Carrier initial course estimation method based on transformation estimator modeling scheme

Abstract

The invention provides a carrier initial course estimation method based on a transformation estimator modeling scheme. The method comprises the steps of (1), data acquisition and synchronization; (2),dynamic identification of a carrier; (3), initialization of a horizontal angle; (4), resolving of a course estimation algorithm; and (5), course estimation convergence evaluation. According to the method, dynamic course initialization of multi-source combined navigation positioning equipment can be realized; the effect that course initialization can be completed for any position configuration ofthe multi-source combined navigation positioning equipment is achieved; the installation requirements of the multi-source combined navigation positioning equipment are reduced; and the scene adaptability and the user experience of the multi-source combined navigation equipment are improved.

Description

technical field [0001] The invention relates to a method for dynamically estimating the initial heading of a carrier equipped with an absolute measurement sensor and an inertial measurement sensor, and is suitable for the field of navigation and positioning by using an absolute measurement sensor, an inertial measurement sensor and a multi-source sensor combined navigation algorithm. Background technique [0002] The multi-source sensor integrated navigation and positioning method with inertial navigation as the main navigation and positioning system needs to provide initial navigation and positioning information, including initial position, initial velocity and initial attitude, where the initial position and initial velocity can generally be given by absolute measurement sensors, mainly It is to confirm the initial attitude. The initial attitude includes pitch angle, roll angle and heading angle. It is the rotation angle of the inertial sensor coordinate axis relative to th...

AI Technical Summary

Problems solved by technology

[0003] The static initialization method uses the gravity vector and the earth's rotation angular velocity as a reference, and measures the gravity vector and the earth's rotation angular velocity through an inertial sensor to calculate the initial attitude. The calculation methods mainly include the two-vector attitude determination algorithm and the kalman filter algorithm, but the static initialization method must be implemented The prerequisite is to be able to distinguish the angular velocity component of the earth's rotation from the observations of the inertial sensor, which puts forward high requirements on the performance of the inertial sensor. At present, most of the optical gyro inertial sensors and rotor gyro inertial sensors that can meet the static initialization requirements , the price is very expensive, the MEMS inertial sensor widely used in the civilian field is cheap, but the performance cannot meet the requirements of static initialization;

[0004] The dynamic initialization method is to complete the initialization of the heading angle in the initial attitude with the heading observation information provided by the external observation, or realize the initialization of the heading angle through a nonlinear algorithm. The initialization of the pitch angle and roll angle can be given by the inertial sensor, and the external information can be It is the displacement vector under the navigation system, the speed vector under the navigation system, the heading of the dual antenna, the heading of the magnetometer, etc., but the initialization through these information all puts forward requirements for the installation of the inertial sensor, that is, the coordinate axis of the inertial sensor needs to be consistent with The reference axis system corresponding to the external information is parallel, otherwise it will cause installation error. The size of the installation error directly affects the accuracy of the external heading information. If the installation error is large (>5°), it will cause the multi-source integrated navigation and positioning algorithm model to be inconsistent Satisfying the assumption of linearization leads to filter divergence. The nonlinear filtering algorithm can avoid the nonlinear problem of the model caused by large installation errors, but the calculation amount of the nonlinear filtering algorithm is very large, which is a great challenge to the processing power of the CPU. , and the nonlinear filtering cannot completely solve the nonlinear problem caused by the installation error. When the error of the heading information given by the external information is very large, the nonlinear filtering may still fail

Images