A polarization vector non-orthogonality error identification method

Abstract

The application discloses a polarized vector non-orthogonal error identification method, comprising the following steps: acquiring an atmospheric polarization image; constructing a polarized vector non-orthogonal error angle model; and substituting prior information and observation information of each observation point in the atmospheric polarization image into the polarized vector non-orthogonal error angle model to estimate the non-orthogonal error angle of each observation point in the atmospheric polarization image. The application is applied to the field of polarized light navigation, and the non-orthogonal error of the polarized vector caused by multiple scattering under cloudy weather is quantitatively introduced into the model construction process. According to the characteristics that the non-orthogonal error angle changes slowly in a short time, the non-orthogonal error of the polarized vector is estimated based on the prior information and the observation information of the observation point, so that the influence of the non-orthogonal error angle on the orientation result can be better compensated, and the application can be effectively applied to the polarized light orientation method under cloudy weather.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of polarized light navigation, and particularly relates to a polarized vector non-orthogonal error identification method. BACKGROUND

[0002] Accurate positioning information is the basis for long-time unmanned platform autonomous navigation, and in the case of satellite positioning or complex electromagnetic interference environment, the existing inertial navigation has error accumulation, and the geomagnetic navigation is easily disturbed, which cannot meet the requirements of high-precision navigation of unmanned platform.

[0003] Sunlight is absorbed and scattered by atmospheric particles during propagation to produce polarized light, and a large amount of polarized light converges in the sky to form an atmospheric polarization pattern with certain distribution rules. The direction information is contained in the atmospheric polarization pattern, and based on the geometric characteristic that the E vector of the polarized light is perpendicular to the scattering plane in the Rayleigh scattering model, the carrier azimuth angle can be estimated by measuring the atmospheric polarization information, that is, the conventional directional method based on the orthogonality of the polarized vector. Although the directional method based on the orthogonality of the polarized vector can effectively improve the directional accuracy under the conditions of sunny weather with shelter, little cloud and cloudy weather, but the effect is not obvious for overcast weather, and the reason is that the polarization degree is not obvious under overcast weather, even if there is a small part of Rayleigh point, but it is difficult to screen, and the E vector of the non-Rayleigh point is no longer perpendicular to the scattering plane, but there is an angle between the E vector of the Rayleigh point, which is called the non-orthogonal error angle of the polarized vector. In addition, the non-orthogonal error angle of the non-Rayleigh point is large under overcast weather, and if the directional method based on the orthogonality of the polarized vector is continued, a large directional error will be introduced. SUMMARY

[0004] In view of the above problems in the prior art, the present application provides a polarized vector non-orthogonal error identification method, which can estimate the non-orthogonal error angle of the polarized vector and be effectively applied to the polarized light directional method under overcast weather.

[0005] To achieve the above object, the present application provides a polarized vector non-orthogonal error identification method, comprising the following steps:

[0006] Step 1, obtaining an atmospheric polarization image;

[0007] Step 2, constructing a polarized vector non-orthogonal error angle model, comprising:

[0008] The E vector e of the observation point is decomposed into the coordinate system composed of the orthogonal axis group of the sun direction vector s, the observation direction vector v and vxs, that is:

[0009] e=r1·s+r2·v+r3·v×s

[0010] Wherein, r1=sinη / / sinτ, r2=-sinη / / tanτ, r3=-cosη / / sinτ are the modulus of the E vector e of the observation point on three axes respectively, η is the non-orthogonal error angle of the observation point, τ is the scattering angle;

[0011] The projection of the E vector e of the observation point to the light compass coordinate system is:

[0012] e c = r1·s c + r2·v c + r3·v c × s c

[0013] Wherein, e c is the projection of the E vector e of the observation point on the light compass coordinate system, s c is the projection of the sun direction vector on the light compass coordinate system, v c is the projection of the observation direction vector on the light compass coordinate system;

[0014] Let f(η)=r1·s c + r2·v c + r3·v c × s c - e c be the objective function, then the polarization vector non-orthogonal error angle model is

[0015] Step 3, substituting the prior information and the observation information of each observation point in the atmospheric polarization image into the polarization vector non-orthogonal error angle model, and estimating the non-orthogonal error angle of each observation point in the atmospheric polarization image.

[0016] In one of the embodiments, in step 1, the scattering angle is specifically:

[0017]

[0018] Wherein, T is the transpose of the matrix.

[0019] In one of the embodiments, the projection of the E vector e of the observation point on the light compass coordinate system is specifically:

[0020]

[0021] is the conversion matrix of the observation point coordinate system to the light compass coordinate system, and is expressed as:

[0022]

[0023] Wherein, α vis the azimuth angle of the observation point in the optical compass coordinate system, γ v is the zenith angle of the observation point in the optical compass coordinate system, α v , γ v is the observation information of the observation point;

[0024] e v is the projection of the E vector e of the observation point on the observation point coordinate system, and is expressed as:

[0025]

[0026] wherein φ is the polarization angle of the observation point.

[0027] In one embodiment, the polarization angle φ of the observation point is calculated by using the Stokes vector method, and specifically:

[0028] The Stokes vector S0 contains four parameters S0 = [S1 S2 S3 S4], wherein S1 represents the total light intensity, S2 is the light intensity difference in the 0° and 90° directions, S3 is the light intensity difference in the 45° and 135° directions, and S4 is the circular polarization component. Then, the light intensity after the polarized light passes through the polarizer is:

[0029]

[0030] wherein θ1 is the polarization angle, is the included angle between the polarization direction of the sensor and the optical axis of the system;

[0031] By omitting the S4 component, the simplified Stokes intensity equation is obtained as:

[0032]

[0033] When the polarization angle θ1 is 0°, 45° and 90° respectively, the Stokes parameters of each pixel point on the image are obtained as:

[0034]

[0035] Without considering the circularly polarized light, the polarization angle of each pixel point is φ = (1 / 2) · arctan (S3 / S2).

[0036] In one embodiment, the projection of the sun direction vector on the optical compass coordinate system is specifically:

[0037]

[0038] is the conversion matrix from the carrier coordinate system to the optical compass coordinate system, and is expressed as:

[0039]

[0040] The conversion matrix of the navigation coordinate system to the light compass coordinate system is represented as:

[0041]

[0042] Wherein, r0, θ0, Ψ0 are the roll angle, pitch angle and heading angle of the carrier provided by other sensors, i.e. prior information;

[0043] s n The sun direction vector in the navigation coordinate system is represented as:

[0044]

[0045] Wherein, T is the transpose of the matrix, is the azimuth angle of the sun direction vector in the navigation coordinate system, is the elevation angle of the sun direction vector in the navigation coordinate system, i.e. prior information.

[0046] In one embodiment, the projection of the observation direction vector on the light compass coordinate system is specifically:

[0047] v c = [siny v ·cosα v sinγ v ·sinα v -cosγ v ] T

[0048] Wherein, T is the transpose of the matrix, α v is the azimuth angle of the observation point in the light compass coordinate system, γ v is the zenith angle of the observation point in the light compass coordinate system, α v , γ v are the observation information of the observation point.

[0049] In one embodiment, in step 3, the polarization vector non-orthogonal error angle model is solved based on the interior point method

[0050] Compared with the prior art, the present application has the following beneficial technical effects:

[0051] The application introduces the non-orthogonal error of the polarization vector caused by multiple scattering under cloudy weather into the model construction process, estimates the non-orthogonal error of the polarization vector based on the prior information and the observation information of the observation point according to the characteristic that the non-orthogonal error angle changes slowly in a short time, so that the influence of the non-orthogonal error angle on the orientation result can be better compensated, and the polarization light orientation method under cloudy weather can be effectively applied. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only show some of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from the structures shown in the drawings without creative labor.

[0053] Figure 1 The flow chart of the polarization vector non-orthogonal error identification method in the embodiment of the present application;

[0054] Figure 2 The scattering process diagram of the sunlight in the sky in the embodiment of the present application;

[0055] Figure 3 The diagram of the polarization vector non-orthogonal error angle in the embodiment of the present application;

[0056] Figure 4 The diagram of the first orthogonal relationship of the polarization vector in the embodiment of the present application;

[0057] Figure 5 The diagram of the second orthogonal relationship of the polarization vector in the embodiment of the present application.

[0058] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the drawings. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0060] It should be noted that all the directionality indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative position relationship, motion condition, etc. between the components in a certain specific posture (as shown in the drawings), and if the specific posture changes, the directionality indications will also change accordingly.

[0061] In addition, the technical solutions among various embodiments of the present application can be combined with each other, but the combination of the technical solutions should be considered not to exist and not within the protection scope of the present application when the combination of the technical solutions appears contradictory or unachievable on the basis that the combination of the technical solutions can be realized by the ordinary skilled in the art.

[0062] The embodiment discloses a polarization vector non-orthogonal error identification method, which quantitatively introduces the polarization vector non-orthogonal error caused by multiple scattering under cloudy weather into the model construction process, estimates the polarization vector non-orthogonal error based on the prior information and the observation information of the observation points according to the characteristic that the non-orthogonal error angle changes slowly in a short time, so that the influence of the non-orthogonal error angle on the orientation result can be better compensated, and the method can be effectively applied to the polarization light orientation method under cloudy weather.

[0063] Reference Figure 1 The polarization vector non-orthogonal error identification method in the embodiment specifically includes the following steps:

[0064] Step 1, acquiring an atmospheric polarization image;

[0065] Step 2, constructing a polarization vector non-orthogonal error angle model;

[0066] Step 3, substituting the prior information and the observation information of each observation point in the atmospheric polarization image into the polarization vector non-orthogonal error angle model to estimate the non-orthogonal error angle of each observation point in the atmospheric polarization image.

[0067] Reference Figure 2 A schematic diagram of the scattering process of sunlight in the sky is shown in FIG. 1, in which O c represents an observer on the ground, V represents an observation point in the sky, S represents the sun, Z represents the zenith point, N represents the geographic north direction, τ represents the scattering angle, γ s , and α s represent the solar zenith angle and the azimuth angle respectively, φ represents the polarization angle of the observation point, α v represents the azimuth angle of the observation point, γ v , and α v represent the zenith angle and the azimuth angle of the observation point V in the polarization light compass coordinate system respectively.

[0068] In the embodiment, the right-front-up coordinate system is defined as the polarization light compass coordinate system (hereinafter referred to as the "c system"), the X c axis and the Y c axis point to the horizontal axis and the vertical axis of the polarization light compass respectively, and the Z c axis is the direction of the optical axis of the polarization light compass. cThe axis points to the zenith direction. Define the North-East-Ground coordinate system as the navigation coordinate system (hereinafter referred to as "n system") with O n -X n Y n Z n , wherein O n is coincident with O c , Figure 2 In order to express simply, Y n axis and Z n axis are omitted in the following description. The direction of X n axis is the geographical north direction, which is represented by N in the following description. Figure 2 Define the Front-Right-Down coordinate system as the carrier coordinate system (hereinafter referred to as "b system") with O b -X b Y b Z b , wherein O b is coincident with O c , Figure 2 In order to express simply, Y b axis and Z b axis are omitted in the following description. The heading angle Ψ of the carrier is the angle between X b axis and N. Define O v -X v Y v Z v as the observation point coordinate system (hereinafter referred to as "v system"), Z v axis points to the observation point direction, X v axis is in the observation plane ΔO v VV' and Y v axis can be determined by the right-hand rule. Figure 2 In order to express simply, Y v axis is omitted in the following description. e is the unit vector of the vibration direction of the polarization light E vector at the observation point V, v is the unit vector of the observation direction, and s is the unit vector of the sun direction.

[0069] Referring to Figure 3 , FIG. 1 is a schematic diagram of the polarization vector non-orthogonal error angle, wherein the coordinate system is consistent with that in Figure 2 . In the figure, the polarization light E vector e' at the Rayleigh point is perpendicular to the scattering plane ΔSO c O v , and the polarization light E vector e at the non-Rayleigh point is not perpendicular to the scattering plane ΔSO c O v . Since light is a transverse wave, e' and e are always perpendicular to the incident light direction vector v, and there is a polarization vector non-orthogonal error angle between e and e', which is denoted as η in the embodiment.

[0070] Referring to Figure 4 , Figure 5Whether the polarization vector orthogonal relationship exists or not, e can be represented by the incident light direction vector v, the sun direction vector s and vxs. Therefore, the specific construction process of the polarization vector non-orthogonal error angle model in the embodiment is as follows:

[0071] First, the E vector e of the observation point is decomposed into the coordinate system composed of the orthogonal axis group of the sun direction vector s, the observation direction vector v and vxs. That is:

[0072] e = r1s + r2v + r3vxs (1)

[0073] Wherein, r1 = sinη / / sinτ, r2 = -sinη / / tanτ, r3 = -cosη / / sinτ are the module lengths of the E vector e of the observation point on the three axes respectively;

[0074] The calculation process of the scattering angle τ is as follows:

[0075]

[0076] Wherein, s c is the projection of the sun direction vector on the light compass coordinate system, v c is the projection of the observation direction vector on the light compass coordinate system, and T is the transpose of the matrix;

[0077] In the ideal case, the non-orthogonal error angle η of the Rayleigh point is 0, at this time, formula (1) degenerates into the polarization vector orthogonal model, that is, e = r3vxs;

[0078] Then, the E vector e of the observation point is projected to the light compass coordinate system, that is:

[0079] e c = r1s c + r2v c + r3v c × s c (3)

[0080] Wherein, e c is the projection of the E vector e of the observation point on the light compass coordinate system;

[0081] Finally, let f(η) = r1s c + r2v c + r3v c × s c - e c be the objective function, and theoretically f(η) = 0, but due to the existence of the polarization vector non-orthogonal error angle, usually f(η) ≠ 0, and therefore the final polarization vector non-orthogonal error angle model is:

[0082]

[0083] In the embodiment, since the non-orthogonal error angle changes slowly in a short time, the projection of the sun direction vector on the optical compass coordinate system s c , the projection of the observation direction vector on the optical compass coordinate system v c , and the projection of the E vector of the observation point on the optical compass coordinate system e c may be calculated by prior information in the process of constructing the non-orthogonal error angle model of the polarization vector.

[0084] In the embodiment, the specific calculation process of the projection of the E vector of the observation point on the optical compass coordinate system is as follows:

[0085]

[0086] is the conversion matrix from the v system to the c system, and is represented as:

[0087]

[0088] The zenith angle γ v and the azimuth angle α v of the observation point in the c system are specifically as follows:

[0089]

[0090] where (x v , y v ) is the coordinate of the observation point in the atmospheric polarization image, (x c , y c ) is the coordinate of the center point of the atmospheric polarization image, f c is the focal length of the polarization optical compass obtained by calibration, and (x v , y v ) and (x c , y c ) are the observation information of the observation point.

[0091] e v is the projection of the E vector e of the observation point on the v system, and is represented as:

[0092]

[0093] where φ is the polarization angle of the observation point, and φ is the prior information.

[0094] In the embodiment, the polarization angle of the observation point in the process of constructing the polarization vector non-orthogonal error angle model can be calculated by using the Stokes vector method. The Stokes vector contains four parameters S0=[S1 S2 S3 S4], wherein S1 represents the total light intensity, S2 is the light intensity difference in the 0° and 90° directions, S3 is the light intensity difference in the 45° and 135° directions, and S4 is the circular polarization component. The light intensity after the polarized light passes through the polarizer is:

[0095]

[0096] wherein θ1 is the polarization angle, is the angle between the polarization direction of the sensor and the system optical axis;

[0097] Since the circular polarization component in nature is very small, the S4 component is generally omitted in actual calculation, and the simplified Stokes intensity equation is obtained as follows:

[0098]

[0099] When the polarization angle θ1 is 0°, 45° and 90° respectively, the Stokes parameters of each pixel point on the image are obtained as follows:

[0100]

[0101] Without considering the circularly polarized light, the polarization angle of each pixel point is obtained as follows:

[0102]

[0103] In the embodiment, the specific calculation process of the projection of the sun direction vector on the optical compass coordinate system is as follows:

[0104]

[0105] is the conversion matrix from the b system to the c system, and is represented as:

[0106]

[0107] is the conversion matrix from the n system to the c system, and is represented as:

[0108]

[0109] wherein r0, θ0 and Ψ0 are respectively the carrier roll angle, pitch angle and heading angle provided by other sensors (for example, satellite double antennas), and r0, θ0 and Ψ0 are prior information;

[0110] s n is the sun direction vector in the navigation coordinate system, and is represented as:

[0111]

[0112] wherein, is the azimuth angle of the sun direction vector in the navigation coordinate system, is the elevation angle of the sun direction vector in the navigation coordinate system, which can be obtained by the sun ephemeris under the condition of known time and geographical position, is also prior information.

[0113] In the embodiment, the specific calculation process of the projection of the observation direction vector on the light compass coordinate system is as follows:

[0114] v c = [sinγ v ·cosα v sinγ v ·sinα v -cosγ v ] T (18)

[0115] After the construction of the non-orthogonal error angle model of the polarization vector is completed, the observation information of each observation point in the atmospheric polarization image is substituted into the non-orthogonal error angle model of the polarization vector, and a series of prior information is combined to solve the non-orthogonal error angle model, so that the non-orthogonal error angle of each observation point can be estimated.

[0116] According to formula (4), the problem of solving the non-orthogonal error angle can be converted into a nonlinear programming problem with boundary constraints, so in the specific implementation process of solving the non-orthogonal error angle model of the polarization vector in step 2, the interior point method is used to solve the nonlinear optimization problem, which has the advantages of low calculation complexity and strong robustness.

[0117] In the specific application process, after the non-orthogonal error angle of each observation point is calculated, the non-orthogonal error angle of each observation point can be substituted into r1, r2 and r3 in formula (3), so that the current sun direction vector can be calculated in combination with the observation information of the observation point and a series of prior information, and the current actual heading angle of the carrier can be derived according to the sun direction vector.

[0118] The above only describes the preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structural transformation made under the inventive concept of the present application, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.​

Claims

1. A method for identifying polarization vector non-orthogonality errors, characterized in that: The steps include: Step 1, obtaining an atmospheric polarization image; Step 2: Constructing a polarization vector non-orthogonal error angle model, including: Decompose the E vector e of the observation point into the coordinate system composed of the sun direction vector s axis, the observation direction vector v axis and the orthogonal axis composed of v×s: e=r1·s+r2·v+r3·v×s Where r1 = sinη / sinτ, r2 = -sinη / tanτ, r3 = -cosη / sinτ, respectively, are the modulus lengths of the E vector e of the observation point on the three axes, η is the non-orthogonality error angle of the observation point, and τ is the scattering angle; Project the E vector e of the observation point to the optical compass coordinate system, which is: yes c =r1·s c +r2·v c +r3·v c ×s c Among them, e c is the projection of the E vector e of the observation point on the optical compass coordinate system, s c is the projection of the sun direction vector on the optical compass coordinate system, v c is the projection of the observation direction vector on the optical compass coordinate system; Let f(η) = r1·s c +r2·v c +r3·v c ×s c -e c As the objective function, the polarization vector non-orthogonal error angle model is Step 3: Substitute the prior information and the observation information of each observation point in the atmospheric polarization image into the polarization vector non-orthogonal error angle model to estimate the non-orthogonal error angle of each observation point in the atmospheric polarization image.

2. The polarization vector non-orthogonality error identification method according to claim 1, characterized in that: In step 1, the scattering angle is specifically: Where T is the transpose of the matrix.

3. The polarization vector non-orthogonality error identification method according to claim 1, characterized in that: The projection of the E vector e of the observation point on the optical compass coordinate system is specifically: is the transformation matrix from the observation point coordinate system to the optical compass coordinate system, expressed as: Among them, α v is the azimuth of the observation point in the optical compass coordinate system, γ v is the zenith angle of the observation point in the optical compass coordinate system, α v , γ v That is the observation information of the observation point; e v is the projection of the E vector e of the observation point on the observation point coordinate system, expressed as: Where φ is the polarization angle of the observation point.

4. The polarization vector non-orthogonality error identification method according to claim 3, characterized in that: The polarization angle φ of the observation point is calculated using the Stokes vector method, specifically: The Stokes vector S0 contains four parameters S0 = [S1 S2 S3 S4], where S1 represents the total light intensity, S2 is the light intensity difference between 0° and 90°, S3 is the light intensity difference between 45° and 135°, and S4 is the circular polarization component. The intensity of polarized light after passing through the polarizer is: Where θ1 is the polarization angle, is the angle between the sensor polarization direction and the system optical axis; Omitting the S4 component, the simplified Stokes intensity equation is: When the polarization angle θ1 is 0°, 45°, and 90° respectively, the Stokes parameter of each pixel on the image is: Without considering circularly polarized light, the polarization angle of each pixel can be obtained as φ = (1 / 2) arctan (S3 / S2).

5. The polarization vector non-orthogonality error identification method according to claim 1, characterized in that: The projection of the sun direction vector on the optical compass coordinate system is specifically: is the transformation matrix from the carrier coordinate system to the optical compass coordinate system, expressed as: is the transformation matrix from the navigation coordinate system to the optical compass coordinate system, expressed as: Among them, r0, θ0, and Ψ0 are respectively provided by other sensors for the carrier roll angle, pitch angle, and heading angle, which are the prior information; s n is the sun direction vector in the navigation coordinate system, expressed as: Where T is the transpose of the matrix, is the azimuth of the sun direction vector in the navigation coordinate system, is the altitude angle of the sun direction vector in the navigation coordinate system, That is the prior information.

6. The polarization vector non-orthogonality error identification method according to claim 1, characterized in that: The projection of the observation direction vector on the optical compass coordinate system is specifically: v c =[sinγ v ·cosα v sinγ v ·sinα v -cosγ v ] T Among them, T is the transpose of the matrix, α v is the azimuth of the observation point in the optical compass coordinate system, γ v is the zenith angle of the observation point in the optical compass coordinate system, α v , γ v That is the observation information of the observation point.

7. The polarization vector non-orthogonality error identification method according to any one of claims 1 to 6, characterized in that: In step 3, the polarization vector non-orthogonality error angle model is solved based on the interior point method.

Images