Micro-array polarization camera parameter online calibration method based on star sensor

Abstract

The invention relates to a micro-array polarization camera parameter online calibration method based on a star sensor, which comprises the following steps of: establishing a state equation by taking a light intensity gain coefficient, a polarization degree coefficient and a polaroid installation angle of each polarization unit of a micro-array polarization camera as state variables; and on the basis of the principle that the vector dot product is not changed under different coordinate systems, the moon vector under the geographical system and the star vector observed by the star sensor are combined, and the moon vector under the carrier system is solved. On the basis, a polarization azimuth angle is represented based on a vertical relation between a moon vector and a polarization vector. Further, combining the polarization azimuth angle with a model of each polarization unit of the microarray polarization camera, and establishing a measurement equation of the polarized light intensity of each light path channel, a light intensity gain coefficient, a polarization degree coefficient and a polaroid installation angle; and finally, online calibration of model parameters of the microarray polarization camera is realized by adopting an unscented Kalman filtering method.

Description

Technical Field

[0001] The present invention relates to an online calibration method for parameters of a micro-array polarization camera based on a star sensor. In combination with star vector information of a star sensor, an online calibration model for parameters of a micro-array polarization camera based on light intensity is established, and online calibration of model parameters of a micro-array polarization camera is realized, thereby improving the environmental adaptability of the micro-array polarization camera under different moon phases and different weather conditions at night. Background Art

[0002] Organisms in nature, such as dung beetles and sweat bees, can sense polarized light field information at night and achieve navigation behaviors such as foraging and catching. Inspired by this navigation method of organisms, nighttime bionic polarized light navigation has received widespread attention. It has the advantages of being free from electromagnetic interference and error accumulation, providing a technical means for autonomous navigation of unmanned systems at night. As a nighttime polarized light field sensing instrument, polarization sensors can detect polarized light and are the core components of nighttime polarized light navigation. Their accuracy is the key to restricting nighttime polarized light navigation. After theoretical analysis and research, the accuracy of polarization sensors depends on the design of the optical path structure of the sensor on the one hand, and the calibration of polarization sensor model parameters on the other hand is also particularly important. The "accuracy and speed" of model parameter calibration is the key link to improve the performance and environmental adaptability of polarization sensors.

[0003] In recent years, many research institutions have conducted extensive research on polarization sensor calibration. The paper "Design and Calibration of a Novel Bio-Inspired Pixelated Polarized Light Compass" provides polarization angle information through an integrating sphere and a high-precision turntable, introduces an iterative least squares algorithm, realizes the calibration of model parameters in an array polarization camera, and improves the accuracy of the heading angle solution of the polarization camera. The patent "A multi-source error calibration method for a bionic polarization sensor based on adaptive EKF" (application number: CN201811421555.8) establishes a polarization sensor model based on multi-source interference such as installation error and measurement noise, and designs an adaptive extended Kalman filter method. While using a turntable and an integrating sphere for filtering calculations, a noise statistical estimator is used to estimate and correct the unknown time-varying noise characteristics, thereby enhancing the anti-interference ability of the sensor calibration. The patent "A calibration method for multi-camera polarization light sensors" (application number: CN201610030919.4) establishes the measurement error equation of the multi-camera polarization light sensor and the linear regression equation of each camera. Then, based on the polarization angle information provided by the turntable and integrating sphere, the target optimization function of the polarizer installation error angle was established. Finally, the installation error angle was calibrated based on Gauss-Newton iteration, which effectively improved the measurement accuracy of the polarization sensor.

[0004] The polarization sensor calibration methods proposed in the above papers and patents all rely on a turntable, which is an offline calibration, and the calibration parameters are fixed parameters. In actual application scenarios, the atmospheric environment, meteorological parameters, etc. will change, and the fixed parameters of offline calibration are difficult to apply to the above environment. In response to the above problems, the patent "A method for online calibration of polarization sensor errors based on the assistance of solar sensors" (application number: CN202311648246.5) uses solar sensors to provide high-precision solar vector information, based on the relationship between the polarization vector perpendicular to the solar vector and the observation vector under the carrier system, combined with the polarization sensor model, to achieve online calibration of the polarization sensor error parameters. However, this method is only applicable to daytime environments and is difficult to apply in nighttime environments. Therefore, around the demand for nighttime polarization sensor parameter calibration, how to combine the star vector information obtained by the star sensor, break through the online calibration technology of the polarization sensor model parameters, and realize the dynamic optimization of the polarization sensor model parameters, thereby improving the angle measurement performance and environmental adaptability of the polarization sensor in the nighttime environment requires further research. Summary of the invention

[0005] In order to solve the above problems and overcome the shortcomings of the prior art, the present invention proposes an online calibration method for parameters of a micro-array polarization camera based on a star sensor, wherein the model parameters of the micro-array polarization camera, namely, the light intensity gain coefficient, the polarization degree coefficient, and the polarization direction angle are used as the state quantities to be estimated, and a state equation is established; the moon vector information is solved by using the star vector measured by the star sensor, and then the functional relationship between the polarization azimuth angle and the moon vector is established; on this basis, combined with the micro-array polarization camera model, a measurement equation based on the output light intensity of the optical path channel of the micro-array polarization camera is established; finally, the unscented Kalman filtering method is used to realize the online calibration of the model parameters of the micro-array polarization camera.

[0006] The technical solution adopted by the present invention is: an online calibration method for parameters of a micro-array polarization camera based on a star sensor, and the implementation steps are as follows:

[0007] Step (1), using the light intensity gain coefficient of each polarization unit of the microarray polarization camera Polarization coefficient Analyzer angle α i,j As the state quantity x, i represents the serial number of the polarization unit, j = 1, ..., 4, represents the j-th optical path channel of the polarization unit. The state quantity of the i-th polarization unit is expressed as: Establishing the state equations for polarization camera parameters Among them, f(·) is the state quantity x and its first-order derivative The functional relationship between them.

[0008] Step (2): Use the star sensor to measure and obtain the star vectors under N carrier systems (b system): Where N ≥ 3. According to the astronomical calendar, the corresponding N star vector information in the geographic coordinate system (n system) is obtained And the moon vector information m n Based on the principle that the dot product of vectors is constant under different coordinate systems, the equation S is established. b ·m b =S n ·m n , get the moon vector m under the carrier system b Further, according to the relationship between the b system and each polarization unit coordinate system (l i The coordinate transformation relationship of the system is obtained. i Moon vector

[0009] Step (3), 1 i Polarization vector under the system Perpendicular to the moon vector and the observation vector Based on the above relationship, the polarization vector is established. Medium polarization azimuth With the moon vector Functional relationship Where h represents the polarization azimuth With the moon vector The functional relationship between them.

[0010] Step (4), based on the model of the i-th polarization unit in the microarray polarization camera Further combined with the polarization azimuth established in step (3) With the moon vector The functional relationship of the camera model parameter measurement equation based on the polarization output light intensity is established. in, represents the incident light intensity, d i represents the degree of polarization, y i,j represents the output light intensity of the jth optical path, g i (·),L i (·) represent the functional relationship between the input parameters and the polarization output light intensity.

[0011] Step (5), combining the state equations and measurement equations of the microarray polarization camera parameters in step (1) and step (4), establish an online calibration model of the microarray polarization camera parameters assisted by a star sensor. According to the characteristics of the model, the unscented Kalman filter method is used to online estimate the model parameters of each polarization unit in the microarray polarization camera, thereby improving the night navigation accuracy and environmental adaptability of the polarization camera.

[0012] Furthermore, in step (1), the light intensity gain coefficient of each polarization unit of the microarray polarization camera is Polarization coefficient Analyzer angle α i,j As the state quantity x, i represents the serial number of the polarization unit, j = 1, ..., 4, represents the j-th optical path channel of the polarization unit. The state quantity of the i-th polarization unit is expressed as: Establishing the state equations for polarization camera parameters Among them, f(·) is the state quantity x and its first-order derivative The functional relationship between them.

[0013] In each polarization unit of the microarray polarization camera, the model parameter can be approximately considered as a constant value per unit time, which is a slowly changing state quantity. Therefore, the state equation can be expressed as:

[0014]

[0015] Furthermore, in step (2), the star vectors of N carrier systems (b system) are measured by using a star sensor. Where N ≥ 3. According to the astronomical calendar, the corresponding N star vector information in the geographic coordinate system (n system) is obtained And the moon vector information m n Based on the principle that the dot product of vectors is constant under different coordinate systems, the equation S is established. b ·m b =S n ·m n , get the moon vector m under the carrier system b Further, according to the relationship between the b system and each polarization unit coordinate system (l i The coordinate transformation relationship of the system is obtained. i Moon vector

[0016] Define the following formula:

[0017]

[0018] in, The matrix is ​​composed of S n and m n The angle between them.

[0019] Based on the least squares method, the moon vector under the carrier system can be expressed as:

[0020]

[0021] Furthermore, the coordinate system of each polarization unit in the microarray polarization camera is defined as l i According to the center point of the microarray polarization camera (xc ,y c ) and the optical focal length f c , l i The coordinate transformation matrix between the a system and the b system can be expressed as:

[0022]

[0023] in, x p ,y p is the center point coordinate of the i-th polarization unit.

[0024] Therefore, l i Moon vector It can be expressed as:

[0025]

[0026] in, The T in the upper right corner is the transpose symbol.

[0027] Furthermore, in step (3), i Polarization vector under the system Perpendicular to the moon vector and the observation vector Based on the above relationship, the polarization vector is established. Medium polarization azimuth With the moon vector Functional relationship Where h represents the polarization azimuth With the moon vector The functional relationship between them.

[0028] In l i Under this system, the moon vector is Observation vector Based on the perpendicular relationship between the polarization vector, the observation vector and the moon vector, the polarization vector It can be expressed as:

[0029]

[0030] According to formula (6), the polarization azimuth and The relationship can be expressed as:

[0031]

[0032] in, Representation vector The mold length.

[0033] Furthermore, in step (4), based on the model of the i-th polarization unit in the microarray polarization camera Further combined with the polarization azimuth established in step (3) With the moon vector The functional relationship of the camera model parameter measurement equation based on the polarization output light intensity is established. in, represents the incident light intensity, d i represents the degree of polarization, y i,j represents the output light intensity of the jth optical path, g i (·),L i (·) represent the functional relationship between the input parameters and the polarization output light intensity.

[0034] The j-th optical path channel model of the i-th polarization unit in the microarray polarization camera is expressed as:

[0035]

[0036] Among them, y i,j Indicates the output light intensity, represents the incident light intensity, d i represents the degree of polarization, Indicates the polarization azimuth; light intensity gain coefficient Polarization coefficient Analyzer angle α i,j are the polarization camera model parameters to be estimated;

[0037] The polarization azimuth in formula (7) With the moon vector Substitute the relationship into formula (8) to establish the measurement equation based on the output light intensity of the optical path channel:

[0038]

[0039] Among them, ν i,j represents the measurement noise.

[0040] Furthermore, in step (5), the state equations and measurement equations of the microarray polarization camera parameters in step (1) and step (4) are combined to establish an online calibration model of the microarray polarization camera parameters assisted by a star sensor. According to the characteristics of the model, the unscented Kalman filter method is used to online estimate the model parameters of each polarization unit in the microarray polarization camera, thereby improving the night navigation accuracy and environmental adaptability of the polarization camera.

[0041] Combining formulas (1) and (9), the online calibration model of the micro-array polarization camera model parameters based on star sensitivity assistance can be expressed as:

[0042]

[0043]

[0044] The advantages of the present invention over the prior art are as follows: a method for online calibration of micro-array polarization camera parameters based on star sensors is proposed for the first time, and an online calibration model for micro-array polarization camera parameters based on light intensity is established in combination with star vector information of the star sensors, so that online calibration of light intensity gain coefficient, polarization coefficient and polarization analysis direction angle can be realized, thereby improving the environmental adaptability of the micro-array polarization camera in different moon phases and different weather conditions at night. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The present invention is a flow chart of a method for online calibration of parameters of a micro-array polarization camera based on a star sensor. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0047] Under different weather conditions and different moon phases at night, the fixed parameters of the microarray polarization camera model are difficult to ensure high-precision inversion and interpretation of polarization information. In view of the above problems, the present invention provides an online calibration method for microarray polarization camera parameters based on star sensors. By using the star vector information measured by the star sensor, the online estimation of the microarray polarization camera model parameters is achieved, eliminating the dependence on the turntable and improving the environmental adaptability of the microarray polarization camera in complex nighttime environments.

[0048] The specific implementation steps of the present invention are as follows:

[0049] Step 1: Use the light intensity gain coefficient of each polarization unit of the microarray polarization camera Polarization coefficient Analyzer angle α i,j As the state quantity x, i represents the serial number of the polarization unit, j = 1, ..., 4, represents the j-th optical path channel of the polarization unit. The state quantity of the i-th polarization unit is expressed as: Establishing the state equations for polarization camera parameters Among them, f(·) is the state quantity x and its first-order derivative The functional relationship between them.

[0050] In each polarization unit of the microarray polarization camera, the model parameter can be approximately considered as a constant value per unit time, which is a slowly changing state quantity. Therefore, the state equation can be expressed as:

[0051]

[0052] Step 2: Use the star sensor to measure and obtain the star vectors under N carrier systems (b system) Where N ≥ 3. According to the astronomical calendar, the corresponding N star vector information in the geographic coordinate system (n system) is obtained And the moon vector information m n Based on the principle that the dot product of vectors is constant under different coordinate systems, the equation S is established. b ·m b =S n ·m n , get the moon vector m under the carrier system b Further, according to the relationship between the b system and each polarization unit coordinate system (l i The coordinate transformation relationship of the system is obtained. i Moon vector

[0053] Define the following formula:

[0054]

[0055] in, The matrix is ​​composed of S n and m n The angle between them.

[0056] Based on the least squares method, the moon vector under the carrier system can be expressed as:

[0057]

[0058] Furthermore, the coordinate system of each polarization unit in the microarray polarization camera is defined as l i According to the center point of the microarray polarization camera (x c ,y c ) and the optical focal length f c , l i The coordinate transformation matrix between the a system and the b system can be expressed as:

[0059]

[0060] in, x p ,y p is the center point coordinate of the i-th polarization unit.

[0061] Therefore, l i Moon vector It can be expressed as:

[0062]

[0063] in, The T in the upper right corner is the transpose symbol.

[0064] Step 3, l i Polarization vector under the system Perpendicular to the moon vector and the observation vector Based on the above relationship, the polarization vector is established. Medium polarization azimuth With the moon vector Functional relationship Where h represents the polarization azimuth With the moon vector The functional relationship between them.

[0065] In l i Under this system, the moon vector is Observation vector Based on the perpendicular relationship between the polarization vector, the observation vector and the moon vector, the polarization vector It can be expressed as:

[0066]

[0067] According to formula (6), the polarization azimuth and The relationship can be expressed as:

[0068]

[0069] in, Representation vector The mold length.

[0070] Step 4: Model based on the i-th polarization unit in the microarray polarization camera Further combined with the polarization azimuth established in step (3) With the moon vector The functional relationship of the camera model parameter measurement equation based on the polarization output light intensity is established. in, represents the incident light intensity, d i represents the degree of polarization, y i,j represents the output light intensity of the jth optical path, g i (·),L i (·) represent the functional relationship between the input parameters and the polarization output light intensity.

[0071] The j-th optical path channel model of the i-th polarization unit in the microarray polarization camera is expressed as:

[0072]

[0073] Among them, y i,j Indicates the output light intensity, represents the incident light intensity, d i represents the degree of polarization, Indicates the polarization azimuth; light intensity gain coefficient Polarization coefficient Analyzer angle α i,j are the polarization camera model parameters to be estimated;

[0074] The polarization azimuth in formula (7) With the moon vector Substitute the relationship into formula (8) to establish the measurement equation based on the output light intensity of the optical path channel:

[0075]

[0076] Among them, ν i,j represents the measurement noise.

[0077] Step 5: Combine the state equations and measurement equations of the microarray polarization camera parameters in step (1) and step (4) to establish an online calibration model for the microarray polarization camera parameters assisted by a star sensor. According to the characteristics of the model, the unscented Kalman filter method is used to online estimate the model parameters of each polarization unit in the microarray polarization camera to improve the night navigation accuracy and environmental adaptability of the polarization camera.

[0078] Combining formulas (1) and (9), the online calibration model of the micro-array polarization camera model parameters based on star sensitivity assistance can be expressed as:

[0079]

[0080] The above is only a preferred embodiment of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention. The contents not described in detail in the specification of the present invention belong to the known technology of professional and technical personnel in this field.

Claims

1. A method for online calibration of micro-array polarization camera parameters based on star sensors, characterized in that: The implementation steps are as follows: Step (1), using the light intensity gain coefficient K of each polarization unit of the microarray polarization camera Ii,j , polarization coefficient K di,j , polarization direction angle α i,j As the state quantity x, i represents the serial number of the polarization unit, j=1,...,4, represents the j-th optical path channel of the polarization unit; the state quantity of the i-th polarization unit is expressed as: Establishing the state equations for polarization camera parameters Among them, f(·) is the state quantity x and its first-order derivative The functional relationship between Step (2): Use the star sensor to measure and obtain the star vectors under N carrier systems (b system): Where N ≥ 3; according to the astronomical calendar, the corresponding N star vector information in the geographic coordinate system (n system) is obtained And the moon vector information m n ; Based on the principle that the dot product of vectors is constant under different coordinate systems, the equation S is established b ·m b =S n ·m n , get the moon vector m under the carrier system b ; Further, according to the b system and each polarization unit coordinate system (l i The coordinate transformation relationship of the system is obtained. i Moon vector Step (3), 1 i Polarization vector under the system Perpendicular to the moon vector and the observation vector Based on the above relationship, the polarization vector Medium polarization azimuth With the moon vector Functional relationship Where h represents the polarization azimuth With the moon vector The functional relationship between Step (4), based on the model of the i-th polarization unit in the microarray polarization camera Further combined with the polarization azimuth established in step (3) With the moon vector The functional relationship of the camera model parameter measurement equation based on the polarization output light intensity is established. in, represents the incident light intensity, d i represents the degree of polarization, y i,j represents the output light intensity of the jth optical path, g i (·),L i (·) respectively represent the functional relationship between input parameters and polarization output light intensity; Step (5): Combine the state equations and measurement equations of the microarray polarization camera parameters in step (1) and step (4) to establish an online calibration model of the microarray polarization camera parameters assisted by a star sensor; according to the characteristics of the model, the unscented Kalman filter method is used to online estimate the model parameters of each polarization unit in the microarray polarization camera to improve the night navigation accuracy and environmental adaptability of the polarization camera.

2. The method for online calibration of parameters of a micro-array polarization camera based on a star sensor according to claim 1, characterized in that: In the step (1), the light intensity gain coefficient of each polarization unit of the microarray polarization camera is Polarization coefficient Analyzer direction angle α i,j As the state quantity x, i represents the serial number of the polarization unit, j=1,...,4, represents the j-th optical path channel of the polarization unit; the state quantity of the i-th polarization unit is expressed as: Establishing the state equations for polarization camera parameters Among them, f(·) is the state quantity x and its first-order derivative The functional relationship between In each polarization unit of the microarray polarization camera, the model parameter can be approximately considered as a constant value per unit time, which is a slowly changing state quantity; therefore, the state equation can be expressed as:

3. The method for online calibration of parameters of a micro-array polarization camera based on a star sensor according to claim 1, characterized in that: In step (2), the star vectors under N carrier systems (b system) are measured by using a star sensor Where N ≥ 3; according to the astronomical calendar, the corresponding N star vector information in the geographic coordinate system (n system) is obtained And the moon vector information m n ; Based on the principle that the vector dot product remains unchanged under different coordinate systems, the equation S is established. b ·m b =S n ·m n , get the moon vector m under the carrier system b ; Further, according to the relationship between the b system and each polarization unit coordinate system (l i The coordinate transformation relationship of the system is obtained. i Moon vector Define the following formula: Among them, the θ matrix is ​​composed of S n and m n The angle between them; Based on the least squares method, the moon vector under the carrier system can be expressed as: m b =[(S b ) T S b ] -1 S b i (3) Furthermore, the coordinate system of each polarization unit in the microarray polarization camera is defined as l i According to the center point of the microarray polarization camera (x c ,y c ) and the optical focal length f c , l i The coordinate transformation matrix between the a system and the b system can be expressed as: in, x p ,y p is the coordinate of the center point of the i-th polarization unit; Therefore, l i Moon vector It can be expressed as: in, The T in the upper right corner is the transpose symbol.

4. The method for online calibration of parameters of a micro-array polarization camera based on a star sensor according to claim 1, characterized in that: In the step (3), i Polarization vector under the system Perpendicular to the moon vector and the observation vector Based on the above relationship, the polarization vector Medium polarization azimuth With the moon vector Functional relationship Where h represents the polarization azimuth With the moon vector The functional relationship between In l i Under this system, the moon vector is Observation vector Based on the perpendicular relationship between the polarization vector, the observation vector and the moon vector, the polarization vector It can be expressed as: According to formula (6), the polarization azimuth and The relationship can be expressed as: in, Represents vector m li The mold length.

5. The method for online calibration of parameters of a micro-array polarization camera based on a star sensor according to claim 1, characterized in that: In the step (4), based on the model of the i-th polarization unit in the microarray polarization camera Further combined with the polarization azimuth established in step (3) With the moon vector The functional relationship of the camera model parameter measurement equation based on the polarization output light intensity is established. in, represents the incident light intensity, d i represents the degree of polarization, y i,j represents the output light intensity of the jth optical path, g i (·),L i (·) respectively represent the functional relationship between input parameters and polarization output light intensity; The j-th optical path channel model of the i-th polarization unit in the microarray polarization camera is expressed as: Among them, y i,j Indicates the output light intensity, represents the incident light intensity, d i represents the degree of polarization, Indicates the polarization azimuth; light intensity gain coefficient Polarization coefficient Analyzer direction angle α i,j are the polarization camera model parameters to be estimated; The polarization azimuth in formula (7) With the moon vector Substitute the relationship into formula (8) to establish the measurement equation based on the output light intensity of the optical path channel: Among them, ν i,j represents the measurement noise.

6. The method for online calibration of parameters of a micro-array polarization camera based on a star sensor according to claim 1, characterized in that: In the step (5), the state equations and measurement equations of the microarray polarization camera parameters in steps (1) and (4) are combined to establish an online calibration model of the microarray polarization camera parameters assisted by a star sensor; based on the characteristics of the model, an unscented Kalman filter method is used to online estimate the model parameters of each polarization unit in the microarray polarization camera, thereby improving the nighttime navigation accuracy and environmental adaptability of the polarization camera; Combining formulas (1) and (9), the online calibration model of the micro-array polarization camera model parameters based on star sensitivity assistance can be expressed as:

Images