CN108088464B - A Deflection Estimation Method for Closed-loop Correction of Installation Errors - Google Patents

Abstract

The invention belongs to the technical field of inertial measurement, and particularly relates to a deflection estimation method for correcting installation errors in a closed loop. The method comprises nine steps, wherein the first step is to define a coordinate system, the second step is to carry out navigation operation based on an inertial system, the third step is to establish a flexural error model, the fourth step is to establish a system error equation, the fifth step is to establish a measurement equation, the sixth step is to calculate a measurement value, the seventh step is to calculate by utilizing a Kalman filtering equation, the eighth step is to carry out closed-loop correction on an installation error angle and a flexural deformation angle, and the ninth step is to repeat the closed-loop correction on the flexural deformation angle. Compared with the traditional method, the method can greatly improve the measurement precision and the estimation speed, enhance the attitude information measurement precision of different parts of the carrier platform and improve the use efficiency of related equipment.

Description

A Deflection Estimation Method for Closed-loop Correction of Installation Errors

technical field

The invention belongs to the technical field of inertial measurement, and particularly relates to a deflection estimation method for closed-loop correction of installation errors.

Background technique

The dynamic deflection and deformation of the carrier platform will affect the use of the main reference information by the sub-inertial navigation equipment, and affect the measurement accuracy of the equipment installed on the carrier platform including ships, aircraft wings and ground vehicles. Therefore, the deflection of the carrier platform must be adjusted. Deformation characteristics are effectively measured and compensated. The deflection deformation is mainly due to the influence of external stress such as temperature change, external impact, airflow, self-load and the surrounding environment. The structure of the carrier platform itself will have different degrees of dynamic deformation. The deflection deformation can be divided into static deflection and dynamic deflection. .

Static flexural deformation is a deformation phenomenon caused by thermal expansion and cold contraction, structural stress release, etc. of the carrier platform during the movement process. It is mainly affected by factors such as load, temperature, and sunlight. The change period is long and relatively stable. The quasi-static flexural deformation can be considered to be constant during the transfer alignment time. British D.H.Titterton and others have pointed out that the load change and structural aging on the ship will lead to the deformation of the hull, which can produce an angle change of about 1° in one day under the action of sunlight.

Dynamic deflection changes continuously with the influence of external conditions such as movement and impact of the carrier, that is, dynamic deflection is a function of time. The static deformation can be approximately regarded as the fixed installation error between the main inertial navigation and the sub-inertial navigation, while the dynamic deflection deformation is a time-varying physical quantity superimposed on the basis of the static deflection. The reasons for formation vary depending on the type of carrier platform.

On the offshore platform, the dynamic flexural deformation of the ship refers to the phenomenon that the hull itself is not a rigid body, and is deformed and structurally vibrated by the action of external force, external moment, and wave impact. The high-frequency deflection deformation is mainly caused by the impact of the waves and the maneuvering shaking of the ship, the missile launcher and the deformation of the hull structure. According to the obtained deformation data when sailing under the sixth-level sea conditions, the deformation around the pitch axis at ±7m along the horizontal axis and ±1.5m along the longitudinal axis from the installation position of the inertial navigation system is 0.05°～0.08° , the deformation around the roll axis is 0.17°~0.2°; the deformation at the installation of the radar antenna base is 0.38°, and the deformation at the installation of the front and rear guns is 0.23°.

On the aircraft platform, flexural deformation refers to the phenomenon that the wing itself is not a rigid body, and is subjected to external forces, external moments, aerodynamic loads, and turbulence to produce deformation and structural vibrations. The deflection deformation of the carrier in flight is mainly divided into two categories, one is the static deflection deformation, and the other is the high frequency deflection deformation. Static flexural deformation is the low-frequency wing deformation phenomenon caused by the change of aircraft load due to aircraft maneuvering or weapon delivery; high-frequency flexural deformation is mainly caused by turbulent flow in the 5-10 Hz aircraft structure.

At present, common deflection measurement methods are divided into two categories, one is optical measurement, and the other is realized by high-precision inertial measurement. Among them, the high-precision inertial measurement mainly uses the Kalman filter to measure the deflection deformation by using the inertial measurement attitude difference distributed in different positions. In addition, the current method cannot realize on-line compensation of deflection error.

Therefore, this paper considers using the difference of attitude measurement information of different parts of the ship to establish the error equation of attitude difference, including inertial navigation gyro drift, fixed installation error and deflection deformation model, etc., and realizes the installation error and deflection deformation through Kalman filter technology. to achieve high-precision pose measurement of different parts.

SUMMARY OF THE INVENTION

Based on the above problems, the present invention proposes a deflection estimation method for closed-loop correction of installation error, which utilizes the difference in attitude measurement information of different parts of the ship to establish an error equation for attitude difference, including inertial navigation gyro drift, fixed installation error and deflection deformation Model, etc., through the Kalman filter technology to realize the estimation of installation error and deflection deformation and closed-loop correction, to achieve high-precision measurement of the required position posture.

In order to achieve this purpose, the technical scheme adopted by the present invention is:

A deflection estimation method for closed-loop correction of installation error, including nine steps, the first step is to define a coordinate system, the second step is to perform a navigation operation based on an inertial frame, the third step is to establish a deflection error model, and the fourth step is to establish a system Error equation, step 5 is to establish the measurement equation, step 6 is to calculate the measurement value, step 7 is to use the Kalman filter equation to calculate, step 8 is to correct the installation error angle and deflection angle closed-loop, and step 9 is to repeat the deflection In the closed-loop correction of the deformation angle, a pair of coordinate systems are defined in the steps. In the attitude matching calculation method, an auxiliary coordinate system needs to be introduced:

Inertial coordinate system i ₁ : obtained by inertial solidification of the sub-inertial navigation calculation carrier system b at time t ₀ ;

Inertial coordinate system i ₂ : obtained by inertial solidification of the main inertial navigation carrier system m at time t ₀ ;

Earth coordinate system e: the x-axis is along the direction of the earth's axis, the y-axis is along the intersection of the equatorial plane and the Greenwich meridian plane, and the z-axis forms a right-handed coordinate system with the x-axis and the y-axis in the equatorial plane;

The earth's inertial coordinate system i: the inertial solidification of the earth's coordinate system e at time t ₀ is obtained.

A deflection estimation method for closed-loop correction of installation error, the step 2 is based on the navigation operation of the inertial frame, and the quaternion is set as:

Q ₁ =[q ₀ q ₁ q ₂ q ₃ ] ^T , and the initial value Q ₁ (0)=[1 0 0 0] ^T .

The analytical formula for quaternion update is:

in:

I is a 4×4 order identity matrix;

为k到k+1时刻计算弹体系相对惯性系的角增量分量，即

Calculate the angular increment component of the missile system relative to the inertial system for the time k to k+1, namely

The angular increments sensed by the gyroscope are transformed into the angular increments after calculating the missile system, which is obtained by the following formula:

为陀螺x，y，z方向的角速度，

姿态矩阵的更新计算公式为：where T _n is the navigation solution period,

is the angular velocity of the gyro in the x, y, and z directions,

The update calculation formula of the attitude matrix is:

即子惯导相对于母惯导的动态形变角为：A deflection estimation method for closed-loop correction of installation error, the step 3 establishes a deflection error model, and sets the deflection deformation of the carrier platform

That is, the dynamic deformation angle of the child INS relative to the parent INS is:

where θ _x , θ _y , θ _z are the dynamic deformation angles of the child inertial navigation relative to the parent inertial navigation in the x, y, and z directions.

μ_x，μ_y，μ_z为在x，y，z方向的挠曲变形角速度，另外假设三轴的形变过程是相互独立的，则Assume that the deflection angular velocity variable is

μ _x , μ _y , μ _z are the angular velocities of deflection in the x, y, and z directions. In addition, assuming that the deformation processes of the three axes are independent of each other, then

Then the second-order Markov equation of motion for flexural deformation can be obtained as

In the formula, β _i =2.146/τ _i , τ _i represents the correlation time constant of the carrier platform i-direction deformation, (i=x, y, z); w _x , w _y and w _z represent the white noise sequence, the white noise The sequence is the driving information of the deflection angle, and its variances are Q _rx , Q _ry and Q _rz respectively, and Q _ri =4β ² _i σ ² _i ;

In summary, the state equation of the deflection deformation of the carrier platform is:

A deflection estimation method for closed-loop correction of installation error, the step 4 establishes a system error equation, and according to the error characteristics of the inertial navigation system, in the attitude matching Kalman filter model, 16 error state variables are selected as

X=[Φ Φ _a ξ ω ε t _A ] ^T

in:

Φ=[φ _x φ _y φ _z ] is the attitude error angle of the inertial system X, Y, and Z axes;

Φ _a =[φ _ax φ _ay φ _az ] is the installation error angle of the three axes X, Y and Z of the carrier system;

ξ=[θ _x θ _y θ _z ] ^T is the deflection deformation of the mounting error angle of the three axes of the carrier system X, Y, and Z;

ω=[μ _x μ _y μ _z ] ^T is the deflection and deformation angular velocity of the mounting error angle of the three axes of the carrier system X, Y, and Z;

ε=[ε _x ε _y ε _z ] gyro drift of X, Y, Z axes;

t _A : Reference attitude delay time.

Written in the form of the equation of state as follows

T表示转置；B为系统噪声向量。其中，

可通过基于惯性系的导航解算获得。in

T represents the transpose; B is the system noise vector. in,

It can be obtained by an inertial frame-based navigation solution.

A deflection estimation method for closed-loop correction of installation error, the step 5 establishes a measurement equation, and the measurement equation is as follows

为陀螺输出三轴角速率；A和B的计算需要利用矩阵

由下式获得：in,

Output triaxial angular rate for the gyroscope; A and B calculations require the use of matrices

Obtained by:

Among them, t ₀ represents the initial time of filtering, and the specific calculation method of each matrix is:

Among them, γ ₀ , ψ ₀ and θ ₀ are the roll angle, heading angle and pitch angle of the reference inertial navigation system at the initial moment, respectively,

Among them, L ₀ and λ ₀ are the longitude and latitude of the reference inertial navigation system at the initial moment, respectively.

为单位阵，即：

is the unit matrix, that is:

为地球系到惯性系的变化矩阵，计算方法为：

is the change matrix from the earth system to the inertial system, and the calculation method is:

where ω _ie is the angular rate of the Earth's rotation, and t is time.

Among them, λ and L are the longitude and latitude of the reference inertial navigation system, respectively.

Among them, γ, ψ and θ are the roll angle, heading angle and pitch angle of the reference inertial navigation system, respectively.

则make

but

A deflection estimation method for closed-loop correction of installation error, the step 6 calculates the measured value,

A deflection estimation method for closed-loop correction of installation error, wherein the step 7 utilizes Kalman filter equation to calculate,

After the above error model is established, the Kalman filter method is selected as the parameter identification method, and the formula is as follows:

State one-step prediction

State estimation

Filter Gain Matrix

One-step forecast error variance matrix

Estimation Error Variance Matrix

P _k =[IK _k H _k ]P _k,k-1

为一步状态预测值，

为状态估计矩阵，Φ_k,k-1为状态一步转移矩阵，H_k为量测矩阵，Z_k为量测量，K_k为滤波增益矩阵，R_k为观测噪声阵，P_k,k-1为一步预测误差方差阵，P_k为估计误差方差阵，Γ_k,k-1为系统噪声驱动阵，Q_k-1为系统噪声阵。in,

is the one-step state prediction value,

is the state estimation matrix, Φ _{k, k-1} is the state one-step transition matrix, H _k is the measurement matrix, Z _k is the quantity measurement, K _k is the filter gain matrix, R _k is the observation noise matrix, P _{k, k-1} is the one-step prediction error variance matrix, P _k is the estimated error variance matrix, Γ _k,k-1 is the system noise driving matrix, and Q _k-1 is the system noise matrix.

A deflection estimation method for closed-loop correction of installation error, the step 8 is for closed-loop correction of the installation error angle and deflection deformation angle, and the correction method is as follows:

为Let γ ^m =X _A (3), ψ ^m =X _A (4), θ ^m =X _A (5), then install the error angle matrix

for

为Let γ ^m =X _A (6), ψ ^m =X _A (7), θ ^m =X _A (8), then the deflection deformation matrix

for

计算新的

计算

以替换原

Depend on

calculate new

calculate

to replace the original

A deflection estimation method for closed-loop correction of installation error, the step 9 repeats the closed-loop correction of deflection deformation angle, because the above steps 1 to 8 correct the error angle and deflection deflection angle, but the error angle and deflection deflection angle are not Convergence, can not reach the required accuracy, so it is necessary to repeat the correction of the error angle and deflection angle, so repeat the above steps 1 to 8, until the error angle and deflection angle estimation results converge, the method ends.

The beneficial effects of the present invention are:

The present invention utilizes the attitude information of multiple sets of inertial navigation systems, and can realize the estimation of error information such as inertial navigation gyro drift, fixed installation error and deflection deformation model through Kalman filtering technology, and close-loop correction of fixed installation error and deflection deformation can be achieved. Compared with the traditional method, the measurement accuracy and estimation speed can be greatly improved, the measurement accuracy of attitude information of different parts of the carrier platform can be enhanced, and the use efficiency of related equipment can be improved.

Detailed ways

Specific embodiments of the present invention are as follows:

The method proposed by the present invention includes nine steps, the first step is to define the coordinate system, the second step is to perform a navigation operation based on the inertial system, the third step is to establish a deflection error model, the fourth step is to establish a system error equation, and the fifth step is to establish For the measurement equation, step 6 is to calculate the measurement value, step 7 is to use the Kalman filter equation for calculation, step 8 is to correct the installation error angle and deflection angle closed-loop, and step 9 is to repeat the closed-loop correction of the deflection angle.

Step 1. Define the coordinate system

In the attitude matching calculation method, an auxiliary coordinate system needs to be introduced:

(1) Inertial coordinate system i ₁ : obtained by inertial solidification of the sub-INS calculation carrier system b at time t ₀ ;

(2) Inertial coordinate system i ₂ : obtained by inertial solidification of the main inertial navigation carrier system m at time t ₀ ;

(3) Earth coordinate system e: the x-axis is along the direction of the earth's axis, the y-axis is along the intersection of the equatorial plane and the Greenwich meridian plane, and the z-axis forms a right-handed coordinate system with the x-axis and the y-axis in the equatorial plane;

(4) Earth inertial coordinate system i: obtained by inertial solidification of earth coordinate system e at time t ₀ .

Step 2. Navigation operation based on inertial frame

Set up a quaternion:

Q ₁ =[q ₀ q ₁ q ₂ q ₃ ] ^T , and the initial value Q ₁ (0)=[1 0 0 0] ^T .

The analytical formula for quaternion update is:

in:

I is a 4×4 order identity matrix;

为k到k+1时刻计算弹体系相对惯性系的角增量分量，即

Calculate the angular increment component of the missile system relative to the inertial system for the time k to k+1, namely

The angular increments sensed by the gyroscope are transformed into the angular increments after calculating the missile system, which is obtained by the following formula:

为陀螺x，y，z方向的角速度，

姿态矩阵的更新计算公式为：where T _n is the navigation solution period,

is the angular velocity of the gyro in the x, y, and z directions,

The update calculation formula of the attitude matrix is:

Step 3. Establish a deflection error model

的原因主要包括各类冲击等因素的作用，这与有白噪声激励的情况比较一致；另外，载体平台挠曲变形既具有惯性同时还具有恢复力矩，可以看作为典型的质量-弹簧系统。二阶Markov过程建立模型可以适应通常的挠曲变形并能够保证相当的精度，因此，将采用二阶马尔可夫过程对在载体平台的挠曲变形建立数学模型。The establishment of the flexural deformation model of the carrier platform involves complex elastic mechanics and other professional knowledge, and the flexural deformation cannot be simply ignored or the flexural deformation can be regarded as a constant value. Deformation of the carrier platform

The reason mainly includes the effect of various shocks and other factors, which is consistent with the case of white noise excitation; in addition, the deflection deformation of the carrier platform has both inertia and restoring moment, which can be regarded as a typical mass-spring system. The model established by the second-order Markov process can adapt to the usual flexural deformation and can ensure considerable accuracy. Therefore, the second-order Markov process will be used to establish a mathematical model for the deflection deformation on the carrier platform.

即子惯导相对于母惯导的动态形变角为：Set the deflection deformation of the carrier platform

That is, the dynamic deformation angle of the child INS relative to the parent INS is:

where θ _x , θ _y , θ _z are the dynamic deformation angles of the child inertial navigation relative to the parent inertial navigation in the x, y, and z directions.

μ_x，μ_y，μ_z为在x，y，z方向的挠曲变形角速度，另外假设三轴的形变过程是相互独立的，则Assume that the deflection angular velocity variable is

μ _x , μ _y , μ _z are the angular velocities of deflection in the x, y, and z directions. In addition, assuming that the deformation processes of the three axes are independent of each other, then

Then the second-order Markov equation of motion for flexural deformation can be obtained as

In the formula, β _i =2.146/τ _i , τ _i represents the correlation time constant of the carrier platform i-direction deformation, (i=x, y, z); w _x , w _y and w _z represent the white noise sequence, the white noise The sequence is the driving information of the deflection angle, the variances of which are respectively Q _rx , Q _ry and Q _rz , and Q _ri =4β ² _i σ ² _i .

In summary, the state equation of the deflection deformation of the carrier platform is:

Step 4. Establish a systematic error equation

According to the error characteristics of the inertial navigation system, in the attitude matching Kalman filter model, 16 error state variables are selected as

X=[Φ Φ _a ξ ω ε t _A ] ^T

in:

Φ=[φ _x φ _y φ _z ] is the attitude error angle of the inertial system X, Y, and Z axes;

Φ _a =[φ _ax φ _ay φ _az ] is the installation error angle of the three axes X, Y and Z of the carrier system;

ξ=[θ _x θ _y θ _z ] ^T is the deflection deformation of the mounting error angle of the three axes of the carrier system X, Y, and Z;

ω=[μ _x μ _y μ _z ] ^T is the deflection and deformation angular velocity of the mounting error angle of the three axes of the carrier system X, Y, and Z;

ε=[ε _x ε _y ε _z ] gyro drift of X, Y, Z axes;

t _A : Reference attitude delay time.

Written in the form of the equation of state as follows

T表示转置；B为系统噪声向量。其中，

可通过基于惯性系的导航解算获得。in

T represents the transpose; B is the system noise vector. in,

It can be obtained by an inertial frame-based navigation solution.

Step 5. Establish the measurement equation

The measurement equation is as follows

为陀螺输出三轴角速率；A和B的计算需要利用矩阵

由下式获得：in,

Output triaxial angular rate for the gyroscope; A and B calculations require the use of matrices

Obtained by:

Among them, t ₀ represents the initial time of filtering, and the specific calculation method of each matrix is:

Among them, γ ₀ , ψ ₀ and θ ₀ are the roll angle, heading angle and pitch angle of the reference inertial navigation system at the initial moment, respectively,

Among them, L ₀ and λ ₀ are the longitude and latitude of the reference inertial navigation system at the initial moment, respectively.

为单位阵，即：

is the unit matrix, that is:

为地球系到惯性系的变化矩阵，计算方法为：

is the change matrix from the earth system to the inertial system, and the calculation method is:

where ω _ie is the angular rate of the Earth's rotation, and t is time.

Among them, λ and L are the longitude and latitude of the reference inertial navigation system, respectively.

Among them, γ, ψ and θ are the roll angle, heading angle and pitch angle of the reference inertial navigation system, respectively.

则make

but

Step 6. Calculate the measured value

Step 7. Use the Kalman filter equation to calculate

After the above error model is established, the Kalman filter method is selected as the parameter identification method. The Kalman filter equation adopts the form in the document "Theory of Kalman Filter and Integrated Navigation" (first edition, edited by Qin Yongyuan, etc.), and the specific formula is as follows:

State one-step prediction

State estimation

Filter Gain Matrix

One-step forecast error variance matrix

Estimation Error Variance Matrix

P _k =[IK _k H _k ]P _k,k-1

为一步状态预测值，

为状态估计矩阵，Φ_k,k-1为状态一步转移矩阵，H_k为量测矩阵，Z_k为量测量，K_k为滤波增益矩阵，R_k为观测噪声阵，P_k,k-1为一步预测误差方差阵，P_k为估计误差方差阵，Γ_k,k-1为系统噪声驱动阵，Q_k-1为系统噪声阵。in,

is the one-step state prediction value,

is the state estimation matrix, Φ _{k, k-1} is the state one-step transition matrix, H _k is the measurement matrix, Z _k is the quantity measurement, K _k is the filter gain matrix, R _k is the observation noise matrix, P _{k, k-1} is the one-step prediction error variance matrix, P _k is the estimated error variance matrix, Γ _k,k-1 is the system noise driving matrix, and Q _k-1 is the system noise matrix.

Step 8. Closed-loop correction of installation error angle and deflection angle

The correction method is as follows:

为Let γ ^m =X _A (3), ψ ^m =X _A (4), θ ^m =X _A (5), then install the error angle matrix

for

为Let γ ^m =X _A (6), ψ ^m =X _A (7), θ ^m =X _A (8), then the deflection deformation matrix

for

计算新的

计算

以替换原

Depend on

calculate new

calculate

to replace the original

Step 9. Repeat the closed-loop correction of the deflection angle

Since the above steps 1 to 8 have corrected the error angle and deflection angle, but the error angle and deflection angle have not converged and the required accuracy cannot be achieved, it is necessary to repeat the correction of the error angle and deflection angle, so The above steps 1 to 8 are repeated until the estimation results of the error angle and the deflection angle are converged, and the method ends.

Claims (6)

1. A deflection estimation method for closed-loop correction of installation error, comprising nine steps, the first step is to define a coordinate system, the second step is to perform a navigation operation based on the inertial system, the third step is to establish a deflection error model, and the fourth step is to Establish the system error equation, step 5 is to establish the measurement equation, step 6 is to calculate the measurement value, step 7 is to use the Kalman filter equation to calculate, step 8 is to correct the installation error angle and deflection angle closed-loop, and step 9 is to repeat The closed-loop correction of deflection angle is characterized in that: the step defines a pair of coordinate systems, and in the attitude matching calculation method, an auxiliary coordinate system needs to be introduced: inertial coordinate system i ₁ : sub-inertial navigation calculation carrier system b at time t ₀ obtained by inertial solidification of the system; inertial coordinate system i ₂ : obtained by inertial solidification of the main inertial navigation system m at time t ₀ ; earth coordinate system e: the x-axis is along the direction of the earth's axis, and the y-axis is along the intersection of the equatorial plane and the Greenwich meridian plane On the equatorial plane, the z-axis forms a right-handed coordinate system with the x-axis and the y-axis; the earth's inertial coordinate system i: the inertial solidification of the earth's coordinate system e at time t ₀ is obtained; Described step 2, the navigation operation based on inertial frame; Set up a quaternion: Q ₁ =[q ₀ q ₁ q ₂ q ₃ ] ^T , and the initial value Q ₁ (0)=[1 0 0 0] ^T ; The analytical formula for quaternion update is:

in: I is a 4×4 order identity matrix;

Calculate the angular increment component of the missile system relative to the inertial system for the time k to k+1, that is, the angular increment sensitive to the gyro is transformed into the angular increment after calculating the missile system, which is obtained by the following formula:

where T _n is the navigation solution period,

is the angular velocity of the gyro in the x, y, and z directions,

The update calculation formula of the attitude matrix is:

The third step is to establish a deflection error model; Set the deflection deformation of the carrier platform

That is, the dynamic deformation angle of the child INS relative to the parent INS is:

where θ _x , θ _y , θ _z are the dynamic deformation angles of the child inertial navigation relative to the parent inertial navigation in the x, y, and z directions; Assume that the deflection angular velocity variable is

μ _x , μ _y , μ _z are the angular velocities of deflection in the x, y, and z directions. In addition, assuming that the deformation processes of the three axes are independent of each other, then

Then the second-order Markov equation of motion for flexural deformation can be obtained as

In the formula, β _i =2.146/τ _i , τ _i represents the relevant time constant of the deformation of the carrier platform in the i direction, i=x, y, z; w _x , w _y and w _z represent the white noise sequence, and the white noise sequence is The driving information of the deflection angle, its variances are Q _rx , Q _ry and Q _rz respectively, and Q _ri =4β ² _i σ ² _i ; In summary, the state equation of the deflection deformation of the carrier platform is:

Described step 4, establish systematic error equation; According to the error characteristics of the inertial navigation system, in the attitude matching Kalman filter model, 16 error state variables are selected as X=[Φ Φ _a ξ ω ε t _A ] ^T in: Φ=[φ _x φ _y φ _z ] is the attitude error angle of the inertial system X, Y, and Z axes; Φ _a =[φ _ax φ _ay φ _az ] is the installation error angle of the three axes X, Y and Z of the carrier system; ξ=[θ _x θ _y θ _z ] ^T is the deflection deformation of the mounting error angle of the three axes of the carrier system X, Y, and Z; ω=[μ _x μ _y μ _z ] ^T is the deflection and deformation angular velocity of the mounting error angle of the three axes of the carrier system X, Y, and Z; ε=[ε _x ε _y ε _z ] gyro drift of X, Y, Z axes; t _A : reference attitude delay time; Written in the form of the equation of state as follows:

in

T represents the transposition; B is the system noise vector; where,

Obtained by the inertial frame-based navigation solution. 2. A kind of deflection estimation method for closed-loop correction of installation error as claimed in claim 1, it is characterized in that: described step 5, establish measurement equation; The measurement equation is as follows:

in,

Output triaxial angular rate for the gyroscope; A and B calculations require the use of matrices

Obtained by:

Among them, t ₀ represents the initial time of filtering, and the specific calculation method of each matrix is:

Among them, γ ₀ , ψ ₀ and θ ₀ are the roll angle, heading angle and pitch angle of the reference inertial navigation system at the initial moment, respectively,

Among them, L ₀ and λ ₀ are the longitude and latitude of the reference inertial navigation system at the initial moment, respectively;

is the unit matrix, that is:

is the change matrix from the earth system to the inertial system, and the calculation method is:

Among them, ω _ie is the angular rate of the earth's rotation, and t is the time;

Among them, λ and L are the longitude and latitude of the reference inertial navigation system, respectively;

Among them, γ, ψ and θ are the roll angle, heading angle and pitch angle of the reference inertial navigation system, respectively; make

but

3. The deflection estimation method for closed-loop correction of installation error according to claim 2, characterized in that: in the step 6, calculating the measured value

4. the deflection estimation method of a kind of closed-loop correction installation error as claimed in claim 1, is characterized in that: described step 7, utilizes Kalman filter equation to calculate; After the above error model is established, the Kalman filter method is selected as the parameter identification method, and the specific formula is as follows: State one-step prediction

State estimation

Filter Gain Matrix

One-step forecast error variance matrix

Estimation Error Variance Matrix P _k =[IK _k H _k ]P _k,k-1 in,

is the one-step state prediction value,

is the state estimation matrix, Φ _{k, k-1} is the state one-step transition matrix, H _k is the measurement matrix, Z _k is the quantity measurement, K _k is the filter gain matrix, R _k is the observation noise matrix, P _{k, k-1} is the one-step prediction error variance matrix, P _k is the estimated error variance matrix, Γ _k,k-1 is the system noise driving matrix, and Q _k-1 is the system noise matrix. 5. A deflection estimation method for closed-loop correction of installation error as claimed in claim 2, characterized in that: in the eighth step, the closed-loop correction of the installation error angle and the deflection angle is performed; The correction method is as follows: Let γ ^m =X(3), ψ ^m =X(4), θ ^m =X(5), then install the error angle matrix

for

Let γ ^m =X(6), ψ ^m =X(7), θ ^m =X(8), then the deflection deformation matrix

for

Depend on

calculate new

calculate

to replace the original

6. A deflection estimation method for closed-loop correction of installation errors as claimed in claim 1, characterized in that: said step 9 repeats the closed-loop correction of deflection deformation angle, since the above steps one to eight pairs of error angles and deflection deflection angles Correction has been made, but the error angle and deflection angle do not converge, and the required accuracy cannot be achieved. Therefore, it is necessary to repeat the correction of the error angle and deflection angle. Therefore, repeat the above steps 1 to 8 until the error angle and deflection angle are repeated. The angle estimation results converge.