is calibrated.This thesis also proposes a geomagnetic interference estimation and correction method based on gyroscope.The method uses a three-axis gyroscope to convert the three-axis error of the magnetometer into two-axis error,and estimates and corrects the interference error of the magnetometer by using the attitude angle from the three-axis gyroscope.(3)A multi-array sensor module is designed and adopted,which uses two similar sensor fusion methods for interference error estimation and correction.The first method is the data fusion of geomagnetic information.The method adopts different weighting coefficients to reduce the interference error.The second method is an online correction method based on multi-array sensor.In the multi-array sensor module,the relatively fixed characteristics of the three-axis magnetometer is utilized to instantly update the sensor's interference error estimate by the point product invariant method.