Method for constructing biaxial rotation inertial derivative character machine based on LSTM (Long Short Term Memory) algorithm

Abstract

The invention relates to a biaxial rotation inertial derivative character machine construction method based on an LSTM algorithm, and belongs to the technical field of inertial navigation, and the method comprises the following steps: S1, constructing a biaxial rotation inertial navigation error propagation equation considering an error modulation effect; s2, setting a transposition sequence, a rotation speed and rotation stop time of double-shaft rotation inertial navigation; s3, performing error modulation on the actual prototype of the double-shaft rotation inertial navigation according to the transposition sequence, the rotation speed and the rotation stopping time to obtain a navigation result of the actual prototype; comparing the navigation result of the actual prototype with the navigation information reference to obtain the navigation error output of the actual prototype; s4, using an LSTM algorithm to obtain inertial device error parameters in the error propagation equation; and substituting the inertial device error parameter, the navigation result, the navigation error output, the transposition sequence, the rotation speed and the rotation and stop time into an error propagation equation to obtain a digital prototype of double-shaft rotation inertial navigation. The method consumes less time and is low in cost.

Description

Technical Field

[0001] The present invention belongs to the technical field of inertial navigation, and in particular relates to a method for constructing a dual-axis rotating inertial navigation digital prototype based on an LSTM algorithm. Background Art

[0002] Rotating inertial navigation uses a rotation modulation strategy to suppress the error of inertial devices within a cycle by rotating, thereby reducing the impact of errors on the navigation accuracy of the system. At present, the mainstream rotation modulation technologies include single-axis rotation modulation technology and dual-axis rotation modulation technology. The single-axis rotating inertial navigation system can only suppress some errors in two directions, and the error in the other direction still propagates according to the original law. The dual-axis rotating inertial navigation system can better suppress errors in all directions and is widely used in long-duration navigation scenarios. In the process of rotation modulation of the dual-axis inertial navigation system, the rotation speed and the rotation-stop time are the most important parameters affecting the error modulation results.

[0003] At present, in the research of dual-axis rotating inertial navigation error modulation, the actual prototype of dual-axis rotating inertial navigation is usually used for error modulation research, which has the problems of long time consumption and high cost. Summary of the invention

[0004] In view of the deficiencies in the prior art, the object of the present invention is to provide a method for constructing a dual-axis rotational inertial navigation digital prototype based on the LSTM algorithm, so as to solve or improve the defects in the prior art.

[0005] To achieve the above object, the present invention adopts the following technical solution: a method for constructing a dual-axis rotating inertial navigation digital prototype based on LSTM algorithm, comprising the following steps: S1. Construct the error propagation equation of dual-axis rotational inertial navigation considering the error modulation effect; S2. Setting the rotation order, rotation speed and rotation stop time of the dual-axis rotation inertial navigation system; S3, placing the actual prototype of the dual-axis rotation inertial navigation on the table, performing error modulation on it according to the indexing order, rotation speed and rotation stop time set in step S2, and obtaining the navigation result of the actual prototype; comparing the navigation result of the actual prototype with the navigation information reference, and obtaining the navigation error output of the actual prototype; S4. Based on the error propagation equation constructed in step S1, the LSTM algorithm is used to obtain the error parameters of the inertial device in the error propagation equation; the obtained inertial device error parameters, the navigation result and navigation error output obtained in step S3, and the indexing order, rotation speed and rotation-stop time set in step S2 are substituted into the error propagation equation constructed in step S1 to obtain a digital prototype of the dual-axis rotation inertial navigation.

[0006] Preferably, in step S1, the specific method of constructing the error propagation equation of the dual-axis rotating inertial navigation system considering the error modulation effect includes: S11, constructing attitude error equation; S12, constructing a velocity error equation; S13. Construct a position error equation.

[0007] Preferably, in step S11, the specific method of constructing the posture error equation is: The posture error equation is constructed as follows: In the formula, is φ n The derivative with respect to time, φ n =[φ E ,φ N ,φ U ] T ,φ E ,φ N ,φ U They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the rotation angular rate error of the navigation system relative to the inertial system, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, Indicates the gyro measurement error in the navigation system; in, In the formula, is the angular velocity of the Earth's rotation relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, V N and V E Respectively represent the northward speed and eastward speed of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the rotation angular rate error of the navigation system relative to the inertial system, is the rotation angular rate error of the Earth relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth; in, In the formula, is the rotation angular rate error of the Earth relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, and δL represents the local latitude error; is the angular velocity of the navigation system relative to the Earth, V N and V E Respectively represent the northward and eastward velocities of the carrier in the geographic coordinate system, δV N and δV E They represent the northward velocity error and eastward velocity error of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the transformation matrix from the turntable coordinate system to the carrier system, is the conversion matrix from the carrier system to the navigation system; in, In the formula, is the transformation matrix from the turntable coordinate system to the carrier system, ω is the rotation speed, t is the inertial navigation operation time, t s The transfer stop time; in, In the formula, is the conversion matrix from the carrier system to the navigation system, θ is the pitch angle, γ is the roll angle, and ψ is the heading angle; in, In the formula, S g is the gyro scale error matrix, δG is the gyro installation error matrix, is the actual angular rate input of the gyroscope, ε is the zero bias of the gyroscope, is the random drift of the gyroscope; in, In the formula, S g is the gyro scale error matrix, is the symmetry scale error of the X-gyro, is the asymmetric scaling error of the X-gyro, is the symmetry scale error of the Y gyro, is the asymmetric scaling error of the Y gyro, is the symmetry scale error of the Z gyro, is the asymmetric scaling error of the Z gyro, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; sign(·) is the sign function, expressed as in, Where δG is the gyro installation error matrix, is the installation error of the Y gyro and the Z gyro, is the installation error of Z gyro and Y gyro, is the installation error of Z gyro and X gyro; in, In the formula, is the actual angular rate input of the gyroscope, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; Where ε=[ε x ,ε y ,ε z ] T ; Where ε is the gyro's zero bias, ε x is the zero bias of the X gyro, ε y is the zero bias of the Y gyro, ε z is the zero bias of the Z gyro; in, In the formula, is the random drift of the gyroscope, ω dx is the random drift of the X gyro, ω dy is the random drift of the Y gyro, ω dz is the random drift of the Z gyro.

[0008] Preferably, in step S12, the specific method of constructing the speed error equation is: The velocity error equation is constructed as follows: In the formula, is δv n The derivative with respect to time, δv n is the velocity error, v n is the speed, f n is the specific force measured by the accelerometer; φ n =[φ E ,φ N ,φ U ] T ,φ E ,φ N ,φU They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the accelerometer measurement error, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the rotation angular rate error of the Earth relative to the inertial system, is the rotation angular rate of the navigation system relative to the earth, δg is the gravity acceleration vector error; in, In the formula, is the accelerometer measurement error, S a is the scale factor error matrix of the accelerometer, δA is the installation error matrix of the accelerometer, is the actual acceleration input of the accelerometer, is the zero bias of the accelerometer, is the random drift of the accelerometer; in, In the formula, S a is the scale factor error matrix of the accelerometer, is the symmetry scale error of the X-accelerometer, is the asymmetric scale error of the X accelerometer, is the symmetry scale error of the Y accelerometer, is the asymmetric scaling error of the Y accelerometer, is the symmetry scale error of the Z accelerometer, is the asymmetric scaling error of the Z accelerometer, is the actual acceleration input of the X accelerometer, is the actual acceleration input of the Y accelerometer, is the actual acceleration input of the Z accelerometer; sign(·) is the sign function, expressed as in, Where δA is the installation error matrix of the accelerometer, η xz is the installation error of the X accelerometer and the Z accelerometer, η xy is the installation error of the X accelerometer and the Y accelerometer, η yz is the installation error of the Y accelerometer and the Z accelerometer, η yx is the installation error between the Y accelerometer and the X accelerometer, η zy is the installation error of the Z accelerometer and the Y accelerometer, η zxis the installation error of the Z accelerometer and the X accelerometer; in, In the formula, is the zero bias of the accelerometer, is the zero bias of the X accelerometer, is the zero bias of the Y accelerometer, is the zero bias of the Z accelerometer; in, In the formula, is the random drift of the accelerometer, f dx is the random drift of the X accelerometer, f dy is the random drift of the Y accelerometer, f dz is the random drift of the Z accelerometer.

[0009] Preferably, in step S13, the specific method of constructing the position error equation is: The position error equation is constructed as follows: In the formula, is the derivative of δL with respect to time, is the derivative of δλ with respect to time, is the derivative of δh with respect to time, L is latitude, λ is longitude, h is height, R is the average radius of the earth, v N is the north velocity, v E is the eastward velocity, δv N is the north velocity error, δv E is the eastward velocity error, δv U is the celestial velocity error, δψ is the heading error, δL is the latitude error, δλ is the longitude error, and δh is the height error.

[0010] Preferably, in step S2, the specific method of setting the indexing order of the dual-axis rotary inertial navigation system is: Set multiple sequences, each sequence rotates 180° forward or reverse around the celestial direction or east direction in the coordinate system at a rotation speed of ω. The directions or rotations of two adjacent sequences are different, with t s The stop time is still; The rotation cycle is repeated with all the set sequences as one rotation cycle.

[0011] Preferably, in step S2, the indexing order of the dual-axis rotary inertial navigation is set as: Order 1: Rotate 180° around the celestial direction at a speed of ω, and then s The stop time is still; Order 2: Rotate 180° around the east direction at a speed of ω, and then rotate 180° at ts The stop time is still; Order 3, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 4, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 5, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 6: Rotate 180° around the celestial direction at a speed of ω, and then rotate 180° at t s The stop time is still; Order 7, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 8, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 9, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 10, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 11, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 12, rotate 180° around the east direction with ω as the rotation speed, and s The stop time is still; Order 13, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 14, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 15, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 16, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; The sequence 1 to sequence 16 is a rotation cycle, and the rotation cycle is repeated.

[0012] Preferably, in step S2, the rotation speed of the dual-axis rotating inertial navigation system is set to 5° / s.

[0013] Preferably, in step S2, the rotation and stop time of the dual-axis rotating inertial navigation system is set to 2s.

[0014] Preferably, in step S3, the navigation information benchmark includes a benchmark for attitude navigation information, a benchmark for speed navigation information and a benchmark for position navigation information, the benchmark for attitude navigation information is the installation reference plane of the actual prototype, the benchmark for speed navigation information is 0, and the benchmark for position navigation information is the local geographic coordinates.

[0015] Preferably, in step S4, the inertial device error parameters include the gyro scale error matrix S g , gyro installation error matrix δG, gyro zero bias ε, gyro random drift The scale factor error matrix S of the accelerometer a , the installation error matrix of the accelerometer δA, the zero bias of the accelerometer Random drift of the accelerometer

[0016] Compared with the prior art, the present invention has the following beneficial effects: a dual-axis rotation inertial navigation digital prototype construction method based on the LSTM algorithm of the present invention first constructs the error propagation equation of the dual-axis rotation inertial navigation considering the error modulation effect, and then sets the indexing order, rotation speed and rotation stop time of the dual-axis rotation inertial navigation, and then obtains the navigation result and the navigation error output through the actual prototype of the dual-axis rotation inertial navigation, and finally obtains the inertial device error parameter in the error propagation equation based on the LSTM algorithm, thereby obtaining the digital prototype of the dual-axis rotation inertial navigation, reducing the time and cost of the dual-axis rotation inertial navigation error modulation research, and having simple operation, less time consumption, low cost and good practicality, and solving the problem that in the current dual-axis rotation inertial navigation error modulation research, the actual prototype of the dual-axis rotation inertial navigation is usually used for error modulation research, which is time-consuming and costly. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings in the following description without paying any creative work.

[0018] Figure 1 The present invention is a flowchart of a method for constructing a dual-axis rotating inertial navigation digital prototype based on an LSTM algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In order to make the above-mentioned features and advantages of the present invention more obvious and easy to understand, the following embodiments are specifically cited and described in detail with the drawings.

[0020] like Figure 1 As shown, an embodiment of the present invention provides a method for constructing a dual-axis rotating inertial navigation digital prototype based on an LSTM algorithm, comprising the following steps: S1. Construct the error propagation equation of dual-axis rotational inertial navigation considering the error modulation effect; S2. Setting the rotation order, rotation speed and rotation stop time of the dual-axis rotation inertial navigation system; S3, placing the actual prototype of the dual-axis rotation inertial navigation on a table (such as a marble table), performing error modulation on it according to the indexing order, rotation speed and rotation stop time set in step S2, and obtaining the navigation result of the actual prototype; comparing the navigation result of the actual prototype with the navigation information reference, and obtaining the navigation error output of the actual prototype; S4. Based on the error propagation equation constructed in step S1, the LSTM algorithm is used to obtain the error parameters of the inertial device in the error propagation equation; the obtained inertial device error parameters, the navigation result and navigation error output obtained in step S3, and the indexing order, rotation speed and rotation-stop time set in step S2 are substituted into the error propagation equation constructed in step S1 to obtain a digital prototype of the dual-axis rotation inertial navigation.

[0021] In this embodiment, in step S1, the specific method of constructing the error propagation equation of the dual-axis rotating inertial navigation system considering the error modulation effect includes: S11, constructing attitude error equation; S12, constructing a velocity error equation; S13. Construct a position error equation.

[0022] In this embodiment, in step S11, the specific method of constructing the posture error equation is: The posture error equation is constructed as follows: In the formula, is φ n The derivative with respect to time, φ n =[φ E ,φ N ,φ U ] T,φ E ,φ N ,φ U They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the rotation angular rate error of the navigation system relative to the inertial system, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, Indicates the gyro measurement error in the navigation system; in, In the formula, is the angular velocity of the Earth's rotation relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, V N and V E Respectively represent the northward speed and eastward speed of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the rotation angular rate error of the navigation system relative to the inertial system, is the rotation angular rate error of the Earth relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth; in, In the formula, is the rotation angular rate error of the Earth relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, and δL represents the local latitude error; is the angular velocity of the navigation system relative to the Earth, V N and V E Respectively represent the northward and eastward velocities of the carrier in the geographic coordinate system, δV N and δV E They represent the northward velocity error and eastward velocity error of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the transformation matrix from the turntable coordinate system to the carrier system, is the conversion matrix from the carrier system to the navigation system; in, In the formula, is the transformation matrix from the turntable coordinate system to the carrier system, ω is the rotation speed, t is the inertial navigation operation time, t s The transfer stop time; in, In the formula, is the conversion matrix from the carrier system to the navigation system, θ is the pitch angle, γ is the roll angle, and ψ is the heading angle; in, In the formula, S g is the gyro scale error matrix, δG is the gyro installation error matrix, is the actual angular rate input of the gyroscope, ε is the zero bias of the gyroscope, is the random drift of the gyroscope; in, In the formula, S g is the gyro scale error matrix, is the symmetry scale error of the X-gyro, is the asymmetric scaling error of the X-gyro, is the symmetry scale error of the Y gyro, is the asymmetric scaling error of the Y gyro, is the symmetry scale error of the Z gyro, is the asymmetric scaling error of the Z gyro, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; sign(·) is the sign function, expressed as in, Where δG is the gyro installation error matrix, is the installation error of the Y gyro and the Z gyro, is the installation error of Z gyro and Y gyro, is the installation error of Z gyro and X gyro; in, In the formula, is the actual angular rate input of the gyroscope, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; Where ε=[ε x ,ε y ,ε z ] T ; Where ε is the gyro's zero bias, ε x is the zero bias of the X gyro, ε y is the zero bias of the Y gyro, ε z is the zero bias of the Z gyro; in, In the formula, is the random drift of the gyroscope, ω dx is the random drift of the X gyro, ω dy is the random drift of the Y gyro, ω dz is the random drift of the Z gyro.

[0023] In this embodiment, in step S12, the specific method of constructing the speed error equation is: The velocity error equation is constructed as follows: In the formula, is δv n The derivative with respect to time, δv n is the velocity error, v n is the speed, f n is the specific force measured by the accelerometer; φ n =[φ E ,φ N ,φ U ] T ,φ E ,φ N ,φ U They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the accelerometer measurement error, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the rotation angular rate error of the Earth relative to the inertial system, is the rotation angular rate of the navigation system relative to the earth, δg is the gravity acceleration vector error; in, In the formula, is the accelerometer measurement error, S a is the scale factor error matrix of the accelerometer, δA is the installation error matrix of the accelerometer, is the actual acceleration input of the accelerometer, is the zero bias of the accelerometer, is the random drift of the accelerometer; in, In the formula, S a is the scale factor error matrix of the accelerometer, is the symmetry scale error of the X-accelerometer, is the asymmetric scale error of the X accelerometer, is the symmetry scale error of the Y accelerometer, is the asymmetric scaling error of the Y accelerometer, is the symmetry scale error of the Z accelerometer, is the asymmetric scaling error of the Z accelerometer, is the actual acceleration input of the X accelerometer, is the actual acceleration input of the Y accelerometer, is the actual acceleration input of the Z accelerometer; sign(·) is the sign function, expressed as in, Where δA is the installation error matrix of the accelerometer, η xz is the installation error of the X accelerometer and the Z accelerometer, η xy is the installation error of the X accelerometer and the Y accelerometer, η yz is the installation error of the Y accelerometer and the Z accelerometer, η yx is the installation error between the Y accelerometer and the X accelerometer, η zy is the installation error of the Z accelerometer and the Y accelerometer, η zx is the installation error of the Z accelerometer and the X accelerometer; in, In the formula, is the zero bias of the accelerometer, is the zero bias of the X accelerometer, is the zero bias of the Y accelerometer, is the zero bias of the Z accelerometer; in, In the formula, is the random drift of the accelerometer, f dx is the random drift of the X accelerometer, f dy is the random drift of the Y accelerometer, f dz is the random drift of the Z accelerometer.

[0024] In this embodiment, in step S13, the specific method of constructing the position error equation is: The position error equation is constructed as follows: In the formula, is the derivative of δL with respect to time, is the derivative of δλ with respect to time, is the derivative of δh with respect to time, L is latitude, λ is longitude, h is height, R is the average radius of the earth, v N is the north velocity, v E is the eastward velocity, δv N is the north velocity error, δv E is the eastward velocity error, δv U is the celestial velocity error, δψ is the heading error, δL is the latitude error, δλ is the longitude error, and δh is the height error.

[0025] In this embodiment, in step S2, the specific method of setting the indexing order of the dual-axis rotary inertial navigation system is: Set multiple sequences, each sequence rotates 180° forward or reverse around the celestial direction or east direction in the coordinate system at a rotation speed of ω. The directions or rotations of two adjacent sequences are different, with t s The stop time is still; The rotation cycle is repeated with all the set sequences as one rotation cycle.

[0026] In this embodiment, in step S2, the indexing order of the dual-axis rotating inertial navigation is preferably but not limited to being set to: Order 1: Rotate 180° around the celestial direction at a speed of ω, and then s The stop time is still; Order 2: Rotate 180° around the east direction at a speed of ω, and then rotate 180° at t s The stop time is still; Order 3, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 4, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 5, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 6: Rotate 180° around the celestial direction at a speed of ω, and then rotate 180° at t s The stop time is still; Order 7, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 8, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 9, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 10, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 11, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 12, rotate 180° around the east direction with ω as the rotation speed, and s The stop time is still; Order 13, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 14, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 15, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 16, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; The sequence 1 to sequence 16 is a rotation cycle, and the rotation cycle is repeated.

[0027] In this embodiment, in step S2, the rotation speed of the dual-axis rotating inertial navigation system is preferably but not limited to being set to 5° / s, and the rotation and stop time of the dual-axis rotating inertial navigation system is preferably but not limited to being set to 2s.

[0028] In this embodiment, in step S3, the navigation information benchmark includes a benchmark for attitude navigation information, a benchmark for speed navigation information and a benchmark for position navigation information. The benchmark for the attitude navigation information is the installation benchmark plane of the actual prototype. Since the actual prototype of the dual-axis rotating inertial navigation is stationary, the benchmark for the speed navigation information is 0. The benchmark for the position navigation information is the local geographic coordinates.

[0029] In this embodiment, in step S4, the inertial device error parameters include the gyro scale error matrix S g , gyro installation error matrix δG, gyro zero bias ε, gyro random drift The scale factor error matrix S of the accelerometer a , the installation error matrix of the accelerometer δA, the zero bias of the accelerometer Random drift of the accelerometer

[0030] In this embodiment, in step S4, the LSTM algorithm is a long short-term memory network neural network algorithm, which belongs to the prior art and will not be described in detail here.

[0031] The method of the present invention firstly constructs an error propagation equation of a dual-axis rotation inertial navigation system considering the error modulation effect, and then sets the indexing order, rotation speed and rotation-stop time of the dual-axis rotation inertial navigation system, and then obtains the navigation result and the navigation error output through the actual prototype of the dual-axis rotation inertial navigation system, and finally obtains the error parameter of the inertial device in the error propagation equation based on the LSTM algorithm, so as to obtain the digital prototype of the dual-axis rotation inertial navigation system, thereby reducing the time and cost of the dual-axis rotation inertial navigation system error modulation research, and has the advantages of simple operation, less time consumption, low cost and good practicality, and solves the problem that in the current dual-axis rotation inertial navigation system error modulation research, the actual prototype of the dual-axis rotation inertial navigation system is usually used for error modulation research, which is time-consuming and costly.

[0032] Parts of the present invention that are not disclosed in detail belong to the common knowledge in the art.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a dual-axis rotating inertial navigation digital prototype based on LSTM algorithm, characterized in that: The steps include: S1. Construct the error propagation equation of dual-axis rotational inertial navigation considering the error modulation effect; S2. Setting the rotation order, rotation speed and rotation stop time of the dual-axis rotation inertial navigation system; S3, placing the actual prototype of the dual-axis rotation inertial navigation on the table, performing error modulation on it according to the indexing order, rotation speed and rotation stop time set in step S2, and obtaining the navigation result of the actual prototype; comparing the navigation result of the actual prototype with the navigation information reference, and obtaining the navigation error output of the actual prototype; S4. Based on the error propagation equation constructed in step S1, the LSTM algorithm is used to obtain the error parameters of the inertial device in the error propagation equation; the obtained inertial device error parameters, the navigation result and navigation error output obtained in step S3, and the indexing order, rotation speed and rotation-stop time set in step S2 are substituted into the error propagation equation constructed in step S1 to obtain a digital prototype of the dual-axis rotation inertial navigation.

2. According to a method for constructing a dual-axis rotating inertial navigation digital prototype based on LSTM algorithm according to claim 1, it is characterized in that: In step S1, the specific method of constructing the error propagation equation of the dual-axis rotating inertial navigation considering the error modulation effect includes: S11, constructing attitude error equation; S12, constructing a velocity error equation; S13. Construct the position error equation.

3. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 2 is characterized in that: In step S11, the specific method of constructing the posture error equation is: The posture error equation is constructed as follows: In the formula, is φ n The derivative with respect to time, φ n =[φ E ,φ N ,φ U ] T ,φ E ,φ N ,φ U They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the rotation angular rate error of the navigation system relative to the inertial system, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, Indicates the gyro measurement error in the navigation system; in, In the formula, is the angular velocity of the Earth's rotation relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, V N and V E Respectively represent the northward speed and eastward speed of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the rotation angular rate error of the navigation system relative to the inertial system, is the rotation angular rate error of the Earth relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth; in, In the formula, is the rotation angular rate error of the Earth relative to the inertial system, ω ie It represents the projection of the Earth's rotation angular velocity in the navigation system, L represents the local latitude, and δL represents the local latitude error; is the angular velocity of the navigation system relative to the Earth, V N and V E Respectively represent the northward and eastward velocities of the carrier in the geographic coordinate system, δV N and δV E They represent the northward velocity error and eastward velocity error of the carrier in the geographic coordinate system, R m and R n They represent the radius of curvature of the Earth's meridian plane and the radius of curvature of the Meridian plane respectively; in, In the formula, is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the transformation matrix from the turntable coordinate system to the carrier system, is the conversion matrix from the carrier system to the navigation system; in, In the formula, is the transformation matrix from the turntable coordinate system to the carrier system, ω is the rotation speed, t is the inertial navigation operation time, t s The transfer stop time; in, In the formula, is the conversion matrix from the carrier system to the navigation system, θ is the pitch angle, γ is the roll angle, and ψ is the heading angle; in, In the formula, S g is the gyro scale error matrix, δG is the gyro installation error matrix, is the actual angular rate input of the gyroscope, ε is the zero bias of the gyroscope, is the random drift of the gyroscope; in, In the formula, S g is the gyro scale error matrix, is the symmetry scale error of the X-gyro, is the asymmetric scaling error of the X-gyro, is the symmetry scale error of the Y gyro, is the asymmetric scaling error of the Y gyro, is the symmetry scale error of the Z gyro, is the asymmetric scaling error of the Z gyro, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; sign(·) is the sign function, expressed as in, Where δG is the gyro installation error matrix, is the installation error of the Y gyro and the Z gyro, is the installation error of Z gyro and Y gyro, is the installation error of Z gyro and X gyro; in, In the formula, is the actual angular rate input of the gyroscope, is the actual angular rate input of the X gyro, is the actual angular rate input of the Y gyro, is the actual angular rate input of the Z gyro; Where, ε = [ε x ,he y ,he z ] T ; Where ε is the gyro's zero bias, ε x is the zero bias of the X gyro, ε y is the zero bias of the Y gyro, ε z is the zero bias of the Z gyro; where, In the formula, is the random drift of the gyroscope, ω dx is the random drift of the X gyro, ω dy is the random drift of the Y gyro, ω dz is the random drift of the Z gyro.

4. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 2 is characterized in that: In step S12, the specific method of constructing the speed error equation is: The velocity error equation is constructed as follows: In the formula, is δv n The derivative with respect to time, δv n is the velocity error, v n is the speed, f n is the specific force measured by the accelerometer; φ n =[φ E ,φ N ,φ U ] T ,φ E ,φ N ,φ U They are pitch attitude error, roll attitude error, and heading attitude error respectively; is the transformation matrix from the turntable coordinate system to the navigation coordinate system, is the accelerometer measurement error, is the angular velocity of the Earth's rotation relative to the inertial system, is the angular rate of rotation of the navigation system relative to the Earth, is the rotation angular rate error of the Earth relative to the inertial system, is the rotation angular rate of the navigation system relative to the earth, δg is the gravity acceleration vector error; in, In the formula, is the accelerometer measurement error, S a is the scale factor error matrix of the accelerometer, δA is the installation error matrix of the accelerometer, is the actual acceleration input of the accelerometer, is the zero bias of the accelerometer, is the random drift of the accelerometer; in, In the formula, S a is the scale factor error matrix of the accelerometer, is the symmetry scale error of the X-accelerometer, is the asymmetric scaling error of the X accelerometer, is the symmetry scale error of the Y accelerometer, is the asymmetric scaling error of the Y accelerometer, is the symmetry scale error of the Z accelerometer, is the asymmetric scaling error of the Z accelerometer, is the actual acceleration input of the X accelerometer, is the actual acceleration input of the Y accelerometer, is the actual acceleration input of the Z accelerometer; sign(·) is the sign function, expressed as in, Where δA is the installation error matrix of the accelerometer, η xz is the installation error of the X accelerometer and the Z accelerometer, η xy is the installation error of the X accelerometer and the Y accelerometer, η yz is the installation error of the Y accelerometer and the Z accelerometer, η yx is the installation error between the Y accelerometer and the X accelerometer, η zy is the installation error of the Z accelerometer and the Y accelerometer, η zx is the installation error of the Z accelerometer and the X accelerometer; in, In the formula, is the zero bias of the accelerometer, is the zero bias of the X accelerometer, is the zero bias of the Y accelerometer, is the zero bias of the Z accelerometer; in, In the formula, is the random drift of the accelerometer, f dx is the random drift of the X accelerometer, f dy is the random drift of the Y accelerometer, f dz is the random drift of the Z accelerometer.

5. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 2, characterized in that: In step S13, the specific method of constructing the position error equation is: The position error equation is constructed as follows: In the formula, is the derivative of δL with respect to time, is the derivative of δλ with respect to time, is the derivative of δh with respect to time, L is latitude, λ is longitude, h is height, R is the average radius of the earth, v N is the north velocity, v E is the eastward velocity, δv N is the north velocity error, δv E is the eastward velocity error, δv U is the celestial velocity error, δψ is the heading error, δL is the latitude error, δλ is the longitude error, and δh is the height error.

6. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 1, characterized in that: In step S2, the specific method of setting the indexing order of the dual-axis rotary inertial navigation is: Set multiple sequences, each sequence rotates 180° forward or reverse around the celestial direction or east direction in the coordinate system at a rotation speed of ω. The directions or rotations of two adjacent sequences are different, with t s The stop time is still; The rotation cycle is repeated with all the set sequences as one rotation cycle.

7. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 1, characterized in that: In step S2, the indexing order of the dual-axis rotating inertial navigation is set as: Order 1: Rotate 180° around the celestial direction at a speed of ω, and then s The stop time is still; Order 2: Rotate 180° around the east direction at a speed of ω, and then rotate 180° at t s The stop time is still; Order 3, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 4, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 5, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 6: Rotate 180° around the celestial direction at a speed of ω, and then rotate 180° at t s The stop time is still; Order 7, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 8, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 9, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; Order 10, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 11, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 12, rotate 180° around the east direction with ω as the rotation speed, and s The stop time is still; Order 13, rotate 180° around the east direction with ω as the rotation speed, and t s The stop time is still; Order 14, rotate 180° around the celestial direction at a speed of ω, and s The stop time is still; Order 15, with the rotation speed ω, reverse 180° around the east direction, with t s The stop time is still; Order 16, with the rotation speed ω, reverse 180° around the celestial direction, with t s The stop time is still; The sequence 1 to sequence 16 is a rotation cycle, and the rotation cycle is repeated.

8. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 1, characterized in that: In step S2, the rotation speed of the dual-axis rotating inertial navigation system is set to 5° / s, and the rotation and stop time of the dual-axis rotating inertial navigation system is set to 2s.

9. The method for constructing a dual-axis rotating inertial navigation digital prototype based on the LSTM algorithm according to claim 1, characterized in that: In step S3, the navigation information reference includes a reference for attitude navigation information, a reference for speed navigation information and a reference for position navigation information. The reference for attitude navigation information is the installation reference plane of the actual prototype, the reference for speed navigation information is 0, and the reference for position navigation information is the local geographic coordinates.

10. The method for constructing a dual-axis rotating inertial navigation digital prototype based on LSTM algorithm according to claim 1, characterized in that: In step S4, the inertial device error parameters include the gyro scale error matrix S g , gyro installation error matrix δG, gyro zero bias ε, gyro random drift The scale factor error matrix S of the accelerometer a , the installation error matrix of the accelerometer δA, the zero bias of the accelerometer Random drift of the accelerometer

Images