An initial rough alignment method for an inertial navigation system on a moving base on water
By using gravity visual motion initial alignment model and CEEMD signal decomposition technology in a ship moving base environment, the problem of limited initial alignment accuracy and speed in traditional methods is solved, and a higher accuracy and rapid initialization of inertial navigation system is achieved. alignment.
Patent Information
- Application Number
- CN202311477529.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-07
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-11-07
AI Technical Summary
In the dynamic base environment of ships, traditional methods are difficult to achieve high-precision initial alignment of the initial inertial navigation system, which is affected by changes in the marine environment and interference from sensor noise, resulting in limited alignment accuracy and speed.
The initial alignment model of gravity visual motion combined with the adaptive signal decomposition method is used to measure the gravity vector and the earth's rotation angular rate through the inertial navigation system accelerometer, and the initial alignment attitude matrix is constructed, and the signal decomposition and core density estimator are used to perform signal decomposition. Noise, realizes signal reconstruction and initial alignment.
The initial alignment accuracy and speed in a moving base environment is improved, and the signal mode aliasing and endpoint effects in traditional methods are overcome, achieving more accurate and fast initial alignment of the inertial navigation system.
Smart Images

Figure CN117553828B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to ship information technology, and in particular to an initial coarse alignment method for an inertial navigation system of an above-water dynamic base. Background Art
[0002] The strapdown inertial navigation system is widely used in the measurement of the ship's own position, speed, attitude and other state information due to its independence and accuracy. Without the assistance of other sensors, the self-alignment technology that uses the data of the inertial measurement unit to accurately and quickly obtain the initial attitude information of the carrier directly affects the startup response speed and accuracy of other systems of the ship. However, due to the changing ocean environment and the interference of sensor measurement noise, it is still very difficult to achieve high-precision initial alignment accuracy in a short time.
[0003] In the initial alignment, the accuracy and speed of the coarse alignment directly determine the convergence time and accuracy of the fine alignment. Due to the influence of factors such as the irregular swaying of the ship under wind and waves, the traditional static base and quasi-static base methods cannot be applied to the initial alignment of the ship's dynamic base. The coarse alignment method of gravity apparent motion is proposed to convert the attitude matrix problem between the carrier coordinate system and the navigation coordinate system in the alignment into the matrix problem between the carrier inertial coordinate system and the navigation inertial coordinate system. However, in practical applications, since the accelerometer data is directly used to calculate the gravity apparent motion, some non-negligible sensor errors are brought into the alignment process, resulting in a hysteresis in the convergence speed and a reduction in alignment accuracy.
[0004] At present, the problem of INS coarse alignment noise reduction is usually solved by using the velocity obtained by integrating the gravity acceleration for initial alignment, but this method also introduces linear velocity errors. Low-pass filtering or wavelet noise reduction methods require frequency prior information of the measurement signal, but it is very difficult to pre-measure the frequency of the information in the water environment and in the case of multiple INS. Summary of the invention
[0005] The technical problem to be solved by the present invention is to provide an initial coarse alignment method for an inertial navigation system of an above-water moving base in view of the defects in the prior art.
[0006] The technical solution adopted by the present invention to solve the technical problem is: an initial rough alignment method of an inertial navigation system of an aquatic moving base, comprising the following steps:
[0007] 1) According to the accelerometer of the inertial navigation system under the dynamic base, the trajectory of the gravity vector rotating with the earth at a fixed point can be measured, and the initial alignment model of the gravity apparent motion under the dynamic base can be constructed;
[0008] According to the accelerometer of the inertial navigation system under the dynamic base, the trajectory of the gravity vector rotating with the earth at a fixed point can be measured, and the initial alignment model of the apparent gravity motion under the dynamic base can be constructed, specifically:
[0009] 1.1) Select the "east-north-sky" geographic coordinate system as the system navigation coordinate system n; select the coordinate system with the center of mass of the carrier as the origin and the directions of "right-front-up" as the carrier coordinate system b; define the attitude matrix calculated by the initial alignment of the inertial navigation system This breaks down into:
[0010]
[0011] Among them, i b0 represents the "right-front-up" initial carrier inertial coordinate system at time t0, is the attitude matrix between the initial carrier coordinate system and the current carrier coordinate system, is the attitude matrix between the inertial navigation system coordinate system and the initial carrier coordinate system,
[0012] in, It is expressed as:
[0013]
[0014] in, is the angular rate measured by the gyro of the inertial navigation system; X represents the antisymmetric matrix operation;
[0015] 1.2) Select the same point on the Earth at time t A ,t B Non-collinear gravity vector g in inertial coordinate system A ,g B , after the Earth's periodic rotation at time t A ,t B The geographic vectors form a cone to construct the gravity apparent motion model;
[0016] in,
[0017] The theoretical measurement of the accelerometer is the component of the gravity vector in the carrier coordinate system, expressed as:
[0018] f b (t) = -g b (t) (3)
[0019] Among them, f b (t) is the theoretical measurement of the accelerometer,
[0020] Actual accelerometer measurement Theoretical measurements and measurement errors are included;
[0021] calculate In the initial carrier inertial coordinate system i b0 The projection below is:
[0022]
[0023] 1.3) Based on the geometric relationship constructed in step 1.2), calculate the midpoint O of AB AB The unit east vector E AB for:
[0024]
[0025] Among them, AB is the same point at time t A ,t B The geometric line segments formed by
[0026] According to formula (5), E AB , the "east-north-sky" vector E, N, U of the navigation coordinate system at B is calculated as:
[0027]
[0028]
[0029] N(t B )=U(t B )×E(t B ) (6)
[0030] Among them, ω ie is the Earth's rotation angular velocity; further construction yields for:
[0031]
[0032] The formula (7) is obtained Substituting into formula (1), we can calculate the initial alignment attitude matrix of the inertial navigation system:
[0033] 2) Adaptively decompose the inertial navigation system measurement signal into intrinsic mode functions (IMF) containing different frequency spectra;
[0034] 2.1) The actual measured signal containing the original signal x and the additional noise w is expressed as:
[0035]
[0036] where y 1 ,y 2 represent the signals with added positive and negative noise, respectively;
[0037] 2.2) Use the empirical modulus decomposition method to completely decompose y 1 ,y 2 , respectively, to obtain IMF i (+) and IMF i (-),i=1,...,I; where i represents the number of intrinsic mode IMFs;
[0038] 2.3) By averaging the relevant modal IMF i =(IMF i (+)+IMF i (-)) / 2, calculate the i-th mode of the original signal x;
[0039] 2.4) Get multiple IMFs and residual signals r of the original signal x after CEEMD decomposition I for:
[0040] 3) Calculate and analyze the similarity between the decomposed signal and the original signal to reconstruct the signal and complete the noise reduction of the inertial navigation system signal;
[0041] 3.1) Calculate the original signal x(t) and each IMF separately through the kernel density estimator i The probability density function P and p i , calculate p i l between P 2 The norm similarity is:
[0042]
[0043]
[0044] Where dist represents l in formula (9) 2 Norm similarity function, D(i) represents the current IMF i Similarity with the original signal x(t);
[0045] 3.2) Calculate the IMF order k required for denoised signal reconstruction th , based on the calculated similarity D(i), k th Determined as:
[0046]
[0047] Among them, first represents the order at which the similarity of the current IMF first decreases compared with the similarity of the previous IMF, l represents the order of the noise term in the original signal, and all IMFs before the lth order need to be deleted in the reconstructed signal:
[0048] l = k th -1 (12)
[0049] 3.3) According to the order k of signal reconstruction th IMF and residual signal after CEEMD decomposition, signal after filtering and noise reduction It is expressed as:
[0050]
[0051] 4) The noise-reduced inertial navigation signal obtained in step 3) is introduced into the gravity visual alignment model constructed in step 1) to achieve initial alignment.
[0052] First, use the inertial navigation system gyro to output the current angular velocity Calculate and update the attitude matrix between the initial carrier coordinate system and the current carrier coordinate system At the same time, the acceleration information output by the accelerometer of the inertial navigation system gyro is obtained after noise reduction filtering Then, After updating Projected to the initial carrier inertial coordinate system i b0 get And calculate through the gravity motion alignment model constructed in step 1 Finally, using and Get the initial posture matrix at the current moment Iterate until the alignment time ends or After multiple iterations, the initial alignment is completed.
[0053] The beneficial effects produced by the present invention are:
[0054] 1. The present invention adopts a gravity apparent motion initial alignment strategy in the initial alignment of the moving base. By introducing the carrier inertial coordinate system, the problem that the general initial alignment method cannot be sensitive to the earth's rotation angle rate in the case of a moving base, and thus cannot calculate the initial attitude matrix, is overcome.
[0055] 2. The present invention uses a data-driven CEEMD signal decomposition method, which can not only decompose the signal without prior information of the signal, but also solves the modal aliasing and endpoint effects of general data decomposition methods, and overcomes the problems of single technical application conditions and environment.
[0056] 3. The present invention uses similarity to reconstruct signals, thereby improving the adaptability and integrity of signal reconstruction and achieving more accurate and rapid initial alignment of the inertial navigation system. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:
[0058] Figure 1is a method flow chart of an embodiment of the present invention;
[0059] Figure 2 is a schematic diagram of an initial alignment coordinate system of an inertial navigation system according to an embodiment of the present invention;
[0060] Figure 3 is a schematic diagram of an initial alignment model of gravity apparent motion of an inertial navigation system according to an embodiment of the present invention;
[0061] Figure 4 is a schematic diagram of an example of an initial alignment model of a ship inertial navigation system according to an embodiment of the present invention;
[0062] Figure 5 This is a diagram showing the noise reduction effect of the accelerometer signal in the Z-axis direction according to an embodiment of the present invention;
[0063] Figure 6 This is a diagram showing the initial alignment effect of a ship mooring dynamic base according to an embodiment of the present invention. DETAILED DESCRIPTION
[0064] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0065] like Figure 1 As shown, an initial rough alignment method for an inertial navigation system of an above-water moving base comprises the following steps:
[0066] 1) According to the accelerometer of the inertial navigation system under the dynamic base, the trajectory of the gravity vector rotating with the earth at a fixed point can be measured, and the initial alignment model of the gravity apparent motion under the dynamic base can be constructed;
[0067] Figure 2 It is the initial alignment coordinate system of the present invention. The coordinate system and its parameters used in the present invention are defined as follows:
[0068] The inertial coordinate system i selects the geocentric inertial coordinate system, that is, the origin of the coordinate is the center of mass of the earth, the axis of rotation of the earth is the Z axis, the X axis points to the vernal equinox in the equatorial plane, and the Y axis forms a right-handed orthogonal coordinate system with the X and Z axes;
[0069] The earth coordinate system e takes the center of mass of the earth as the coordinate origin, the Z axis is the earth's rotation axis, its X axis points to the prime meridian in the equatorial plane, and the Y axis forms a right-handed orthogonal coordinate system with the X and Z axes. The angular motion of the earth coordinate system relative to the inertial coordinate system is the angular velocity of the earth's rotation ω ie ;
[0070] The navigation coordinate system n selects the center of gravity of the carrier as the coordinate origin, and the X, Y and Z axes are east, north and sky directions respectively based on the ground level;
[0071] The carrier coordinate system b selects the carrier's center of gravity as the coordinate origin, and the X, Y and Z axes point to the right, front and top directions of the carrier respectively;
[0072] Initial carrier inertial coordinate system i b0 The carrier coordinate system at time t=0 during the gravity alignment process is selected.
[0073] According to the accelerometer of the inertial navigation system under the dynamic base, the trajectory of the gravity vector rotating with the earth at a fixed point can be measured, and the initial alignment model of the apparent gravity motion under the dynamic base can be constructed, specifically:
[0074] 1.1) Select the "east-north-sky" geographic coordinate system as the system navigation coordinate system n; select the coordinate system with the center of mass of the carrier as the origin and the directions of "right-front-up" as the carrier coordinate system b; define the attitude matrix calculated by the initial alignment of the inertial navigation system This breaks down into:
[0075]
[0076] Among them, i b0 represents the "right-front-up" initial carrier inertial coordinate system at time t0, is the attitude matrix between the initial carrier coordinate system and the current carrier coordinate system, is the attitude matrix between the inertial navigation system coordinate system and the initial carrier coordinate system,
[0077] in, It is expressed as:
[0078]
[0079] in, is the angular rate measured by the gyro of the inertial navigation system; X represents the antisymmetric matrix operation;
[0080] 1.2) Select Figure 3 The same point on the Earth at different times t A ,t B , the non-collinear gravity vector g in its inertial coordinate system A ,g B , respectively, obtained by measuring the specific force through IMU and After the Earth's periodic rotation, at time t A ,t B The geographic vectors form a cone to construct the gravity apparent motion model; Figure 3 It is the gravity visual alignment model of the moving base on the water surface;
[0081] in,
[0082] The theoretical measurement of the accelerometer is the component of the gravity vector in the carrier coordinate system, expressed as:
[0083] f b (t) = -g b (t) (3)
[0084] Among them, f b (t) is the theoretical measurement of the accelerometer,
[0085] Actual accelerometer measurement Theoretical measurements and measurement errors are included;
[0086] calculate In the initial carrier inertial coordinate system i b0 The projection below is:
[0087]
[0088] 1.3) Based on the geometric relationship constructed in step 1.2), calculate the midpoint O of AB AB The unit east vector E AB for:
[0089]
[0090] Among them, AB is the same point at time t A ,t B The geometric line segments formed by
[0091] According to formula (5), E AB , the "east-north-sky" vector E, N, U of the navigation coordinate system at B is calculated as:
[0092]
[0093]
[0094] N(t B )=U(t B )×E(t B ) (6)
[0095] Among them, ω ie is the Earth's rotation angular velocity; further construction yields for:
[0096]
[0097] The formula (7) is obtained Substituting into formula (1), we can calculate the initial alignment attitude matrix of the inertial navigation system:
[0098] 2) Adaptively decompose the inertial navigation system measurement signal into intrinsic mode functions (IMF) containing different frequency spectra;
[0099] 2.1) The actual measured signal including the original signal x and the additional noise w is expressed as:
[0100]
[0101] where y 1 ,y 2 represent the signals with added positive and negative noise, respectively;
[0102] 2.2) Use the empirical modulus decomposition method to completely decompose y 1 ,y 2 , respectively, to obtain IMF i (+) and IMF i (-),i=1,...,I; where i represents the number of intrinsic mode IMFs;
[0103] 2.3) By averaging the relevant modal IMF i =(IMF i (+)+IMF i (-)) / 2, calculate the i-th mode of the original signal x;
[0104] 2.4) Get multiple IMFs and residual signals r of the original signal x after CEEMD decomposition I for:
[0105] 3) Calculate and analyze the similarity between the decomposed signal and the original signal to reconstruct the signal and complete the noise reduction of the inertial navigation system signal;
[0106] 3.1) Calculate the original signal x(t) and each IMF separately through the kernel density estimator i The probability density function P and p i , calculate p i l between P 2 The norm similarity is:
[0107]
[0108]
[0109] Where dist represents l in formula (9) 2 Norm similarity function, D(i) represents the current IMF i Similarity with the original signal x(t);
[0110] 3.2) Calculate the IMF order k required for denoised signal reconstruction th , based on the calculated similarity D(i), k th Determined as:
[0111]
[0112] Among them, first represents the order at which the similarity of the current IMF first decreases compared with the similarity of the previous IMF, l represents the order of the noise term in the original signal, and all IMFs before the lth order need to be deleted in the reconstructed signal:
[0113] l = k th -1 (12)
[0114] 3.3) According to the order k of signal reconstruction th IMF and residual signal after CEEMD decomposition, signal after filtering and noise reduction
[0115] It is expressed as:
[0116]
[0117] 4) The noise-reduced inertial navigation signal obtained in step 3) is introduced into the gravity visual alignment model constructed in step 1) to achieve initial alignment.
[0118] First, use the inertial navigation system gyro to output the current angular velocity Calculate and update the attitude matrix between the initial carrier coordinate system and the current carrier coordinate system At the same time, the acceleration information output by the accelerometer of the inertial navigation system gyro is obtained after noise reduction filtering Then, After updating Projected to the initial carrier inertial coordinate system i b0 get And calculate through the gravity motion alignment model constructed in step 1 Finally, using and Get the initial posture matrix at the current moment Iterate until the alignment time ends or After multiple iterations, the initial alignment is completed.
[0119] Comparison of actual results:
[0120] like Figure 4In the initial alignment example of ship mooring inertial navigation, the attitude as a reference benchmark is generated by combining GPS and a reference strapdown inertial navigation system; another device with the same parameters as the reference strapdown inertial navigation system is used for initial self-alignment, and the two devices are fixed adjacently in the cabin.
[0121] like Figure 5 In the example shown, during the 250s alignment process, the blue signal is the actual acceleration measurement output by the Z axis of the strapdown inertial navigation system carrier, and the red signal is the acceleration signal processed by the noise reduction filter of the present invention. It can be found that the noise of the original signal can be significantly reduced.
[0122] like Figure 6 The initial alignment process effect diagram is shown in the figure, δθ and δγ are the pitch angle error and roll angle error respectively, is the heading angle error, and the blue and purple represent the initial alignment accuracy of the low-pass filtering method and the velocity vector, respectively, and the red and green represent the initial alignment effects of the filtering method based on the common empirical modulus decomposition and the improved filtering method of the present invention, respectively. It can be seen that in the alignment process of 250s, the alignment accuracy and speed of the present invention are both optimal.
[0123] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.
Claims
1. A method for initial coarse alignment of an inertial navigation system on a moving base above water, characterized in that it comprises the following steps: 1) According to the accelerometer of the inertial navigation system under the dynamic base, the trajectory of the gravity vector rotating with the earth at a fixed point can be measured, and the initial alignment model of the gravity apparent motion under the dynamic base can be constructed; Construct the initial alignment model of gravity apparent motion under the dynamic base, specifically: 1.1) Select the "east-north-sky" geographic coordinate system as the system navigation coordinate system n; select the coordinate system with the center of mass of the carrier as the origin and the directions of "right-front-up" as the carrier coordinate system b; define the attitude matrix calculated by the initial alignment of the inertial navigation system This breaks down into: Among them, i b0 represents the "right-front-up" initial carrier inertial coordinate system at time t0, is the attitude matrix between the initial carrier coordinate system and the current carrier coordinate system, is the attitude matrix between the inertial navigation system coordinate system and the initial carrier coordinate system, in, It is expressed as: in, is the angular rate measured by the gyro of the inertial navigation system; × represents antisymmetric matrix operation; 1.2) Select the same point on the Earth at time t A ,t B Non-collinear gravity vector g in inertial coordinate system A ,g B , after the Earth's periodic rotation at time t A ,t B The geographic vectors form a cone to construct the gravity apparent motion model; in, The theoretical measurement of the accelerometer is the component of the gravity vector in the carrier coordinate system, expressed as: f b (t) = -g b (t) Among them, f b (t) is the theoretical measurement of the accelerometer; Actual accelerometer measurement Theoretical measurements and measurement errors are included; calculate In the initial carrier inertial coordinate system i b0 The projection below for: 1.3) Based on the geometric relationship constructed in step 1.2), calculate the midpoint O of AB AB The unit east vector E AB for: in, AB is the same point at time t A ,t B The geometric line segments formed by According to the obtained E AB , the "east-north-sky" vector E, N, U of the navigation coordinate system at B is calculated as: N(t B )=U(t B )×E(t B ) Among them, ω ie is the Earth's rotation angular velocity; further construction yields for: Will get Substitute into the formula The calculated initial alignment attitude matrix of the inertial navigation system is 2) Adaptively decompose the inertial navigation system measurement signal into intrinsic mode functions containing different frequency spectra; 2.1) The actual measured signal including the original signal x and the additional noise w is expressed as: Where y1 and y2 represent the signals with added positive and negative noise respectively; 2.2) Use the empirical modulus decomposition method to completely decompose y1 and y2 and obtain IMF respectively i (+) and IMF i (-),i=1,...,I; where i represents the number of intrinsic mode IMFs; 2.3) By averaging the relevant modal IMF i =(IMF i (+)+IMF i (-)) / 2, calculate the i-th mode of the original signal x; 2.4) Get multiple IMFs and residual signals r of the original signal x(t) at time t after CEEMD decomposition I for: 3) Calculate and analyze the similarity between the decomposed signal and the original signal to reconstruct the signal and complete the noise reduction of the inertial navigation system signal; 4) The noise-reduced inertial navigation signal obtained in step 3) is introduced into the gravity visual alignment model constructed in step 1) to achieve initial alignment.
2. The initial coarse alignment method of an underwater dynamic base inertial navigation system according to claim 1, characterized in that: In step 3), the similarity between the decomposed signal and the original signal is calculated and analyzed to reconstruct the signal and complete the noise reduction of the inertial navigation system signal, as follows: 3.1) The probability density function P of the original signal x(t) at time t and each IMF are calculated separately by the kernel density estimator i The probability density function p i , calculate the probability density function p i The l2-norm similarity between and P is: Where dist represents the l2 norm similarity function, and D(i) represents the current IMF i Similarity with the original signal x(t) at time t; 3.2) Calculate the IMF order k required for denoised signal reconstruction th , based on the calculated similarity D(i), k th Determined as: Among them, first represents the order at which the similarity of the current IMF first decreases compared with the similarity of the previous IMF, l represents the order of the noise term in the original signal, and all IMFs before the lth order need to be deleted in the reconstructed signal: l=k th -1 3.3) According to the order k of signal reconstruction th IMF and residual signal after CEEMD decomposition, signal after filtering and noise reduction It is expressed as:
Citation Information
Patent Citations
The initial alignment method for sway base
AU2020101268A4
Full-digital simulation method and device for global positioning system (GPS) / strapdown inertial navigation system (SINS) combined navigation
CN102508954A
Initial alignment method of speedometer-assisted strapdown inertial navigation system during running
CN103743414A
Off-line wavelet denoising fast initial alignment method
CN107270937A
Blind separation algorithm-based strapdown inertial navigation signal extraction method and system
CN109443393A
