RINS fast zero offset estimation and alignment method based on indexing mechanism angle increment observation

Abstract

The invention relates to the technical field of inertial navigation, in particular to an indexing mechanism angle increment observation-based RINS (Remote Inertial Navigation System) rapid zero offset estimation and alignment method, which comprises the following steps: S1, defining a used reference coordinate system; s2, powering on the rotary inertial navigation system, resetting the module, and returning the sky axis to the zero position; s3, a data acquisition module acquires data output by the k-th frame of inertial measurement unit and an indexing mechanism photoelectric encoder; performing attitude, speed and position calculation on the data of the kth frame by using a data processing computer to obtain an inertia calculation result of the kth frame; s4, the data processing computer performs initial alignment of the kth frame; according to the method, the observability of the zero offset and course angle error of the sky-direction gyroscope in the initial alignment process can be remarkably improved, so that the zero offset and course angle error estimation of the gyroscope is faster in convergence and higher in precision; and the initial alignment time of the rotary inertial navigation system can be obviously shortened, the starting preparation time of the carrier is greatly shortened, and the quick response capability of the carrier is effectively enhanced.

Description

Technical Field

[0001] The present invention relates to the field of inertial navigation technology, and in particular to a method for fast zero bias estimation and alignment of a Rotational Inertial Navigation System (RINS) based on angular increment observation of a transfer mechanism, which can be used for fast startup and navigation of the RINS. Background Art

[0002] Inertial Navigation System (INS) is abbreviated as INS. In the process of navigation, INS is not affected by signal interference, masking and deception, and will not radiate signals outward, so it has strong concealment. Based on the above characteristics, INS is widely used in various land, sea, air and space vehicles in the military field to ensure that the vehicles accurately execute relevant instructions, complete and achieve established tasks. The traditional INS is fixed to the vehicle, which is called a strapdown INS. Its navigation error accumulates rapidly over time, so the accuracy is poor in long-distance navigation. In order to overcome the above defects, the concept of rotating INS was proposed, which periodically modulates the error through the mechanical rotation of the indexing mechanism, and can still maintain high-precision navigation in long-distance navigation. In the application process of rotating INS, initial alignment is an indispensable step, and high-precision initial alignment accuracy is the basis of high-precision navigation. In order to obtain a high-precision initial alignment attitude, the initial alignment takes a long time (usually 6 hours to 12 hours for rotating INS for navigation). Long initial alignment is usually used to estimate the gyro bias of the inertial navigation system using a Kalman filter, and feed the bias back to the alignment loop for correction, ultimately obtaining a high-precision attitude. If the alignment time is too short, the gyro bias estimation of the inertial navigation system is inaccurate, resulting in poor accuracy of the final alignment result. Current technical means make it difficult to simultaneously ensure that the initial alignment of a rotating inertial navigation system has both high accuracy and high speed. Therefore, in order to ensure that the vehicle has high-precision navigation while responding quickly, it is urgent to study high-precision and fast alignment methods. Summary of the invention

[0003] The present invention proposes a RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism, which can achieve fast initial alignment with high precision and effectively solve the problems existing in the prior art.

[0004] The present invention is implemented by the following technical solution: a RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism, which is divided into the following steps:

[0005] S1: Define the reference coordinate system used: n represents the navigation coordinate system, b represents the carrier coordinate system, s represents the measurement coordinate system of the inertial measurement unit, e represents the earth coordinate system, and i represents the inertial coordinate system. The coordinate system diagram is shown in 1. The detailed definition of the above coordinate systems can be found in the literature "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019";

[0006] S2: The rotating inertial navigation system is powered on, the module is reset, and the azimuth axis returns to zero position:

[0007] S2.1 Rotational inertial navigation system is powered on;

[0008] S2.2 The inertial measurement unit, data processing computer, data acquisition module, indexing mechanism control module, and indexing mechanism photoelectric encoder module in the rotating inertial navigation system are reset; the counter is reset to k=0, where k represents the number of frames of data acquisition;

[0009] S2.3 The transfer mechanism control module controls the azimuth axis to rotate to the initial zero position;

[0010] S2.4 performs initial parameter settings:

[0011] Set the coarse alignment time T c With precise alignment time T f , usually the coarse alignment time T c Set to 300 seconds, precise alignment time T f Set to 3600 seconds;

[0012] S3: The data acquisition module collects the data output by the k-th frame inertial measurement unit and the photoelectric encoder of the indexing mechanism; the data of the k-th frame is solved for attitude, speed and position by the data processing computer to obtain the k-th frame inertial solution result; the details are as follows:

[0013] S3.1 The data acquisition module collects the output data of the inertial measurement unit and the photoelectric encoder of the indexing mechanism of the kth frame. k =[Δθ k Δv k ω k ] T , where the superscript "T" represents the transpose of the matrix, and its sampling time interval is Δt. k =[Δθ x,k ,Δθ y,k ,Δθ z,k ] T is the k-th frame angular increment of the gyro output in the inertial measurement unit, Δv k =[Δv x,k ,Δv y,k ,Δv z,k ] Tis the k-th frame velocity increment output by the accelerometer in the inertial measurement unit, ω k The angular velocity information output by the photoelectric encoder. If it is a single-axis RINS, the system has a celestial rotation axis, and the rotation angular velocity of the axis is expressed as ω k =[0 0ω z,k ] T ,ω z,k Indicates the rotational angular velocity of the celestial axis; if it is a dual-axis RINS, the system includes a horizontal rotation axis (y-axis) and a celestial rotation axis, and the rotational angular velocity is expressed as ω k =[0 ω y,k ω z,k ] T ,ω y,k represents the angular velocity of the horizontal rotation axis y-axis; if it is a three-axis RINS, the system includes two horizontal rotation axes (x-axis and y-axis) and one celestial rotation axis, then ω k =[ω x,k ω y,k ω z,k ] T ,ω x,k Indicates the rotational angular velocity of the two horizontal axes x-axis; the present invention only needs to rotate the celestial axis to enhance the convergence speed of the gyro zero bias and heading angle error. Single-axis, dual-axis and three-axis RINS all contain the celestial rotation axis, so the present invention is applicable to the above three systems;

[0014] S3.2 If k <T c , then save the k-th frame output data A k To the data processing computer, k=k+1, and then return to S3.1 to collect a new frame of data;

[0015] If k = T c , use the strapdown inertial navigation rough alignment method to obtain the rough initial attitude of the vehicle Will Save to the data processing computer, set k = k + 1, and return to S3.1 to collect a new frame of data; the strapdown inertial navigation coarse alignment method can refer to the literature "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019";

[0016] If k>T c , skip step S3.2 and go to S3.3;

[0017] S3.3 If k = T c +1, initial speed of data processing computer binding carrier and the local coordinate position of the vehicle Where L0 represents the local latitude, λ0 represents the local longitude, and h0 represents the local altitude. The initial attitude matrix of the vehicle is defined as

[0018] If k>T c +1, go directly to S3.4;

[0019] S3.4 The transfer mechanism control module controls the celestial axis of the transfer mechanism to rotate at a uniform speed, and the angular velocity is ω z,k , take ω z,k =10° / s;

[0020] S3.5 Using the inertial measurement unit data Δθ of the kth frame k , Δv k And the initial posture matrix of the carrier at frame k-1 Initial Speed And the local coordinates Perform inertial solution to obtain the inertial solution result of the kth frame, including the attitude matrix of the kth frame speed and location

[0021] and are the eastward, northward and celestial components of the velocity of the kth frame, Respectively represent the latitude, longitude and altitude values ​​of the kth frame. The above inertial solution method can be referred to the literature "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019";

[0022] It should be noted that the k-1th frame used The k-th frame calculated by the solution There are horizontal lines and wavy lines above the letters. The wavy line indicates the result obtained by direct solution, which contains the solution error; the horizontal line indicates the result after error correction at the k-1th moment;

[0023] The pose matrix of the kth frame It can be expressed as follows:

[0024]

[0025] The physical meaning of the matrix elements in formula (1) can be found in the description of the attitude matrix in Appendix B of the document "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019";

[0026] S4: The data processing computer performs initial alignment of the kth frame:

[0027] S4.1 Calculate the discretized state transfer matrix Φ from time k-1 to time k k|k-1 :

[0028] Φ k|k-1 ≈I 12×12 +F k Δt (2)

[0029] I 12×12 is the 12th-order identity matrix;

[0030] F k is the continuous state transfer matrix at time k:

[0031]

[0032] In formula (3), F 11 ,F 12 ,F 13 ,F 21 ,F 22 ,F 24 Each block matrix is ​​calculated by the following formula:

[0033]

[0034]

[0035] In the formula, ω ie is the angular velocity of the Earth's rotation; R N and R M are the radius of the earth's meridian circle and the radius of its circumpolar circle respectively;; f E,k 、f N,k and f U,k They are the velocity increment Δv output by the accelerometer k =[Δv x,k ,Δv y,k ,Δv z,k ] T Divide by the sampling time Δt and then convert to the navigation coordinate system. The calculation formula is as follows:

[0036]

[0037] S4.3 Time update process for initial alignment

[0038] Calculate the one-step prediction value of the state vector from time k-1 to time k And the corresponding mean square error matrix P kk-1 :

[0039]

[0040] In formula (5), is the posterior estimate of the state vector at time k-1, P k-1|k-1 is the mean square error matrix of the state estimation at time k-1, Q is the mean square error matrix of the white noise output by the inertial measurement unit; Represents Φ k|k-1 The transposed matrix of

[0041] A posteriori estimate of the state vector Initial value of Set to:

[0042]

[0043] State estimation mean square error matrix P k-1k-1 The initial value is set to:

[0044]

[0045] The mean square error matrix Q of the inertial measurement unit output white noise is set as:

[0046]

[0047] S4.4 Constructing the observation matrix H k :

[0048]

[0049] Compared with the traditional observation matrix based on speed observation and speed position observation, the observation matrix H constructed by the present invention is k In the fourth row, a change was made to add the error related to the misalignment angle and Three items, including It is related to the heading angle error, which is equivalent to being able to directly observe the heading angle error. A Δt term related to the gyro celestial bias is also added, which can effectively improve the observability of the heading angle error and the gyro celestial bias during the initial alignment process.

[0050] S4.5 Construct the observation vector Z including the angular increment of the indexing mechanism k :

[0051]

[0052] In formula (10), ω z,k Δt is the angular increment of the indexing mechanism; δV E,k , δV N,k and δV U,k are the observed values ​​of velocity errors in the east, north and celestial directions, respectively. Since the velocity of the vehicle is 0 during the initial alignment, it can be considered that the velocity error obtained by S3.5 is is equal to the speed error, that is:

[0053]

[0054] In addition, in formula (10), for The transpose of is the projection of the earth's rotation angular velocity in the navigation coordinate system, express and The third row of the 3×1 matrix obtained after multiplication; is the increment of the heading angle at time k:

[0055]

[0056] In formula (12), is the heading angle calculated at the kth moment, which can be obtained by Calculated; atan2(.,.) is the inverse tangent function in the standard C language function library; k is the time The first row and second column element and the second row and second column element; is the element corresponding to the k-1th moment;

[0057] S4.6 Measurement update process for initial alignment

[0058] The measurement update process is shown in formula (13), which is the standard formula for the Kalman filter measurement update process. The filter gain K at the kth moment is calculated in sequence: k , the posterior estimate of the state vector The state vector posterior estimate mean square error matrix P k|k :

[0059]

[0060] In formula (13), I is a 12×12 unit matrix, and the measurement noise covariance matrix R is set to: R = diag{(0.01 m / s) 2 ,(0.01m / s) 2 ,(0.01m / s) 2 ,(0.001°) 2}.

[0061] S4.7 Feedback correction for initial alignment

[0062] The pose matrix of the kth frame obtained by S3.5 is speed Correction is performed to obtain the corrected attitude matrix and speed

[0063]

[0064] The pointed hat above the letter indicates the estimated value of the corresponding element. is the estimated value of the misalignment angle error in the east, north and sky directions at time k calculated by S4.4, are the estimated values ​​of velocity errors in the east, north and sky directions at time k calculated by S4.4, and their values ​​are:

[0065]

[0066] in, Represents the posterior estimate of the 12-dimensional state vector at time k The values ​​of the 1st to 6th dimensions of ; In addition, The three values ​​are the estimated gyro bias values ​​of the x, y, and z axes at time k. The three values ​​are the estimated values ​​of the accelerometer zero bias of the x, y, and z axes at time k;

[0067] S5 when k <T f , counter k=k+1, return to S3.1 and execute the loop;

[0068] When k = T f , output the initial alignment result The initial alignment process is completed, the celestial axis stops rotating, and the subsequent navigation process is started.

[0069] At this point, a high-precision initial alignment attitude is obtained, which will be used to set the initial value of the subsequent pure inertial navigation of the rotating inertial navigation system.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. The present invention can significantly improve the observability of the gyro bias in the initial alignment process, making the gyro bias estimation converge faster and more accurate;

[0072] 2. The present invention can significantly improve the observability of the heading angle error during the initial alignment process, so that the heading angle error estimation of the initial alignment converges faster and has higher accuracy;

[0073] 2. The present invention can significantly shorten the initial alignment time of the rotating inertial navigation system, thereby greatly shortening the start-up preparation time of the vehicle (the alignment time of the rotating inertial navigation system for navigation can be shortened from 6 hours at port to 1 hour), effectively enhancing the rapid response capability of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 : Flow chart of initial alignment of the present invention;

[0075] Figure 2 : A curve diagram of the celestial gyro bias estimation of the embodiment;

[0076] Figure 3 : Graph of heading angle alignment results of the embodiment. DETAILED DESCRIPTION

[0077] In order to explain the technical solution disclosed in the present invention in detail, further elaboration is made below in conjunction with specific embodiments. The specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0078] Figure 1 The implementation process of the present invention is divided into the following steps:

[0079] S1: Define the reference coordinate system used: n represents the navigation coordinate system, b represents the carrier coordinate system, s represents the measurement coordinate system of the inertial measurement unit, e represents the earth coordinate system, and i represents the inertial coordinate system. The coordinate system diagram is shown in 1. The detailed definition of the above coordinate systems can be found in the literature "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019";

[0080] S2: The rotating inertial navigation system is powered on, the module is reset, and the azimuth axis returns to zero position;

[0081] S3: The data acquisition module collects the data output by the k-th frame inertial measurement unit and the photoelectric encoder of the indexing mechanism; the data of the k-th frame is solved for attitude, speed and position by the data processing computer to obtain the inertial solution result of the k-th frame;

[0082] S4: The data processing computer performs initial alignment of the kth frame.

[0083] The feasibility of the present invention can be verified by the following simulation experiments:

[0084] First, a set of simulation experimental data is generated by using the strapdown inertial navigation inversion algorithm (see the literature "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019"), and the local coordinate position of the simulation experiment is [28.22°112.99°60m] T The error parameters of the gyroscope and accelerometer added to the simulation experimental data are shown in Table 1:

[0085] Table 1 Simulation error added value

[0086]

[0087] Next, the embodiment of the present invention is based on Figure 1 Execute, where:

[0088] The initial parameter settings of S2.4 are as follows:

[0089] Set the coarse alignment time to 5 minutes, then T c =5*60*200=120000;

[0090] Set the fine alignment time to 1 hour, then T f =1*3600*200=720000;

[0091] Set the initial value of the state vector to

[0092]

[0093] Set the initial value of the state estimation mean square error matrix to:

[0094] P 00 =diag{(0.001) 2 ,(0.001°) 2 ,(0.001°) 2 ,(0.01m / s) 2 ,(0.01m / s) 2 ,(0.01m / s) 2 ,(0.01° / h) 2 ,(0.01° / h) 2 ,(0.01° / h) 2 ,(1×10 -4 m / s) 2 ,(1×10 -4 m / s) 2 ,(1×10 -4 m / s) 2}

[0095] Set the mean square error matrix Q of the inertial measurement unit output white noise to:

[0096] Q = diag {(0.0005° / h) 2 ,(0.0005° / h) 2 ,(0.0005° / h) 2 ,

[0097] (10×10 -6 m / s) 2 ,(10×10 -6 m / s) 2 ,(10×10 -6 m / s) 2 ,

[0098] 0,0,0,0,0,0}

[0099] In S3.1, set the sampling time interval Δt = 0.005s;

[0100] In S3.4, the angular velocity of the celestial axis is set to ω z,k =10° / s, uniform rotation;

[0101] After 1 hour of initial alignment, the initial alignment zero bias estimation result is as follows: Figure 2 As shown in Table 2, the initial alignment heading angle estimation results are as follows Figure 3 As shown in Table 3. As can be seen from Table 2 and Table 3, compared with the traditional initial alignment without the angle increment observation of the indexing mechanism, the zero bias estimation accuracy and the heading angle estimation result accuracy of the present invention are significantly improved, which verifies the effectiveness of the present invention.

[0102] Table 2 Initial alignment bias estimation results

[0103] Initial Alignment Method Estimated gyro bias value Simulation settings real value Estimation accuracy Traditional methods 0.00066° / h 0.001° / h 66％ The present invention 0.00092° / h 0.001° / h 92％

[0104] Table 3 Initial alignment heading angle estimation results

[0105] Initial Alignment Method Heading angle alignment result Simulation settings real value Alignment error value Traditional methods 25.0053° 25° 0.0053° The present invention 25.0003° 25° 0.0003°

Claims

1. A RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism, characterized in that: The method is divided into the following steps: S1: Define the reference coordinate system used: n represents the navigation coordinate system, b represents the carrier coordinate system, s represents the measurement coordinate system of the inertial measurement unit, e represents the earth coordinate system, and i represents the inertial coordinate system; S2: The rotating inertial navigation system is powered on, the module is reset, and the azimuth axis returns to zero position: S2.1 Rotational inertial navigation system is powered on; S2.2 The inertial measurement unit, data processing computer, data acquisition module, indexing mechanism control module, and indexing mechanism photoelectric encoder module in the rotating inertial navigation system are reset; the counter is reset to k=0, where k represents the number of frames of data acquisition; S2.3 The transfer mechanism control module controls the azimuth axis to rotate to the initial zero position; S2.4 performs initial parameter settings: Set the coarse alignment time T c With precise alignment time T f ; S3: The data acquisition module collects the data output by the k-th frame inertial measurement unit and the photoelectric encoder of the indexing mechanism; the data of the k-th frame is solved for attitude, speed and position by the data processing computer to obtain the k-th frame inertial solution result; the details are as follows: S3.1 The data acquisition module collects the output data of the inertial measurement unit and the photoelectric encoder of the indexing mechanism of the kth frame. k =[Δθ k Δv k ω k ] T , where the superscript "T" represents the transpose of the matrix, and the sampling time interval is Δt; where Δθ k =[Δθ x,k ,Δθ y,k ,Δθ z,k ] T is the k-th frame angular increment of the gyro output in the inertial measurement unit, Δv k =[Δv x,k ,Δv y,k ,Δv z,k ] T is the k-th frame velocity increment output by the accelerometer in the inertial measurement unit, ω k The angular velocity information output by the photoelectric encoder. If it is a single-axis RINS, the system has a celestial rotation axis, and the rotation angular velocity of the axis is expressed as ω k =[00ω z,k ] T ,ω z,k Indicates the rotational angular velocity of the celestial axis; if it is a dual-axis RINS, the system includes a horizontal rotation axis and a celestial rotation axis, and the rotational angular velocity is expressed as ω k =[0ω y,k ω z,k ] T ,ω y,k represents the angular velocity of the horizontal rotation axis y-axis; if it is a three-axis RINS, the system includes two horizontal rotation axes and one celestial rotation axis, then ω k =[ω x,k ω y,k ω z,k ] T ,ω x,k Represents the angular velocity of rotation of the two horizontal axes x-axis; S3.2 If k <T c , then save the k-th frame output data A k To the data processing computer, k=k+1, and then return to S3.1 to collect a new frame of data; If k = T c , use the strapdown inertial navigation rough alignment method to obtain the rough initial attitude of the vehicle Will Save to the data processing computer, set k=k+1, and return to S3.1 to collect a new frame of data; If k>T c , skip step S3.2 and go to S3.3; S3.3 If k = T c +1, initial speed of data processing computer binding carrier and the local coordinate position of the vehicle Where L0 represents the local latitude, λ0 represents the local longitude, and h0 represents the local altitude. The initial attitude matrix of the vehicle is defined as If k>T c +1, go directly to S3.4; S3.4 The transfer mechanism control module controls the celestial axis of the transfer mechanism to rotate at a uniform speed, and the angular velocity is ω z,k ; S3.5 Using the inertial measurement unit data Δθ of the kth frame k , Δv k And the initial posture matrix of the carrier at frame k-1 Initial Speed And the local coordinates Perform inertial solution to obtain the inertial solution result of the kth frame, including the attitude matrix of the kth frame speed and location and are the eastward, northward and celestial components of the velocity of the kth frame, Respectively represent the latitude, longitude and altitude values ​​of the kth frame; The pose matrix of the kth frame It can be expressed as follows: The physical meaning of the matrix elements in formula (1) can be found in the description of the attitude matrix in Appendix B of the document "Yan Gongmin, Weng Jun. Strapdown Inertial Navigation Algorithm and Integrated Navigation Principle [M]. Northwestern Polytechnical University Press, 2019"; S4: The data processing computer performs initial alignment of the kth frame: S4.1 Calculate the discretized state transfer matrix Φ from time k-1 to time k k|k-1 : F k|k-1 ≈I 12×12 +F k Δt (2) I 12×12 is the 12th-order identity matrix; F k is the continuous state transfer matrix at time k: In formula (3), F 11 ,F 12 ,F 13 ,F 21 ,F 22 ,F 24 Each block matrix is ​​calculated by the following formula: In the formula, ω ie is the angular velocity of the Earth's rotation; R N and R M are the radius of the earth's meridian circle and the radius of its circumpolar circle respectively;; f E,k 、f N,k and f U,k They are the velocity increment Δv output by the accelerometer k =[Δv x,k ,Δv y,k ,Δv z,k ] T Divide by the sampling time Δt and then convert to the navigation coordinate system. The calculation formula is as follows: S4.3 Time update process for initial alignment Calculate the one-step prediction value of the state vector from time k-1 to time k And the corresponding mean square error matrix P k|k-1 : In formula (5), is the posterior estimate of the state vector at time k-1, P k-1|k-1 is the mean square error matrix of the state estimation at time k-1, Q is the mean square error matrix of the white noise output by the inertial measurement unit; Represents Φ k|k-1 The transposed matrix of S4.4 Constructing the observation matrix H k : S4.5 Construct the observation vector Z including the angular increment of the indexing mechanism k : In formula (10), ω z,k Δt is the angular increment of the indexing mechanism; δV E,k , δV N,k and δV U,k are the observed values ​​of velocity errors in the east, north and celestial directions, respectively. Since the velocity of the vehicle is 0 during the initial alignment, it can be considered that the velocity error obtained by S3.5 is is equal to the speed error, that is: In addition, in formula (10), for The transpose of is the projection of the earth's rotation angular velocity in the navigation coordinate system, express and The third row of the 3×1 matrix obtained after multiplication; is the increment of the heading angle at time k: In formula (12), is the heading angle calculated at the kth moment, which can be obtained by Calculated; atan2(.,.) is the inverse tangent function in the standard C language function library; k is the time The first row and second column element and the second row and second column element; is the element corresponding to the k-1th moment; S4.6 Measurement update process for initial alignment The measurement update process is shown in formula (13), which is the standard formula for the Kalman filter measurement update process. The filter gain K at the kth moment is calculated in sequence: k , the posterior estimate of the state vector The state vector posterior estimate mean square error matrix P k|k : In formula (13), I is a 12×12 unit matrix, and the measurement noise covariance matrix R is set to: R = diag{(0.01 m / s) 2 ,(0.01m / s) 2 ,(0.01m / s) 2 ,(0.001°) 2 }; S4.7 Feedback correction for initial alignment The pose matrix of the kth frame obtained by S3.5 is speed Correction is performed to obtain the corrected attitude matrix and speed The pointed hat above the letter indicates the estimated value of the corresponding element. is the estimated value of the misalignment angle error in the east, north and sky directions at time k calculated by S4.4, are the estimated values ​​of velocity errors in the east, north and sky directions at time k calculated by S4.4, and their values ​​are: in, Represents the posterior estimate of the 12-dimensional state vector at time k The values ​​of the 1st to 6th dimensions of ; In addition, The three values ​​are the estimated gyro bias values ​​of the x, y, and z axes at time k. The three values ​​are the estimated values ​​of the accelerometer zero bias of the x, y, and z axes at time k; S5 when k <T f , counter k=k+1, return to S3.1 and execute the loop; When k = T f , output the initial alignment result The initial alignment process is completed, the celestial axis stops rotating, and the subsequent navigation process is started; At this point, a high-precision initial alignment attitude is obtained, which will be used to set the initial value of the subsequent pure inertial navigation of the rotating inertial navigation system.

2. The RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism according to claim 1 is characterized by: In S2.4, the coarse alignment time T c Set to 300 seconds, precise alignment time T f Set to 3600 seconds.

3. The RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism according to claim 1 is characterized by: In S3.4, the angular velocity ω z,k =10° / s.

4. The RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism according to claim 1 is characterized in that: In S4.3, the posterior estimate of the state vector Initial value of Set to:

5. According to claim 1, the RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism is characterized by: In S4.3, the state estimation mean square error matrix P k-1|k-1 The initial value is set to:

6. The RINS fast zero bias estimation and alignment method based on the angular increment observation of the indexing mechanism according to claim 1 is characterized by: In S4.3, the mean square error matrix Q of the white noise output by the inertial measurement unit is set to:

Images