CN116519013A - High-precision inertial navigation initial alignment method based on virtual IMU prediction - Google Patents

Abstract

The invention provides a high-precision inertial navigation initial alignment method based on virtual IMU prediction, which comprises the following steps: navigation alignment of a plurality of positions is carried out based on a biaxial rotation inertial navigation system, and alignment reference data, IMU data and a code disc rotation angle are obtained; acquiring the angular velocity and the acceleration of the virtual IMU gyroscope according to the alignment reference data; preprocessing IMU data to obtain a gyro angular rate increment and a accelerometer acceleration increment; setting model training input and output; performing model training by adopting a deep learning LSTM model according to model training input and model training output to obtain a virtual IMU prediction training model; based on the virtual IMU prediction training model, acquiring real-time predicted virtual IMU gyro angular velocity and acceleration according to real-time IMU data calculated gyro angular velocity increment, accelerometer acceleration increment and real-time code wheel rotation angle; and performing navigation alignment according to the real-time predicted angular velocity and acceleration of the virtual IMU gyroscope. The invention can solve the technical problem of insufficient inertial navigation initial alignment precision in the prior art.

Description

A high-precision inertial navigation initial alignment method based on virtual IMU prediction

technical field

The invention relates to the technical field of inertial navigation systems, in particular to a high-precision inertial navigation initial alignment method based on virtual IMU prediction.

Background technique

Due to its good concealment and strong autonomy, inertial navigation technology has been continuously applied in various weapons and equipment on land, sea and air. Alignment research on platform inertial navigation systems has been quite mature. Usually Kalman filtering is used to solve its initial alignment problem. However, due to the poor observability of the inertial navigation system, the convergence speed and estimation accuracy of the linear filter state estimation are affected, which in turn affects the alignment accuracy and rapidity, and there is a problem of conflicting alignment accuracy and rapidity , therefore, it is still a very valuable problem to study how to improve the speed and accuracy of the initial alignment of the inertial navigation system.

The dual-axis rotary inertial navigation system can modulate the zero error of the gyroscope and the meter through the dual-axis rotation, and improve the alignment navigation accuracy. However, after the IMU rotates, the coordinate conversion calculation of the code disc is required to obtain the attitude result of the final navigation system, so During the rotation of the IMU, the error of the code disc will also lead to an error in the attitude transformation matrix, which will eventually be introduced into the navigation system to align the attitude angle. Therefore, in order to further improve the alignment accuracy of the high-precision inertial navigation system, it is necessary to propose an improved method based on the original rotation modulation alignment to meet the needs of long-range strategic weapons and equipment at this stage.

Contents of the invention

The present invention aims to solve at least one of the technical problems existing in the prior art.

The present invention provides a high-precision inertial navigation initial alignment method based on virtual IMU prediction. The high-precision inertial navigation initial alignment method based on virtual IMU prediction includes: performing navigation alignment of multiple positions based on a dual-axis rotary inertial navigation system. Acquire the alignment reference data, IMU data and code wheel rotation angle; obtain the virtual IMU gyro angular velocity and acceleration according to the alignment reference data; preprocess the IMU data to obtain the gyro angular rate increment and the acceleration increment; according to the virtual IMU gyro angular velocity and acceleration, gyro angular rate increment and meter acceleration increment, and code disc rotation angle are used to set model training input and model training output; according to model training input and model training output, deep learning LSTM model is used for model training to obtain Virtual IMU prediction training model; based on the virtual IMU prediction training model, the real-time predicted virtual IMU gyro angular velocity and acceleration can be obtained according to the gyro angular rate increment and acceleration increment calculated by the real-time IMU data, and the real-time code wheel rotation angle; Real-time predicted virtual IMU gyro angular velocity and acceleration for navigation alignment.

Further, according to Get the virtual IMU gyro angular velocity, where, /> is the virtual IMU gyro angular velocity, /> is the projection of the angle change of the carrier coordinate system b relative to the navigation coordinate system n under the b system, Y is the IMU sampling period, γ, θ, ψ are reference roll angle, pitch angle, heading angle respectively; /> is the projection of the earth's rotation angular velocity in the navigation system n; L is the latitude, ω _ie is the angular rate of the earth's rotation; /> is the projection along the east direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, /> is the projection along the north direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, R _n is the curvature radius of the Maoyou circle, H is the navigation height, and R _m is the curvature radius of the meridian circle.

Further, according to Get virtual IMU acceleration, where, /> Acceleration for the virtual IMU, /> is the projection under the n-system of the velocity variation of the carrier system b relative to the inertial system i-system, /> is the acceleration due to gravity, /> is the velocity of the navigation coordinate system n relative to the earth coordinate system e, />

Further, according to Get the gyro angular rate increment, where, /> is the gyro measurement angular rate increment in the ΔT time period, ΔT is the data preprocessing interval time period, /> is the angular velocity measured by the gyroscope, and T is the IMU sampling period.

Further, according to Get the acceleration increment of the table, where, /> is the acceleration increment of the meter in the ΔT time period, ΔT is the data preprocessing interval time period, /> In order to add a meter to measure the specific force, T is the IMU sampling period.

Further, set the model training input as Among them, Z( ) means that z-score is used for data normalization processing, which satisfies /> μ represents the mean value of the original data x, and σ represents the standard deviation of the original data x;/> Respectively, the angular rate increments measured by the gyroscope /> Components along the IMU x-axis, y-axis, z-axis, /> Respectively add table acceleration increment /> Components along the IMU x-axis, y-axis, and z-axis, α is the inner ring angle of the code wheel, and β is the outer ring angle of the code wheel.

Further, set the model training output as where, /> and /> are the virtual IMU gyro angular velocity and acceleration values, respectively.

Further, the loss function of the virtual IMU prediction training model is Among them, Y _OUT is the model output, Y _pred is the predicted value during the model training process, and a is the number of all data items for training.

Further, after normalizing the gyro angular rate increment and acceleration increment obtained by real-time calculation, as well as the real-time code disc rotation angle, input the virtual IMU prediction training model for real-time prediction of virtual IMU gyro angular velocity and acceleration output.

Applying the technical solution of the present invention, a high-precision inertial navigation initial alignment method based on virtual IMU prediction is provided. The high-precision inertial navigation initial alignment method based on virtual IMU prediction obtains the virtual IMU gyro angular velocity and Acceleration, combined with IMU data and code wheel rotation angle, using deep learning LSTM model (Long short-termmemory, long-short-term memory network model) for model training to obtain a virtual IMU prediction training model, and calculate the gyro angular rate increment based on real-time IMU data Acceleration increment and real-time code disc rotation angle to obtain real-time predicted virtual IMU gyro angular velocity and acceleration, and perform navigation alignment. The inertial navigation initial alignment method of the present invention not only ensures the effect of the rotation modulation to eliminate the zero error, but also can estimate and compensate the error of the code disc, so as to achieve the purpose of fast and high-precision alignment. Compared with the prior art, the technical solution of the present invention can solve the technical problem of insufficient initial alignment accuracy of the inertial navigation system in the prior art.

Description of drawings

The accompanying drawings are included to provide further understanding of the embodiments of the invention, and constitute a part of the specification, are used to illustrate the embodiments of the invention, and together with the description, explain the principle of the invention. Apparently, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings according to these drawings without creative efforts.

Fig. 1 shows a schematic flow chart of a high-precision inertial navigation initial alignment method based on virtual IMU prediction provided according to a specific embodiment of the present invention.

Detailed ways

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. The following description of at least one exemplary embodiment is merely illustrative in nature and in no way taken as limiting the invention, its application or uses. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and/or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and/or combinations thereof.

The relative arrangements of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise. At the same time, it should be understood that, for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. Techniques, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such techniques, methods and devices should be considered part of the description. In all examples shown and discussed herein, any specific values should be construed as illustrative only, and not as limiting. Therefore, other examples of the exemplary embodiment may have different values. It should be noted that like numerals and letters denote like items in the following figures, therefore, once an item is defined in one figure, it does not require further discussion in subsequent figures.

As shown in Figure 1, according to a specific embodiment of the present invention, a high-precision inertial navigation initial alignment method based on virtual IMU prediction is provided, and the high-precision inertial navigation initial alignment method based on virtual IMU prediction includes:

Based on the dual-axis rotary inertial navigation system, the navigation alignment of multiple positions is performed, and the alignment reference data, IMU data and code wheel rotation angle are obtained; the virtual IMU gyro angular velocity and acceleration are obtained according to the alignment reference data;

Preprocess the IMU data to obtain the gyro angular rate increment and the acceleration increment of the meter;

Set the model training input and model training output according to the virtual IMU gyro angular velocity and acceleration, gyro angular rate increment and meter acceleration increment, and code disc rotation angle;

According to the model training input and model training output, the deep learning LSTM model is used for model training to obtain the virtual IMU prediction training model;

Based on the virtual IMU prediction training model, the real-time predicted angular velocity and acceleration of the virtual IMU gyro can be obtained according to the gyro angular rate increment and the acceleration increment calculated by the real-time IMU data, and the real-time code wheel rotation angle; according to the real-time predicted virtual IMU gyro Angular velocity and acceleration for navigation alignment.

Applying this configuration method, a high-precision inertial navigation initial alignment method based on virtual IMU prediction is provided. The high-precision inertial navigation initial alignment method based on virtual IMU prediction obtains virtual IMU gyro angular velocity and acceleration based on alignment reference data , combined with the IMU data and the rotation angle of the code disc, the deep learning LSTM model is used for model training to obtain the virtual IMU prediction training model, and the gyro angular rate increment and acceleration increment calculated according to the real-time IMU data, as well as the real-time code disc rotation angle Get real-time predicted virtual IMU gyro angular velocity and acceleration, and perform navigation alignment. The inertial navigation initial alignment method of the present invention not only ensures the effect of the rotation modulation to eliminate the zero error, but also can estimate and compensate the error of the code disc, so as to achieve the purpose of fast and high-precision alignment.

Further, in the present invention, in order to realize the initial alignment of high-precision inertial navigation based on virtual IMU prediction, firstly, the navigation alignment of multiple positions is performed based on the dual-axis rotary inertial navigation system, and the alignment reference data, IMU data and codes are obtained. Disk rotation angle; obtain virtual IMU gyro angular velocity and acceleration based on alignment reference data.

As a specific embodiment of the present invention, the virtual IMU gyro angular velocity and acceleration values can be reversely calculated based on the reference attitude results, specifically, according to Get the virtual IMU gyro angular velocity, where, /> is the virtual IMU gyro angular velocity, /> is the projection of the angle change of the carrier coordinate system b relative to the navigation coordinate system n in the b system, T is the IMU sampling period, γ, θ, ψ are reference roll angle, pitch angle, heading angle respectively; /> is the projection of the earth's rotation angular velocity in the navigation system n; L is the latitude, ω _ie is the angular rate of the earth's rotation; /> is the projection along the east direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, /> is the projection along the north direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, R _n is the curvature radius of the Maoyou circle, H is the navigation height, and R _m is the curvature radius of the meridian circle.

according to Get virtual IMU acceleration, where, /> Acceleration for the virtual IMU, /> is the projection under the n-system of the velocity variation of the carrier system b relative to the inertial system i-system, /> is the acceleration due to gravity, /> is the velocity of the navigation coordinate system n relative to the earth coordinate system e, />

Further, in the present invention, after obtaining the virtual IMU gyro angular velocity and acceleration, the IMU data is preprocessed to obtain the gyro angular rate increment and the surface acceleration increment.

As a specific embodiment of the present invention, according to and Obtain the gyro angular rate increment and the accelerometer increment, where, /> Measure the angular rate increment for the gyro during the ΔT time period, /> is the acceleration increment of the meter in the ΔT time period, ΔT is the data preprocessing interval time period, /> Measure angular velocity for gyro, /> In order to add a meter to measure the specific force, T is the IMU sampling period. When t=n×ΔT, the interval period needs to restart the data, and perform IMU incremental initialization. At this time />

Further, in the present invention, after obtaining the gyro angular rate increment and the surface acceleration increment, the model is set according to the virtual IMU gyro angular velocity and acceleration, the gyro angular rate increment and the surface acceleration increment, and the code disc rotation angle Training input and model training output.

As a specific embodiment of the present invention, the gyroscope angular rate increment, the surface acceleration increment, and the code disc rotation angle are respectively normalized, and the normalized data is used as model training input, and the virtual The IMU gyro angular velocity and acceleration are used as model outputs.

Specifically, the model training input can be set as Among them, X _IN-z is the model training input, and Z( ) means that z-score is used for data normalization processing, which satisfies /> μ represents the mean value of the original data x, and σ represents the standard deviation of the original data x;/> Respectively, the angular rate increments measured by the gyroscope /> Components along the IMU x-axis, y-axis, z-axis, /> Respectively add table acceleration increment /> Components along the IMU x-axis, y-axis, and z-axis, α is the inner ring angle of the code wheel, and β is the outer ring angle of the code wheel.

Set the model training output to where, /> and /> are the virtual IMU gyro angular velocity and acceleration values, respectively.

Further, in the present invention, after setting the model training input and output, the deep learning LSTM model is used for model training according to the model training input and output, and the virtual IMU prediction training model is obtained.

As a specific embodiment of the present invention, the look-back value of the virtual IMU prediction training model is set to 1, and the loss function is Among them, Y _pred is the predicted value during the model training process, and a is the number of all data items for training.

This embodiment uses a square loss function. In addition, set the number of training times for model training as epoch. After the number of training times is satisfied, the virtual IMU prediction training model can be obtained as F(p1, p2, p3...ps), where p1, p2 , p3...ps are various parameters obtained from training, and s is an integer.

Further, in the present invention, after obtaining the virtual IMU prediction training model, based on the virtual IMU prediction training model, the gyroscope angular rate increment and the acceleration increment and the real-time code disc rotation angle obtained according to the real-time IMU data solution Real-time predicted virtual IMU gyro angular velocity and acceleration; navigation alignment based on real-time predicted virtual IMU gyro angular velocity and acceleration.

As a specific embodiment of the present invention, the trained virtual IMU prediction training model F(p1, p2, p3...ps) is imported into the alignment application software for application. During the alignment process, the gyro angular rate increment and meter acceleration increment are calculated according to the real-time IMU data, and the gyro angular rate increment and meter acceleration increment obtained by real-time calculation, as well as the real-time code wheel rotation angle Normalized and used as model input. Set the prediction period to ΔT. When the time is greater than or equal to l×ΔT, the real-time prediction will be performed according to the virtual IMU prediction training model, and the real-time prediction output will be where, /> and /> are the virtual IMU gyro angular velocity and acceleration output by real-time prediction, respectively.

Furthermore, on the basis of rough alignment, navigation calculation and Kalman filter alignment are performed according to the real-time predicted virtual IMU gyro angular velocity and acceleration, and the obtained attitude angle can be used as the final alignment result.

The high-precision inertial navigation initial alignment method based on virtual IMU prediction of the present invention can be applied to the field of initial alignment of platform-type dual-axis rotary inertial navigation systems. According to the alignment reference, the present invention reversely calculates the virtual IMU data during the rotation process, and then Using the deep learning method based on LSTM, based on the IMU data during the rotation process and the rotation angle of the inner and outer ring code discs, the virtual IMU prediction model training is carried out. During the alignment process, the actual IMU and the inner and outer ring angle information of the code disc are used to predict the error-free Virtual IMU information to eliminate various errors in IMU measurement. Real-time alignment based on virtual IMU information predicted by the present invention not only ensures the effect of rotation modulation to eliminate zero error, but also estimates and compensates code disc error, so as to achieve the purpose of fast and high-precision alignment.

In order to have a further understanding of the present invention, the method for initial alignment of high-precision inertial navigation based on virtual IMU prediction of the present invention will be described in detail below in conjunction with FIG. 1 .

As shown in FIG. 1 , according to a specific embodiment of the present invention, a high-precision inertial navigation initial alignment method based on virtual IMU prediction is provided, including the following steps.

Step 1: Carry out navigation alignment at multiple positions based on the dual-axis rotary inertial navigation system, and obtain alignment reference data, IMU data, and code wheel rotation angle.

according to Obtain the virtual IMU gyro angular velocity; according to Get virtual IMU acceleration.

Step two, according to and Get the gyro angular rate increment and accelerometer increment.

Step 3, set the model training input as Set the model training output to />

Step 4: According to the model training input and output, the deep learning LSTM model is used for model training. The virtual IMU predicts the look-back value of the training model as l, and the loss function is

Step five, import the trained virtual IMU prediction training model F(p1, p2, p3...ps) into the alignment application software for application. During the alignment process, the gyro angular rate increment and meter acceleration increment are calculated according to the real-time IMU data, and the gyro angular rate increment and meter acceleration increment obtained by real-time calculation, as well as the real-time code wheel rotation angle Normalized and used as model input. Set the prediction period to ΔT. When the time is greater than or equal to l×ΔT, the real-time prediction will be performed according to the virtual IMU prediction training model, and the real-time prediction output will be

Step 6: On the basis of rough alignment, navigation calculation and Kalman filter alignment are performed according to the real-time predicted virtual IMU gyro angular velocity and acceleration, and the obtained attitude angle can be used as the final alignment result.

In summary, the present invention provides a high-precision inertial navigation initial alignment method based on virtual IMU prediction. The high-precision inertial navigation initial alignment method based on virtual IMU prediction obtains virtual IMU gyro angular velocity and Acceleration, combined with IMU data and code wheel rotation angle, adopts deep learning LSTM model for model training to obtain virtual IMU prediction training model, and calculates gyro angular rate increment and meter acceleration increment based on real-time IMU data, as well as real-time code wheel rotation Angle Obtain real-time predicted virtual IMU gyro angular velocity and acceleration, and perform navigation alignment. The inertial navigation initial alignment method of the present invention not only ensures the effect of the rotation modulation to eliminate the zero error, but also can estimate and compensate the error of the code disc, so as to achieve the purpose of fast and high-precision alignment.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (9)

1. A high-precision inertial navigation initial alignment method based on virtual IMU prediction, characterized in that, the high-precision inertial navigation initial alignment method based on virtual IMU prediction comprises: Based on the dual-axis rotary inertial navigation system, the navigation alignment of multiple positions is carried out, and the alignment reference data, IMU data and code wheel rotation angle are obtained; the virtual IMU gyro angular velocity and acceleration are obtained according to the alignment reference data; Preprocessing the IMU data to obtain gyro angular rate increments and surface acceleration increments; Set model training input and model training output according to the virtual IMU gyro angular velocity and acceleration, the gyro angular rate increment and the acceleration increment, and the code wheel rotation angle; Adopt deep learning LSTM model to carry out model training according to described model training input and model training output, obtain virtual IMU prediction training model; Based on the virtual IMU prediction training model, according to the real-time IMU data solution gyro angular rate increment and meter acceleration increment and the real-time code disc rotation angle to obtain real-time predicted virtual IMU gyro angular velocity and acceleration; according to the real-time prediction The virtual IMU gyro angular velocity and acceleration are used for navigation alignment. 2. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, according to Get the virtual IMU gyro angular velocity, where, /> is the virtual IMU gyro angular velocity, /> is the projection of the angle change of the carrier coordinate system b relative to the navigation coordinate system n in the b system, T is the IMU sampling period, /> γ, θ, ψ are reference roll angle, pitch angle, heading angle respectively; is the projection of the earth's rotation angular velocity in the navigation system n; L is the latitude, ω _ie is the angular rate of the earth's rotation; /> is the projection along the east direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, R _n is the radius of curvature of the Maoyou circle, /> is the projection along the north direction of the velocity of the navigation coordinate system n relative to the earth coordinate system e, H is the navigation height, and R _m is the radius of curvature of the meridian. 3. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1 or 2, is characterized in that, according to Get virtual IMU acceleration, where, /> Acceleration for the virtual IMU, /> is the projection under the n-system of the velocity variation of the carrier system b relative to the inertial system i-system, /> is the acceleration of gravity, /> is the velocity of the navigation coordinate system n relative to the earth coordinate system e, /> 4. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, according to Get the gyro angular rate increment, where, /> is the gyro measurement angular rate increment in the ΔT time period, ΔT is the data preprocessing interval time period, /> is the angular velocity measured by the gyroscope, and T is the IMU sampling period. 5. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, according to Get the acceleration increment of the table, where, /> is the acceleration increment of the meter in the ΔT time period, ΔT is the data preprocessing interval time period, /> In order to add a meter to measure the specific force, T is the IMU sampling period. 6. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, setting model training input is Among them, Z( ) means that z-score is used for data normalization processing, which satisfies /> μ represents the mean value of the original data x, and σ represents the standard deviation of the original data x;/> are the angular rate increments measured by the gyro Components along the IMU x-axis, y-axis, z-axis, /> Respectively add table acceleration increment /> Components along the IMU x-axis, y-axis, and z-axis, α is the inner ring angle of the code wheel, and β is the outer ring angle of the code wheel. 7. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, setting model training output is where, /> and /> are the virtual IMU gyro angular velocity and acceleration values, respectively. 8. the high precision inertial navigation initial alignment method based on virtual IMU prediction according to claim 1, is characterized in that, the loss function of virtual IMU prediction training model is Among them, Y _OUT is the model output, Y _pred is the predicted value during the model training process, and a is the number of all data items for training. 9. The high-precision inertial navigation initial alignment method based on virtual IMU prediction according to any one of claims 1 to 8, characterized in that, the gyro angular rate increment and the surface acceleration increment obtained by real-time solution , and the real-time code wheel rotation angle are normalized, and then input into the virtual IMU prediction training model for real-time prediction output of the virtual IMU gyro angular velocity and acceleration.