A three-dimensional attitude determination method for a vehicle based on dual-vector characteristics of polarized light fields at night

Abstract

The present invention relates to a method for three-dimensional attitude determination of a carrier based on dual-vector characteristics of a nighttime polarized light field. The method comprises the following steps: obtaining nighttime polarized light field information, obtaining the coordinates of the neutral point of an image plane coordinate system and the coordinates of the moon based on the neutral plane characteristics of a polarization distribution pattern; inversely solving the polarized neutral point vector and the moon vector in the carrier coordinate system based on an imaging model, and further calculating the neutral plane normal vector of the carrier system; obtaining the moon vector in the navigation system based on time information and geographic location information, and calculating the neutral plane normal vector of the navigation system using the zenith direction vector; constructing orthogonal coordinate systems in the carrier system and the navigation system using the corresponding moon vector and normal vector; and calculating an attitude conversion matrix based on the two orthogonal coordinate systems to obtain a conversion relationship between the carrier system and the navigation system, thereby obtaining a three-dimensional attitude angle. The present invention achieves autonomous attitude determination based on the sky polarized light field without external input, eliminating external error sources and improving autonomy.

Description

Technical Field

[0001] The present invention belongs to the field of attitude determination and relates to a three-dimensional attitude determination method for a carrier based on double-vector characteristics of a nighttime polarized light field. Background Art

[0002] Bionic polarization navigation, characterized by its passive autonomy and lack of error accumulation, has seen rapid development. Current research on polarization navigation primarily focuses on daytime sunlight polarization, with limited research on moonlight polarization navigation for nighttime environments. Polarization navigation offers a viable solution for autonomous nighttime navigation, particularly when satellite navigation signals are limited.

[0003] However, existing nighttime polarimetric attitude determination algorithms require known horizontal attitude information and external sensors such as inertial navigation systems and horizon instruments to provide a horizontal attitude reference. Consequently, their application is limited to providing heading information for the vehicle and can introduce errors from external navigation information. For example, in Chinese patent ZL201911250896.8 (Autonomous heading and attitude determination method based on polarimetric-astronomical angle observation), star sensors provide roll and pitch angle measurements to determine the vehicle's three-dimensional attitude, but this also introduces errors from the star sensors.

[0004] Furthermore, existing methods for 3D attitude estimation based on polarization information fail to account for factors such as low illumination and poor visibility in nighttime scenes. Chinese Patent ZL201310731899.X (A Method for 3D Attitude Determination Using Atmospheric Polarized Light) can determine the 3D attitude of an aircraft based on atmospheric polarization distribution patterns. However, this method requires matching horizontal polarization information, making it unsuitable for applications at lower altitudes. Chinese Patent ZL201611078923.4 (A Method for 3D Attitude Determination Based on Dual Neutral Point Vectors) relies on two neutral point vectors obtained from the carrier and navigation systems to estimate the vehicle's attitude. However, simultaneous observation of both neutral points is difficult in nighttime scenes, limiting its applicability. Chinese Patent ZL202210828890.X (A Method for 3D Attitude Determination Based on Underwater Downstream Radiant Intensity and Polarized Light Field) can determine the 3D attitude of underwater vehicles, but because it requires the optical refraction characteristics of the atmosphere and water, this method is not suitable for atmospheric environments. Summary of the Invention

[0005] To address the issue of polarization navigation requiring horizontal attitude information from other sensors in nighttime scenarios, and considering that existing methods for three-dimensional attitude estimation based on polarization information do not consider the characteristics of nighttime scenarios, the present invention proposes a method for three-dimensional carrier attitude determination based on the dual-vector characteristics of the nighttime polarization light field. This method analyzes the main characteristics of the polarization distribution pattern of moonlight at nighttime, and based on the measurement of the moon vector and neutral point vector in the nighttime polarization light field, obtains the coordinate transformation relationship between the carrier system and the navigation system, and calculates the three-dimensional attitude angle information. The present invention can provide a three-dimensional carrier attitude estimation result by measuring the moon vector and neutral point vector in the nighttime sky polarization light field, providing a solution for autonomous polarization attitude determination in nighttime scenarios that does not rely on external horizontal reference information. The present invention can achieve three-dimensional carrier attitude estimation using only polarization navigation information without introducing errors from other sensors, thereby improving the autonomy and accuracy of nighttime polarization navigation.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for three-dimensional attitude determination of a carrier based on dual-vector characteristics of a nighttime polarized light field comprises the following steps:

[0008] The first step is to use an image polarization sensor to obtain the polarized light field of the sky, and obtain the image plane coordinate system based on the neutral point characteristics in the atmospheric polarization distribution pattern. Coordinates of the neutral point in the system ;

[0009] The second step is to calculate the carrier system according to the camera imaging model and the polarization distribution pattern information in the image plane coordinate system. Neutral point coordinates under the system Coordinates with the moon , get the neutral point vector and moon vector , further solve Normal vector of the neutral plane where the moon is located ;

[0010] The third step is to import the time information and geographic location information into the astronomical calendar to obtain the navigation system Moon vector , based on the zenith direction vector ,get Normal vector of the neutral plane where the line connecting the zenith and the moon is located ;

[0011] Step 4: Under the load system, based on the moon vector and the midplane normal vector Construct an orthogonal coordinate system. Similarly, construct a moon vector based on the navigation system. and the midplane normal vector Construct an orthogonal coordinate system;

[0012] Step 5: Based on Department and The coordinate transformation relationship between the carrier system and the navigation system is obtained by calculating the two orthogonal coordinate systems under the system, and the three-dimensional attitude information is obtained.

[0013] Compared with the existing technology, the present invention has the following beneficial effects:

[0014] (1) The present invention only provides three-dimensional attitude information based on the image-based polarization sensor, and the nighttime polarization navigation technology is no longer limited to heading estimation.

[0015] (2) The present invention takes into account the situation that the moon pattern cannot be directly observed in actual application scenarios, and obtains moon information based on global polarization light field information, thereby improving the environmental adaptability of using polarization light field for three-dimensional attitude determination.

[0016] (3) The present invention only uses polarized light navigation technology to solve three-dimensional spatial information, avoiding the introduction of external sensor errors and reducing the impact that external errors may bring. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a three-dimensional attitude determination method for a carrier based on the dual-vector characteristics of the nighttime polarized light field according to the present invention.

[0018] Figure 2 Schematic diagram of the image plane coordinate system involved in the present invention.

[0019] Figure 3 Schematic diagram of the carrier system involved in the present invention.

[0020] Figure 4 Schematic diagram of the navigation coordinate system and polarized light field involved in the present invention. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0022] like Figure 1 As shown, a method for three-dimensional attitude determination of a carrier based on dual-vector features of a nighttime polarized light field according to an embodiment of the present invention includes the following steps:

[0023] The first step, such as Figure 2As shown in the figure, an image polarization sensor is used to collect the polarized light field of the sky, obtain a polarization image, and calculate the polarization angle (AOP) information. The edge contour lines of different polarization angle areas are drawn in the polarization angle image. Based on the characteristic that the contour lines of each polarization angle of the polarized light field intersect at the neutral point, the image plane coordinate system is obtained. Coordinates of the neutral point in the system .

[0024] When the entire moon pattern is visible in the polarization image, the image plane coordinate system is obtained by extracting the centroid. Moon coordinates in the system .

[0025] The second step is to calculate the conversion relationship between the image plane coordinate system and the carrier coordinate system based on the camera model, including: using the calibrated camera model, according to the coordinates of each group of pixels in the image plane coordinate system, calculate the azimuth and zenith angle of the detection point corresponding to the group of pixels in the carrier coordinate system, so as to obtain the azimuth and zenith angle of the detection point in the carrier coordinate system. Based on the coordinates of the neutral point , obtained by solving the camera model Neutral point coordinates . Based on the camera model The coordinate solution process under the system is as follows:

[0026] First, calculate the coordinates of the neutral point of the image plane coordinate system and the image center coordinates obtained by camera model calibration The relative position relationship between , considering that the coordinate axis directions of the image plane coordinate system and the carrier coordinate system are different:

[0027] .

[0028] Next, calculate The coordinates of the neutral point are:

[0029] ;

[0030] ;

[0031] .

[0032] in, is the focal length of the camera lens used.

[0033] If we can get Moon coordinates of the system , can be calculated using the camera model The moon coordinates in the system .

[0034] like Figure 3 As shown in , if the moon pattern in the polarization image is incomplete or invisible, the perpendicular relationship between the polarization E vector and the moon vector can be used to achieve it. For the polarization image, the AOP corresponding to each group of pixels is , the polarization E vector obtained by solution satisfy ,in, , is the moon vector obtained by connecting the moon and the origin, and the superscript T represents the transpose of the matrix. Based on the effective polarization E vector information and its perpendicular relationship with the moon vector, we can get Moon vector .

[0035] Neutral point of the system With the moon position They are all located on the neutral plane of the sky polarized light field and can be determined based on the moon vector and the neutral point vector. Neutral plane under the system ,in , represents the cross product of vectors. Further, based on You can get its normal vector .

[0036] Step 3: Carrier Coordinate System Zenith direction vector under the system and moon vector Determine the neutral plane normal vector in the carrier coordinate system, including:

[0037] like Figure 4 As shown, the time information and longitude and latitude information are input into the astronomical calendar to obtain the moon vector in the navigation system n. . The zenith point of the system is the position directly above the carrier, with coordinates , you can get the zenith direction vector . The moon and the zenith point are located on the moon-antimoon meridian, which is included in the neutral plane of the polarized light field. and It can be determined Neutral plane under the system . Further, the corresponding normal vector can be obtained .

[0038] Step 4: Use the moon vector and the neutral surface normal as anchor vectors to construct orthogonal coordinate systems in the carrier system and navigation system respectively, including:

[0039] According to the definitions of the moon vector and the neutral plane normal vector, these two vectors must be non-parallel, so they can be used as the basic vectors for constructing an orthogonal coordinate system.

[0040] exist Under the system, based on the moon vector and the normal vector The orthogonal matrix can be constructed as follows:

[0041] ;

[0042] in, and Represent two vectors used to construct an orthogonal matrix, 、 and are three mutually perpendicular unit vectors obtained by calculation and point to one direction of the coordinate axis respectively.

[0043] Similarly, in The system can be based on the moon vector and the normal vector Construct the following orthogonal matrix:

[0044] ;

[0045] in, and Represent two vectors used to construct an orthogonal matrix, 、 and are three mutually perpendicular unit vectors obtained by calculation and point to one direction of the coordinate axis respectively.

[0046] The coordinate transformation relationship matrix A can be obtained:

[0047] ;

[0048] in, and is the anchor vector, and To calculate the vector, select and The coordinate transformation relationship matrix calculated as the anchor vector is: Similarly, select and The coordinate transformation relationship matrix calculated as the anchor vector is: .

[0049] Step 5: Select the standard deviation (SD) of the anchor vector as the coordinate transformation matrix. and The accuracy measure is used to establish the weight coefficient. The attitude matrix obtained by using the moon vector as the anchor vector SD corresponding to the moon vector measurement , similarly, using the normal vector as the pose matrix of the anchor vector SD corresponding to normal vector measurement .

[0050] The attitude transformation matrix can be obtained based on SD as the accuracy weight The linear unbiased minimum variance estimate of for:

[0051] ;

[0052] Using the unknown SD with error and Instead of exact and , we can get:

[0053] ;

[0054] The weights of the two attitude transformation matrices are obtained by SD, and the weights are defined as 、 They are:

[0055] ;

[0056] That is .

[0057] definition for The variance of , that is:

[0058] ;

[0059] From the above formula we can see that the smallest You can select the weight When obtained.

[0060] Based on and The posture transformation matrix with the minimum variance is obtained as:

[0061] ;

[0062] in, To be better than and The pose transformation matrix estimation result.

[0063] Based on the attitude conversion matrix, the three-dimensional attitude information of the carrier can be calculated - pitch angle, roll angle and heading angle.

[0064] The contents not described in detail in the specification of the present invention belong to the common knowledge of professionals in this field.

[0065] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand the present invention, but the scope of protection of the present invention is not limited thereto. Any other changes or replacements that can be easily thought of by those skilled in the art should be included in the scope of protection of the present invention.

Claims

1. A three-dimensional attitude determination method for a carrier based on the dual-vector characteristics of polarized light fields at night, characterized in that: The steps include: The first step is to use an image polarization sensor to obtain the polarized light field of the sky, and obtain the image plane coordinate system based on the neutral point characteristics in the atmospheric polarization distribution pattern. Coordinates of the neutral point in the system ; The second step is to calculate the carrier system according to the camera imaging model and the polarization distribution pattern information in the image plane coordinate system. Neutral point coordinates under the system Coordinates with the moon , get the neutral point vector and moon vector , further solve Normal vector of the neutral plane where the moon is located ; The third step is to import the time information and geographic location information into the astronomical calendar to obtain the navigation system Moon vector , based on the zenith direction vector ,get Normal vector of the neutral plane where the line connecting the zenith and the moon is located ; Step 4: Under the load system, based on the moon vector and the midplane normal vector Construct an orthogonal coordinate system. Similarly, construct a moon vector based on the navigation system. and the midplane normal vector Construct an orthogonal coordinate system; Step 5: Based on Department and The coordinate transformation relationship between the carrier system and the navigation system is obtained by calculating the two orthogonal coordinate systems under the system, and the three-dimensional attitude information is obtained.

2. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 1 is characterized in that: The first step comprises: Use an image-based polarization sensor to collect the polarized light field of the sky to obtain a polarization image, calculate the polarization angle information, and draw the edge contour lines of different polarization angle areas in the polarization angle image. Based on the characteristic that the contour lines of each polarization angle of the polarized light field intersect at the neutral point, the image plane coordinate system is obtained. Coordinates of the neutral point in the system ; When the entire moon pattern is visible in the polarization image, the image plane coordinate system is obtained by extracting the centroid. Coordinates of the Moon in the system .

3. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 1 is characterized in that: The second step includes: Using the calibrated camera model, according to the coordinates of each group of pixels in the image plane coordinate system, the azimuth and zenith angle of the detection point corresponding to the group of pixels in the carrier coordinate system are calculated, thereby obtaining the zenith angle of the point in the carrier coordinate system. Coordinates of the system; based on the neutral point coordinates , obtained by solving the camera model Neutral point coordinates ; Camera model based The coordinate solution process under the system is as follows: First, calculate the coordinates of the neutral point of the image plane coordinate system and the image center coordinates obtained by camera model calibration The relative position relationship between , considering that the coordinate axis directions of the image plane coordinate system and the carrier coordinate system are different: ； Next, calculate The coordinates of the neutral point are: ； ； ； in, is the focal length of the camera lens used.

4. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 3 is characterized in that: If you get in the first step Moon coordinates , then the camera model is used to calculate Set the moon coordinates When the moon pattern in the polarized image is incomplete or invisible, the perpendicular relationship between the polarized E vector and the moon vector is used to achieve this; For polarized images, the polarization angle corresponding to each group of pixels is , the polarization E vector obtained by solution satisfy ,in, , The moon vector is obtained by connecting the moon and the origin. Based on the effective polarization E vector information and its perpendicular relationship with the moon vector, it is solved to get Moon vector ; Neutral point With the moon position They are all located on the neutral plane of the sky polarized light field, determined based on the moon vector and the neutral point vector Neutral plane under the system ,in ; based on Get its normal vector .

5. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 1 is characterized in that: The third step includes: Input the time information and longitude and latitude information into the astronomical calendar to obtain the moon vector in the navigation system n ; The zenith point of the system is the position directly above the carrier, with coordinates , that is, to obtain the zenith direction vector ; The moon and the zenith point are located on the moon-antimoon meridian, and the moon-antimoon meridian is located on the neutral plane of the polarized light field. and Sure Neutral plane under the system , and thus obtain the corresponding normal vector .

6. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 1 is characterized in that: The fourth step includes: exist Under the system, based on the moon vector and the normal vector Construct the following orthogonal matrix: ； in, and Represent two vectors used to construct an orthogonal matrix, 、 and are three mutually perpendicular unit vectors obtained by calculation and each points to a direction of the coordinate axis; exist Based on the moon vector and the normal vector Construct the following orthogonal matrix: ； in, and Represent two vectors used to construct an orthogonal matrix, 、 and are three mutually perpendicular unit vectors obtained by calculation and each points to a direction of the coordinate axis; Thus, the coordinate transformation relationship matrix A is obtained: ； in, and is the anchor vector, and To calculate the vector; select and The coordinate transformation relationship matrix calculated as the anchor vector is: ;choose and The coordinate transformation relationship matrix calculated as the anchor vector is: .

7. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 1 is characterized in that: The fifth step includes: Select the error standard deviation corresponding to the anchor vector as the coordinate transformation relationship matrix obtained by solving it and Accuracy measurement indicators, establish weight coefficients; use the moon vector as the anchor vector to obtain the attitude matrix Corresponding standard deviation of the moon vector measurement , using the normal vector as the pose matrix of the anchor vector The standard deviation of the error corresponding to the normal vector measurement .

8. The method for determining the three-dimensional attitude of a carrier based on the dual-vector characteristics of the nighttime polarized light field according to claim 7, characterized in that: Get the attitude transformation matrix based on the error standard deviation as the accuracy weight The linear unbiased minimum variance estimate of is: ； Use the unknown error standard deviation with error and Instead of exact and ,get: ； The weights of the two attitude transformation matrices are obtained by the error standard deviation, and the weights are defined as follows: ； That is ; definition for The variance of , that is: ； From the above formula, we can get the minimum In the selection weight When you get Based on and The posture transformation matrix with the minimum variance is obtained as: ； in, To be better than and The pose transformation matrix estimation result; Based on the attitude conversion matrix, the three-dimensional attitude information of the carrier is calculated. The three-dimensional attitude information includes pitch angle, roll angle and heading angle.

Images