A Three-Dimensional Real-Time Attitude Determination Method for a Polarization Compass Based on the Polarization Meridian Plane
Through the polarization meridian method, the real-time attitude information of the polarization compass is solved using the solar polarization field imaging mode, and the problem that the polarization compass autonomous posture technology needs to rely on external horizontal reference instruments is solved, achieving high-precision and adaptable three-dimensional real-time posture.
Patent Information
- Application Number
- CN202210620572.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-02
- Publication Date
- 2025-06-03
- Estimated Expiration
- 2042-06-02
AI Technical Summary
The existing polarization compass autonomous posture technology needs to rely on external horizontal reference instruments, resulting in limited navigation accuracy and adaptability, especially in high-speed environments.
Using a polarization meridian method, the solar polarization neutral point and solar vector are extracted through the solar polarization field imaging mode, a polarization meridian model is established, and the real-time attitude information of the polarization compass is solved through this model to achieve three-dimensional real-time pose.
No need to rely on external horizontal reference instruments, improves navigation accuracy and adaptability of polarization autonomous posture, and is suitable for different environments, including high-speed states.
Smart Images

Figure CN115164889B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of three-dimensional real-time attitude determination methods of polarized compasses, and particularly relates to a three-dimensional real-time attitude determination method of a polarized compass based on a polarized meridian plane. Background Art
[0002] As an autonomous navigation instrument applied in the Earth's atmosphere, the three-dimensional real-time attitude determination technology of polarized compasses has always been a key technology that countries around the world have been continuously researching, exploring, and seeking breakthroughs. This technology can solve the problem of real-time acquisition of high-precision carrier three-dimensional attitude information in the case of satellite navigation rejection or interference.
[0003] As analyzed in the paper "Limitation of Rayleigh sky model for bio-inspired polarized skylight navigation in three-dimensional attitude determination", it is impossible to achieve three-dimensional real-time attitude determination by polarized navigation when only considering a single solar vector. In order to achieve real-time attitude determination of a polarized compass, existing autonomous navigation methods of polarized compasses mostly rely on other devices such as horizon sensors, magnetometers, gravity pendulums, inertial navigation systems, or inclinometers to provide horizontal reference measurement information. The patent number CN110231025A, "A dynamic orientation method and system based on a strapdown polarized light compass", uses a strapdown inertial navigation to obtain the pitch angle and roll angle of the polarized compass, and then obtains the carrier heading angle by solving the polarized meridian. The patent number CN108917749A, "A method for extracting the solar meridian based on polarized imaging", obtains the meridian in the polarized compass imaging by fitting and estimating the polarized meridian under the horizontal placement condition, and then the heading angle can be calculated. The article "Integrated Polarized Skylight Sensor and MIMU With a Metric Map for Urban Ground Navigation" obtains the heading attitude of the polarized compass with the help of a micro inertial measurement unit (MIMU).
[0004] However, the introduction of external horizontal reference instruments has the following impacts on the polarized compass measurement system. First, the horizontal measurement system error of external horizontal measurement instruments (such as inclinometers, inertial navigation, etc.) becomes an important part determining the accuracy of polarized navigation; second, the installation and calibration error between the horizontal reference instrument and the polarized system is directly coupled in the polarized navigation system; finally, most horizontal reference instruments are based on accelerometers and are not suitable for high-speed environments. Ultimately, the navigation accuracy and adaptability of the polarized navigation system are limited.
[0005] In summary, it is urgent to break through the limitations of external horizontal reference instruments on the autonomous navigation of the polarization compass and realize the three-dimensional real-time autonomous attitude determination of the polarization compass that does not rely on the external horizontal reference. Summary of the invention
[0006] In order to solve the above problems and overcome the difficulty that the existing polarization compass autonomous attitude determination technology needs to rely on external instruments to provide horizontal reference information, the present invention provides a three-dimensional real-time attitude determination method for a polarization compass based on the polarization meridian plane, which can achieve high-precision polarization autonomous attitude determination without relying on external horizontal reference instruments.
[0007] The present invention adopts the following technical solution:
[0008] A three-dimensional real-time attitude determination method of a polarization compass based on a polarization meridian plane, characterized in that it comprises the following steps:
[0009] The first step is to extract the solar polarization neutral point and solar vector based on the solar polarization field imaging mode, and establish a polarization meridian plane model;
[0010] The second step is to establish a measurement model of the polarization meridian plane based on the conversion relationship between the geographic coordinate system and the polarization coordinate system.
[0011] The third step is to calculate the real-time attitude information of the polarization compass, namely the heading angle, pitch angle and roll angle, based on the polarization meridian plane measurement model and the solar vector coordinate system conversion model.
[0012] As a preferred technical solution of the present invention, in the first step, based on the solar polarization field imaging mode, the solar polarization neutral point and the solar vector are extracted respectively, and a polarization meridian plane model is established, and the implementation steps are as follows:
[0013] (a1) Sunlight interacts with atmospheric molecules or aerosol particles in the Earth's atmosphere and scatters to generate polarization. Polarization compasses are used to analyze the atmospheric polarization field and obtain the polarization degree and polarization vector corresponding to different observation directions at the time of observation.
[0014] (a2) Based on the Stokes vector formula, the polarization degree and polarization direction corresponding to different coordinates in the polarization imaging are calculated, and then the polarization degree image and polarization direction image of the polarization compass analysis are obtained;
[0015] (a3) Based on the polarization image, the polarization threshold segmentation method is used to obtain the neutral point area (minimum polarization area), and then the polarization neutral point coordinate Ne (x NE ,y NE ) and the neutral point vector Ne in the polarized compass coordinate system;
[0016] (a4) Based on the polarization direction image, according to the principle that the polarization vector is perpendicular to the solar vector, establish an optimal solution model for the solar vector to obtain the optimal solar vector (S B ) = [x b , y b , z b = argmin(S B T EE T S B ), where S B is the solar vector in the polarization compass coordinate system, E is the polarization E vector in the polarization compass coordinate system, S B T and E T represent the transpose of the vector;
[0017] (a5) Based on the neutral point vector Ne and the solar vector S B , establish the polarization meridian plane in the polarization compass coordinate system corresponding to this observation moment, that is, Y
[0018] As a preferred technical solution of the present invention, in the second step, based on the conversion relationship between the polarization meridian plane in the geographical coordinate system and the polarization coordinate system, establish a measurement model of the polarization meridian plane, and the implementation steps are as follows:
[0019] (b1) Based on the observation moment t and the observation position information, calculate the reference solar vector S N of the solar vector in the east-north-up geographical coordinate system from the solar ephemeris, that is, the altitude angle h N of the solar vector S S and the azimuth angle
[0020] (b2) In the N-system of the geographical coordinate system, the polarization meridian plane is expressed as
[0021] (b3) Based on the polarization meridian plane of the B-system in the polarization coordinate system obtained in the first step, establish its conversion models between the N-system and the B-system respectively: where R 2 , R 1 and R 3 are the rotation matrices around the Y-axis, X-axis and Z-axis respectively, the pitch angle θ X , the roll angle θ Y and the heading angle φ, is the conversion matrix between the polarization compass coordinate system B-system and the geographical coordinate system N-system.
[0022] As a preferred technical solution of the present invention, in the third step, based on the polarization meridian plane measurement model, the solar vector coordinate system conversion model is combined to solve the real-time attitude information of the polarization compass, that is, the pitch angle θ X , roll angle θ Y and heading angle φ, the implementation steps are as follows:
[0023] (c1) Based on the reference sun vector S obtained in the second step N =[x n ,y n ,z n ] and the solar vector S observed in the first step B =[x b ,y b ,z b ], establish the standard system conversion model:
[0024] (c2) Based on the measurement model of the polarization meridian plane obtained in the second step, the solar vector coordinate system conversion model is constructed:
[0025]
[0026] (c3) Solve the above equation to obtain the three-axis attitude angle of the polarization compass as the pitch angle θ X , roll angle θ Y and the heading angle φ.
[0027] The advantages of the present invention compared with the prior art are:
[0028] (1) The main disadvantage of the existing technology is that it relies on external instruments to provide horizontal reference information, which limits the navigation accuracy and adaptability of polarization autonomous attitude determination. First, the horizontal measurement system error of external horizontal measurement instruments (such as inclinometers and inertial navigation) becomes an important component that determines the accuracy of polarization navigation; second, the installation calibration error of the horizontal reference instrument and the polarization system is directly coupled in the polarization navigation system; finally, the horizontal reference instrument is mostly based on accelerometers, which is not suitable for high-speed environments.
[0029] (2) Compared with the shortcomings of the prior art, the present invention solves the problem that polarization navigation cannot achieve three-dimensional real-time attitude determination when only considering a single sun vector, and its implementation does not rely on external instruments to provide horizontal reference information;
[0030] (3) Compared with the shortcomings of the prior art, the autonomous orientation of the present invention overcomes the requirement of horizontal placement for the symmetric characteristics of the orientation method based on the polarization meridian, and is also adaptable in an inclined state;
[0031] (4) All the above advantages of the present invention are due to the fact that the present invention proposes to characterize the propagation process of incident polarized light by the polarization meridian plane, and expand the two-dimensional characteristic solar vector into a three-dimensional characteristic polarization meridian plane, so as to realize the autonomous attitude determination of the polarization compass without relying on external horizontal reference instruments. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a design flow chart of a three-dimensional real-time attitude determination method for a polarization compass based on the polarization meridian plane of the present invention;
[0033] Figure 2 It is the polarization degree image of the polarization compass of the present invention;
[0034] Figure 3 It is the extraction result of the neutral point area after K-means clustering and polarization degree centroid method of the present invention. DETAILED DESCRIPTION OF THE INVENTION
[0035] The present invention will be described in detail below with reference to the drawings and embodiments.
[0036] As Figure 1 、 Figure 2 、 Figure 3 shown, the present invention discloses a three-dimensional real-time attitude determination method for a polarization compass based on the polarization meridian plane. This method characterizes the propagation process of incident polarized light by the polarization meridian plane, and expands the two-dimensional characteristic solar vector into a three-dimensional characteristic polarization meridian plane.
[0037] Sunlight is scattered by interacting with atmospheric molecules or aerosol particles in the Earth's atmosphere to generate polarization. The polarization compass is used to image and analyze the atmospheric polarization field to obtain the degree of polarization DOP and the polarization direction AOP corresponding to different observation directions at the observation moment:
[0038]
[0039] In the formula, S 0 、S 1 and S 2 are known polarization analysis directions, and are calculated from the imaging gray values I(0°), I(45°), I(90°) and I(135°) corresponding to 0°, 45°, 90° and 135° respectively:
[0040]
[0041] S 1 = I(0°) - I(90°)
[0042] S 2 = I(45°) - I(135°)
[0043] Every four polarization direction pixels form a degree of polarization pixel. Therefore, for a polarization image with a resolution of N×N, its degree of polarization image is DOP = D(N / 2, N / 2). Then, the neutral point is obtained based on the degree of polarization.
[0044] First, set the degree of polarization threshold as DOP TH , so as to segment the degree of polarization image and obtain the neutral point area:
[0045]
[0046] Then, use the K-means clustering method to classify the degree of polarization dot map after threshold segmentation. For example, the degree of polarization binary map is segmented into three parts. Then, select the two clustering results with the largest number as the candidates for the neutral point area, and the result is expressed as N i (x, y). Where i is the number of clustering areas, and (x, y) are the pixel coordinates.
[0047] Calculate the Euclidean distance between the center of the clustering area and each pixel, and perform a secondary screening with 2 / 3 of the maximum Euclidean distance:
[0048]
[0049] In the formula, (x c , y c ) is the clustering center of the i-th neutral point clustering area.
[0050] Use the degree of polarization centroid method to calculate the polarization neutral point coordinates Ne(x NE , y NE ) from the neutral point area:
[0051]
[0052] In the formula, D(x, y) is the degree of polarization of the pixel with coordinates (x, y) in the neutral point area.
[0053] In the polarization compass imaging coordinate system, the polarization neutral point vector Ne:
[0054] Ne = [x NE - x 0 , y NE - y 0 , -f]
[0055] In the formula, (x 0 , y 0 ) is the principal point of the polarization compass imaging, and f is the focal length.
[0056] At the same time, from the polarization direction AOP, the set E of E-vectors E in different observation directions in the polarization compass coordinate system can be obtained, E = [e 1 , e2 ,...,e m . Based on the principle that the polarization vector is perpendicular to the solar vector, an optimal solution model for the solar vector is established, and the optimal solar vector S in the polarization compass coordinate system is obtained B :
[0057] (S B ) = [x b , y b , z b = argmin(S B T EE T S B )
[0058] In the formula, S B is the solar vector in the polarization compass coordinate system, E is the polarization E vector in the polarization compass coordinate system, S B T and E T represent the transpose of the vector.
[0059] In the B-frame of the polarization coordinate system, based on the neutral point vector Ne and the solar vector S B , the corresponding polarization meridian plane at the observation time is established, that is
[0060]
[0061] In the N-frame of the geographic coordinate system, the polarization meridian plane is composed of the solar vector and the zenith direction, that is
[0062]
[0063] In the formula, S N is the direction vector of the solar vector in the N-frame.
[0064] Based on the observation time t and the longitude and latitude of the observation position information, S N is calculated from the solar ephemeris:
[0065]
[0066] where, h S and are the altitude angle and azimuth angle of the solar vector S N , respectively.
[0067] From the N-frame to the B-frame, the coordinate transformation of the polarization meridian plane can be expressed as:
[0068]
[0069] In the formula, R 2 , R 1 and R3 Rotation matrices about the Y-axis, X-axis, and Z-axis respectively, pitch angle θ X , roll angle θ Y and heading angle φ, is the transformation matrix between the polarization compass coordinate system B-frame and the geographic coordinate system N-frame.
[0070] The polarization meridian plane can be represented by its normal vector, so there is
[0071]
[0072] In the formula, F YB and F YN are the unit normal vectors of the polarization meridian plane and respectively.
[0073] On the other hand, in the polarization compass observation system, the coordinate transformation relationship between the observed solar vector S B in the polarization observation coordinate system and the reference solar vector S N in the geographic coordinate system is expressed as:
[0074]
[0075] Since the above formula lacks a measurement relationship, the measurement model of the polarization meridian plane proposed according to the present invention can complete this measurement relationship, so there is:
[0076]
[0077] Based on the above formula, the three-axis attitude angles of the polarization compass are solved to be the pitch angle θ X , roll angle θ Y and heading angle φ.
[0078] The content not described in detail in the specification of the present invention belongs to the prior art well known to those skilled in the art.
Claims
1. A three-dimensional real-time attitude determination method of a polarization compass based on the polarization meridian plane. It is characterized in that The following steps are involved: The first step is to extract the solar polarization neutral point and solar vector based on the solar polarization field imaging mode, and establish a polarization meridian plane model; The second step is to establish a measurement model of the polarization meridian plane based on the conversion relationship between the geographic coordinate system and the polarization coordinate system. The third step is to calculate the real-time attitude information of the polarization compass, namely the heading angle, pitch angle and roll angle, based on the polarization meridian plane measurement model and the solar vector coordinate system conversion model. In the first step, based on the solar polarization field imaging mode, the solar polarization neutral point and the solar vector are extracted respectively, and a polarization meridian plane model is established. The implementation steps are as follows: (a1) Sunlight interacts with atmospheric molecules or aerosol particles in the Earth's atmosphere and scatters to generate polarization. Polarization compasses are used to analyze the atmospheric polarization field and obtain the polarization degree and polarization vector corresponding to different observation directions at the time of observation. (a2) Based on the Stokes vector formula, the polarization degree and polarization direction corresponding to different coordinates in the polarization imaging are calculated, and then the polarization degree image and polarization direction image of the polarization compass analysis are obtained; (a3) Based on the degree of polarization image, the neutral point region is obtained by using the method of degree of polarization threshold segmentation. Then, the K-means clustering and the center of gravity method of the degree of polarization are used to calculate the coordinates Ne(x NE , y NE ) of the polarization neutral point and the neutral point vector Ne in the polarization compass coordinate system; (a4) Based on the polarization direction image, according to the principle that the polarization vector is perpendicular to the solar vector, an optimized solution model for the solar vector is established to obtain the optimal solar vector (S B ) = [x b , y b , z b = argmin(S B T EE T S B ), where S B is the solar vector in the polarization compass coordinate system, E is the polarization E vector in the polarization compass coordinate system, S B T and E T represent the transpose of the vector; (a5)Based on the neutral point vector Ne and the solar vector S B , establish the polarization meridian plane in the polarization compass coordinate system corresponding to this observation moment, that is, Y 2. The three-dimensional real-time attitude determination method of a polarization compass based on a polarization meridian plane according to claim 1, Features: In the second step, based on the conversion relationship between the polarization meridian plane in the geographic coordinate system and the polarization coordinate system, a measurement model of the polarization meridian plane is established, and the implementation steps are as follows: (b1) Based on the observation time t and the observation position information, calculate the reference solar vector S of the solar vector in the east-north-up geodetic coordinate system from the solar ephemeris N , that is, the solar vector S N 's altitude angle h S and azimuth angle (b2) In the N-system under the geographical coordinate system, the polarization meridian plane is represented as (b3) Based on the polarization meridian plane of the B-frame in the polarization coordinate system obtained in the first step Establish the conversion model between it in the N-frame and the B-frame respectively: In the formula, R 2 , R 1 and R 3 are the rotation matrices around the Y-axis, X-axis and Z-axis respectively, the pitch angle θ X , the roll angle θ Y and the heading angle , is the conversion matrix between the polarization compass coordinate system B-frame and the geographic coordinate system N-frame.
3. The three-dimensional real-time attitude determination method of a polarization compass based on a polarization meridian plane according to claim 2, Features: In the third step, based on the polarization meridian measurement model, the real-time attitude information of the polarization compass, i.e., the pitch angle θ X , roll angle θ Y and heading angle are calculated by combining the solar vector coordinate system conversion model. The implementation steps are as follows: (c1) Based on the reference solar vector S obtained in the second step N = [x n , y n , z n and the observed solar vector S in the first step B = [x b , y b , z b , establish a coordinate system transformation model: (c2) Based on the measurement model of the polarization meridian plane obtained in the second step, the solar vector coordinate system conversion model is constructed: (c3) Solve the above equation to obtain the three-axis attitude angles of the polarization compass, which are the pitch angle θ X , the roll angle θ Y and the heading angle .
Citation Information
Patent Citations
Method for extracting solar meridian based on polarization imaging
CN108917749A
Method used for obtaining three dimensional attitude of carrier and based on horizon and polarized light
CN108387206A
Dynamic orientation method and system based on strapdown polarized light compass
CN110231025A