On-orbit calibration method for micro-variations of intrinsic parameters in high-precision star sensor optical systems

Abstract

This invention relates to an on-orbit calibration method for the micro-variations of intrinsic parameters of a high-precision star sensor optical system, belonging to the field of high-precision on-orbit calibration technology for spacecraft. First, using the ground-calibrated intrinsic parameters of the optical system as initial values, the relationship between various coordinate systems and the angular distance model of the star sensor points is combined to establish the equality between the angle of the direction vector in the star sensor camera coordinate system and the angle of the navigation star in the celestial coordinate system. Then, based on this, and considering the characteristic that the angular distance between observed stars and the angular distance of the navigation star remain unchanged after environmental changes, a calibration model for the micro-variations of the intrinsic parameters of the star sensor optical system used in this invention is established. Finally, the extended Kalman filter (EKF) algorithm is used to calibrate the micro-variations of the intrinsic parameters of the optical system. After the micro-variation calibration is completed, the initial intrinsic parameter values ​​are added to obtain the intrinsic parameters of the optical system after environmental changes.

Description

Technical Field

[0001] This invention relates to the field of high-precision on-orbit calibration technology for spacecraft, and in particular to an on-orbit calibration method for micro-variations of internal parameters of a high-precision star sensor optical system. It is a method for on-orbit calibration of higher-precision internal parameters of the optical system of a star sensor during operation. Background Technology

[0002] A star sensor is a high-precision optical navigation device that uses stars as reference points and measures the attitude of spacecraft in space. It detects stars on the celestial sphere, combines high-precision astronomical coordinates of the stars with the intrinsic parameters of the star sensor's optical system (focal length, principal point) to perform attitude calculations, and possesses autonomous navigation capabilities. It is widely used in the aerospace field. A star sensor is an optical device that uses its optical system to image stars, obtaining an observed star map. Through star map preprocessing and centroid extraction, the image coordinates of the stars are obtained. Star map recognition algorithms are used to match star points in the observed star map with star points in the navigation star catalog. Based on the known high-precision celestial coordinates of the stars and the image coordinates of the centroids, combined with the intrinsic parameters of the star sensor's optical system, the three-axis attitude calculation of the star sensor's optical system coordinate system in the celestial coordinate system is achieved. Accurate intrinsic parameters of the optical system are one of the necessary conditions for ensuring high-precision attitude measurement during the operation of the star sensor.

[0003] Before a star sensor is put into use, its optical system parameters need to be calibrated on the ground. However, during launch and use, the star sensor's optical system is affected by environmental conditions such as vibration and thermal radiation, which increases aberrations and causes structural deformation, resulting in slight changes in the optical system parameters during actual use compared to the ground calibration. These slight changes in the internal parameters alter various calibration parameters of the star sensor, reducing its attitude measurement accuracy, especially since changes in the principal point position directly affect attitude measurement. Therefore, the slight changes in the internal parameters of the star sensor's optical system are one of the main factors limiting the accuracy of star sensor measurements. Traditional on-orbit calibration methods use the optical system's internal parameters themselves as the calibration object, resulting in a relatively coarse model. These slight changes are typically treated as random errors, leading to insufficient accuracy in the calibration results. Therefore, it is necessary to perform high-precision on-orbit calibration of the slight changes in the internal parameters during the star sensor's operation, and to compensate for these changes to ensure attitude measurement accuracy. Summary of the Invention

[0004] The purpose of this invention is to provide an on-orbit calibration method for micro-changes in the intrinsic parameters of a star sensor's optical system, solving the problem of the inability to handle micro-changes in existing star sensor on-orbit calibration techniques. This invention calibrates the micro-changes in the intrinsic parameters of a star sensor's optical system, primarily for determining the principal point position and actual imaging focal length. Using ground-calibrated intrinsic parameters as initial values, the invention adds the initial intrinsic parameter values ​​after calibrating the micro-changes to obtain the changed intrinsic parameters. Compared to traditional methods that directly calibrate intrinsic parameters, this invention enables more accurate and efficient calibration of the star sensor's optical system intrinsic parameters, especially for the principal point, thereby improving the accuracy of star sensor attitude calculation.

[0005] The above-mentioned objective of the present invention is achieved through the following technical solution:

[0006] An on-orbit calibration method for micro-variations of intrinsic parameters in a high-precision star sensor optical system includes the following steps:

[0007] Step 1: Perform ground calibration, and use the optical system intrinsic parameters obtained from the ground calibration as the initial optical system intrinsic parameter values;

[0008] Step 2: During the on-orbit operation of the star sensor, images of stars are obtained through the optical lens, and then star image preprocessing, barycenter extraction, and the position of the star's barycenter in the image are obtained.

[0009] Step 3: After obtaining the positions of the stars in the image, the captured star map is compared with the navigation star catalog for star map recognition. The navigation stars in the celestial coordinate system and the observed stars in the image coordinate system are obtained through the matching of star map recognition.

[0010] Step 4: Calibrate the minute changes in the parameters of the optical system.

[0011] Let V be the navigation star in the celestial coordinate system. i =(X i Y i Z i V j =(X j Y j Z j ), as in equation (1); the coordinates of the projection point of star i in the image coordinate system are (u i v i If the observed star is S, then the observed star is S. i =(u i v i ), S j =(u j v jThe direction vector W is obtained by inverting the position of the point in the camera coordinate system using the principal point and focal length of the observed star's standard optical system parameters. i and W j As shown in equation (2); where (u0, v0) represent the principal point coordinates, f represents the focal length, and α i α j With δ i δ j They represent right ascension and declination, respectively.

[0012]

[0013]

[0014] Based on the principle that the star angular distance remains unchanged under coordinate transformation, the coordinate system W of the star sensor camera is obtained. i and W j The angle between the direction vectors and the navigation star V in the celestial coordinate system i and V j The included angles are equal, as expressed in equation (3):

[0015] V i T V j =W i T W j (3)

[0016] Substituting equation (2) into equation (3), we can find the angular distance between the stars, as shown in equation (4);

[0017]

[0018] in,

[0019]

[0020] Suppose that the changes in the intrinsic parameters of the optical system after the environmental changes are as follows: As in equation (6)

[0021]

[0022] Where δu0, δv0, and δf are the micro-variables of the internal parameters of the optical system;

[0023] Suppose that due to changes in the parameters of the optical system after environmental changes, the position of the star points in the image changes as follows:

[0024]

[0025] Where, δu i ,δv i ,δu j ,δv jThis represents the change in the position of star points in the image;

[0026] The star distance after the environmental change is

[0027]

[0028] That is, the position of the star point and the parameters within the optical system change simultaneously, while the angular distance between the observed star and the angular distance between the navigation star remain unchanged;

[0029] in,

[0030]

[0031] From equation (8), we can obtain that

[0032]

[0033] Equation (10) uses a multivariate Taylor expansion to remove higher-order terms;

[0034] in,

[0035]

[0036]

[0037]

[0038] in,

[0039]

[0040] According to equation (10), it can be written as follows:

[0041]

[0042] When a captured star map contains n identified navigation stars, then:

[0043] M = A * [δu0δv0δf] T (16)

[0044] in,

[0045]

[0046] Step 5: The captured star image is processed using the Extended Kalman Filter algorithm to calibrate the minute changes in δu0, δv0, and δf (i.e., the principal point coordinates and focal length). Specifically:

[0047] Equation (15) is used as the measurement equation. Since this equation is nonlinear, it is calibrated using an extended Kalman filter. The state equation is:

[0048] xk =I 3×3 ·x k-1 (18)

[0049] x k =[δu0δv0δf] T (19)

[0050] Where x k The parameters δu0, δv0, and δf are the small variations of the principal point and focal length to be calibrated, as shown in equation (19); k-1 and k represent the (k-1)th and kth images, respectively, I 3×3 Given the identity matrix, the measurement equation is:

[0051] z k =h(x k )+n c (20)

[0052] Among them, z k The matrix is ​​formed by subtracting two star distances, which are the star distances calculated from the position vectors in the celestial coordinate system and the star distances calculated by back-projecting the transformed star points with the initial intrinsic parameter values, respectively; h(x k ) is A*[δu0δv0δf] T The simplified representation of n c The measurement error is caused by noise. Starting with the initial estimates of the stellar covariance P0 and parameter x0 obtained from ground calibration, the calibration is performed frame by frame. For the k-th star image, the EKF prediction equation is:

[0053]

[0054]

[0055] in Let Q be the prior estimate of the covariance at time k, and Q be the covariance matrix of the system process. The EKF update equation is:

[0056]

[0057]

[0058]

[0059] Where R is the covariance matrix of the observation noise, H k It is a Jacobian matrix.

[0060] The beneficial effects of this invention are as follows: By adding the micro-changes in the intrinsic parameters of the star sensor optical system calibrated using the method of this invention to the initial intrinsic parameter values ​​of the optical system obtained after ground calibration, the current intrinsic parameter values ​​of the star sensor optical system after environmental changes can be obtained. The accuracy of the intrinsic parameters of the star sensor optical system calibrated using the micro-change technique is higher than that based on the traditional star angular distance calibration method, especially in terms of the accuracy of the principal point, thereby improving the accuracy of the star sensor attitude information. Compared with the traditional on-orbit calibration model, the micro-change on-orbit calibration method used in this invention calibrates the intrinsic parameters of the star sensor optical system with higher accuracy, and the attitude accuracy is also greatly improved compared with the traditional on-orbit calibration model. The attitude residual is reduced, thus improving the accuracy of star sensor attitude calculation. Attached Figure Description

[0061] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate the invention and are used to explain it, but do not constitute an undue limitation of the invention.

[0062] Figure 1 This is a diagram showing the relationship between the camera coordinate system and the physical image coordinate system of the present invention;

[0063] Figure 2 This is a diagram showing the relationship between the pixel coordinate system and the physical image coordinate system of this invention;

[0064] Figure 3 This is a model diagram of the star point angular distance of the star sensor of the present invention;

[0065] Figure 4 This is a schematic diagram of the on-orbit operation principle of the star sensor of the present invention;

[0066] Figure 5 This is a schematic diagram illustrating the process of calibrating the internal parameters of the star sensor optical system of the present invention;

[0067] Figure 6 The simulation calibration results obtained using the method of this invention;

[0068] Figure 7 The attitude residual obtained using the method of this invention;

[0069] Figure 8 The simulation calibration results obtained using traditional on-orbit calibration methods;

[0070] Figure 9 This is to utilize the attitude residuals obtained using traditional on-orbit calibration methods. Detailed Implementation

[0071] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0072] See also Figures 1 to 5 As shown, the on-orbit calibration method for the micro-variations of the intrinsic parameters of the high-precision star sensor optical system of this invention first uses the ground-calibrated intrinsic parameters of the optical system as the initial values. Based on the relationships between various coordinate systems and the angular distance model of the star points of the star sensor, the relationship that the angle between the direction vectors in the star sensor camera coordinate system is equal to the angle between the navigation star in the celestial coordinate system is established. Then, based on the characteristic that the angular distance between observed stars and the angular distance between the navigation star remain unchanged after environmental changes, a calibration model for the micro-variations of the intrinsic parameters of the star sensor optical system used in this invention is established. Finally, the extended Kalman filter (EKF) algorithm is used to calibrate the micro-variations of the intrinsic parameters of the optical system. After the micro-variation calibration is completed, the initial intrinsic parameter values ​​of the optical system are added to obtain the intrinsic parameters of the optical system after environmental changes.

[0073] See also Figures 1 to 5 As shown, the on-orbit calibration method for micro-variations of internal parameters in a high-precision star sensor optical system according to the present invention includes the following steps:

[0074] Step 1: Perform ground calibration, and use the optical system intrinsic parameters obtained from the ground calibration as the initial optical system intrinsic parameter values;

[0075] Step 2: During the on-orbit operation of the star sensor, images of stars are obtained through the optical lens, and then star image preprocessing, barycenter extraction, and the position of the star's barycenter in the image are obtained.

[0076] Step 3: After obtaining the positions of stars in the image, the captured star map is compared with the navigation star catalog for star map recognition. The triangular star map recognition method is the most commonly used method for star map recognition. In the recognition process, the angular distance feature is most often used. The reference angular distance is obtained from the vector of the star in the prepared navigation star catalog in the celestial coordinate system. The angular distance feature to be recognized is calculated by combining the centroid coordinates of the star points in the image with the parameter values ​​of the initial optical system, thus completing the star map recognition. The navigation star in the celestial coordinate system and the observed star in the image coordinate system are obtained through the matching of star map recognition.

[0077] Step four: After obtaining the positions of the stars through star map recognition and matching, it is necessary to calibrate the parameters of the star sensor optical system and perform relevant calculations. This step involves the core technology of this invention, namely, the calibration of the minute changes in the parameters within the optical system:

[0078] The imaging process of a star sensor involves multiple coordinate systems, primarily the celestial coordinate system, camera coordinate system, image coordinate system, and pixel coordinate system. This invention mainly studies the calibration of micro-variations of parameters within the optical system of a star sensor, involving three coordinate systems. Figure 1 This relates the camera coordinate system to the physical image coordinate system. Figure 2 This represents the relationship between the pixel coordinate system and the physical image coordinate system.

[0079] Figure 1 China O C -X C Y C Z C Let o be the camera coordinate system, and o′-xy be the physical image coordinate system.

[0080] Figure 2 In the diagram, o′-xy represents the physical image coordinate system, and o-uv represents the pixel coordinate system.

[0081] This invention uses the angular distance, which is mostly used in current star sensor calibration methods, as the calibration reference, such as... Figure 3 The image shows the star point angular distance model for a star sensor, θ. ij It is the angular distance between stars i and j.

[0082] Let V be the navigation star in the celestial coordinate system. i =(X i Y i Z i V j =(X j Y j Z j ), as in equation (1); the coordinates of the projection point of star i in the image coordinate system are (u i v i If the observed star is S, then the observed star is S. i =(u i v i ), S j =(u j v j The direction vector W is obtained by inverting the position of the point in the camera coordinate system using the principal point and focal length of the observed star's standard optical system parameters. i and W j As shown in equation (2). Where (u0, v0) represent the principal point coordinates, f represents the focal length, and α... i α k With δ iδ j These represent right ascension and declination, respectively.

[0083]

[0084]

[0085] Based on the principle that the star angular distance remains unchanged under coordinate transformation, we can deduce that W in the star sensor camera coordinate system... i and W j The angle between the direction vectors and the navigation star V in the celestial coordinate system i and V j The included angles are equal, as expressed in equation (3):

[0086] V i T V j =W i T W j (3)

[0087] Substituting Equation 2 into Equation 3, we can find the angular distance between the stars, as shown in Equation (4).

[0088]

[0089] in,

[0090]

[0091] The method of this invention is to calibrate the micro-changes of the star sensor after environmental changes. Assuming the internal parameters of the optical system change after environmental changes, the position of the star point in the image will also change due to the change in the internal parameters of the optical system. Since the position of the star point and the internal parameters of the optical system change simultaneously, the angular distance of the observed star and the angular distance of the navigation star remain unchanged.

[0092] Suppose that the changes in the intrinsic parameters of the optical system after the environmental changes are as follows: As shown in equation (6).

[0093]

[0094] Where δu0, δv0, and δf are the micro-variables of the intrinsic parameters of the optical system.

[0095] Suppose that due to changes in the parameters of the optical system after environmental changes, the position of the star points in the image changes as follows:

[0096]

[0097] Where, δu i ,δv i ,δu j ,δv jThis represents the change in the position of star points in the image;

[0098] The star distance after the environmental change is

[0099]

[0100] That is, the position of the star point and the parameters within the optical system change simultaneously, while the angular distance between the observed star and the navigation star remains unchanged.

[0101] in,

[0102]

[0103] From equation (8), we can obtain that

[0104]

[0105] Equation (10) uses a multivariate Taylor expansion to remove higher-order terms.

[0106] in,

[0107]

[0108]

[0109]

[0110] in,

[0111]

[0112] According to equation (10), it can be written as follows:

[0113]

[0114] When a star map we take contains n identified navigation stars, then:

[0115] M = A * [δu0δv0δf] T (16)

[0116] in,

[0117]

[0118] Step 5: We use the Extended Kalman Filter (EKF) algorithm to calculate the star map and calibrate the minute changes in δu0, δv0, and δf, i.e., the principal point coordinates and focal length.

[0119] Equation (15) is used as the measurement equation. Since this equation is nonlinear, an extended Kalman filter (EKF) is used for calibration. The state equation is:

[0120] x k =I 3×3 ·x k-1 (18)

[0121] x k =[δu0δv0δf] T (19)

[0122] where x k The parameters δu0, δv0, and δf are the small variations of the principal point and focal length to be calibrated, as shown in equation (19). k-1 and k represent the (k-1)th and kth images, respectively. 3×3 Given the identity matrix, the measurement equation is:

[0123] z k =h(x k )+n c (20)

[0124] Among them, z k h(x) is the matrix formed by subtracting two angular distances, which are the angular distances calculated from the position vectors in the celestial coordinate system and the angular distances calculated by back-projecting the transformed star points with the initial intrinsic parameter values. k ) is A*[δu0δv0δf] T The simplified representation of n c The measurement error is caused by noise. Starting with the initial estimates of the stellar covariance P0 and parameter x0 obtained from ground calibration, the calibration is performed frame by frame. For the k-th star image, the EKF prediction equation is:

[0125]

[0126]

[0127] in Let Q be the prior estimate of the covariance at time k, and Q be the covariance matrix of the system process. The EKF update equation is:

[0128]

[0129]

[0130]

[0131] Where R is the covariance matrix of the observation noise, H k It is a Jacobian matrix.

[0132] By adding the micro-changes in the intrinsic parameters of the star sensor optical system calibrated using the method of this invention to the initial intrinsic parameter values ​​obtained after ground calibration, the current intrinsic parameter values ​​of the star sensor optical system after environmental changes can be obtained. The accuracy of the intrinsic parameters of the star sensor optical system calibrated by the micro-change technique of this invention is higher than that of the traditional star-angle distance calibration method, especially in terms of the accuracy of the principal point. Therefore, with the high accuracy of the principal point parameters obtained by the micro-change technique used in this invention, the accuracy of the star sensor attitude information can be improved.

[0133] Example:

[0134] In this embodiment, the evaluation criterion for the on-orbit calibration method of the star sensor optical system internal parameters is attitude residual: the attitude matrix and optical axis pointing are calculated using the star map simulation data used for calibration and the calibration results of the optical system internal parameters. The attitude matrix and optical axis pointing are compared with the attitude matrix and optical axis pointing calculated by the set standard parameters. The angle between the two optical axis pointing obtained in each image is calculated. The final evaluation index is the average of the angle between the two optical axis pointing in the last 100 images, i.e., attitude residual.

[0135] Simulations were performed using a simulated star chart. The values ​​of the intrinsic parameters of the optical system in the simulation data were set to f = 16.04, u0 = 900, and v0 = 500. The initial values ​​of the intrinsic parameters of the optical system were set to f = 16.00, u0 = 910, and v0 = 510, meaning the minute changes in the intrinsic parameters were: δf = +0.04, δu0 = -10, and δv0 = -10.

[0136] The simulation calibration results and attitude residuals are obtained using the micro-change on-orbit calibration method of this embodiment, such as... Figure 6 and Figure 7 As shown; simulation calibration results and attitude residuals are obtained using traditional on-orbit calibration methods, such as... Figure 8 and Figure 9 As shown.

[0137] When the average number of stars per frame is 15.1, the results of the optical system's intrinsic parameter calibration are shown in Table 1 and Table 2.

[0138] Table 1: Experimental Results of Calibration of Micro-Changes in Internal Parameters

[0139]

[0140] Table 2: Comparison of calibration experiment results

[0141]

[0142] As can be seen from Table 2, compared with the traditional on-orbit calibration method, the on-orbit calibration method for micro-variable intrinsic parameters used in this invention improves the accuracy of the intrinsic parameters of the optical system, reduces the attitude residual, and improves the accuracy of star sensor attitude calculation.

[0143] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made to the present invention should be included within the scope of protection of the present invention.

Claims

1. An on-orbit calibration method for micro-variations of intrinsic parameters in a high-precision star sensor optical system, characterized in that: Includes the following steps: Step 1: Perform ground calibration, and use the optical system intrinsic parameters obtained from the ground calibration as the initial optical system intrinsic parameter values; Step 2: During the on-orbit operation of the star sensor, images of stars are obtained through the optical lens, and then star image preprocessing, barycenter extraction, and the position of the star's barycenter in the image are obtained. Step 3: After obtaining the positions of the stars in the image, the captured star map is compared with the navigation star catalog for star map recognition. The navigation stars in the celestial coordinate system and the observed stars in the image coordinate system are obtained through the matching of star map recognition. Step 4: Calibrate the minute changes in the parameters of the optical system. Let V be the navigation star in the celestial coordinate system. i =(X i ,Y i Z i V j =(X j ,Y j Z j ), as in equation (1); the coordinates of the projection point of star i in the image coordinate system are (u i ,v i If the observed star is S, then the observed star is S. i =(u i ,v i ), S j =(u j ,v j The direction vector W is obtained by inverting the position of the point in the camera coordinate system using the principal point and focal length of the observed star's standard optical system parameters. i and W j As shown in equation (2); where (u0, v0) represents the principal point coordinates, f represents the focal length, and α i α j With δ i δ j They represent right ascension and declination, respectively. Based on the principle that the star angular distance remains unchanged under coordinate transformation, the coordinate system W of the star sensor camera is obtained. i and W j The angle between the direction vectors and the navigation star V in the celestial coordinate system i and V j The included angles are equal, as expressed in equation (3): V i T V j =W i T W j (3) Substituting equation (2) into equation (3), we can find the angular distance between the stars, as shown in equation (4); in, Suppose that the changes in the intrinsic parameters of the optical system after the environmental changes are as follows: As in equation (6) Where δu0, δv0, and δf are the micro-variables of the internal parameters of the optical system; Suppose that due to changes in the parameters of the optical system after environmental changes, the position of the star points in the image changes as follows: Where, δu i ,δv i ,δu j ,δv j This represents the change in the position of star points in the image; The star distance after the environmental change is That is, the position of the star point and the parameters within the optical system change simultaneously, while the angular distance between the observed star and the angular distance between the navigation star remain unchanged; in, From equation (8), we can obtain that Equation (10) uses a multivariate Taylor expansion to remove higher-order terms; in, in, According to equation (10), it can be written as follows: When a captured star map contains n identified navigation stars, then: M=A*[δu0 δv0 δf] T (16) in, Step 5: Use the extended Kalman filter algorithm to calculate the star map and calibrate the minute changes in δu0, δv0, and δf, i.e., the principal point coordinates and focal length.

2. The on-orbit calibration method for micro-variations of internal parameters in a high-precision star sensor optical system according to claim 1, characterized in that: The process of using the extended Kalman filter algorithm to calculate the star image and calibrate the minute changes in δu0, δv0, and δf (i.e., the principal point coordinates and focal length) is as follows: Equation (15) is used as the measurement equation. Since this equation is nonlinear, it is calibrated using an extended Kalman filter. The state equation is: x k =I 3×3 ·x k-1 (18) x k =[δu0 δv0 δf] T (19) Where x k The parameters δu0, δv0, and δf are the small variations of the principal point and focal length to be calibrated, as shown in equation (19); k-1 and k represent the (k-1)th and kth images, respectively, I 3×3 Given the identity matrix, the measurement equation is: z k =h(x k )+n c (20) Among them, z k The matrix is ​​formed by subtracting two star distances, which are the star distances calculated from the position vectors in the celestial coordinate system and the star distances calculated by back-projecting the transformed star points with the initial intrinsic parameter values, respectively; h(x k ) is A*[δu0δv0δf] T The simplified representation of n c The measurement error is caused by noise. Starting with the initial estimates of the stellar covariance P0 and parameter x0 obtained from ground calibration, the calibration is performed frame by frame. For the k-th star image, the EKF prediction equation is: in Let Q be the prior estimate of the covariance at time k, and Q be the covariance matrix of the system process. The EKF update equation is: Where R is the covariance matrix of the observation noise, H k It is a Jacobian matrix.

3. The on-orbit calibration method for micro-variations of internal parameters in a high-precision star sensor optical system according to claim 2, characterized in that: By adding the micro-changes in the intrinsic parameters of the calibrated star sensor optical system to the initial intrinsic parameter values ​​obtained after ground calibration, the current intrinsic parameter values ​​of the star sensor optical system after environmental changes can be obtained. In terms of the accuracy of the principal point, the accuracy of the intrinsic parameters of the star sensor optical system calibrated by the micro-change technique is higher than that of the traditional star angular distance calibration method, thereby improving the accuracy of the star sensor attitude information.

Images