A method for identifying and calibrating a gyro error characteristic parameter in orbit

Abstract

The application provides a gyro error characteristic parameter on-orbit identification and calibration method, comprising constructing an error measurement model of the gyro error characteristic parameter; using a high-precision geometric calibration field and high-precision GPS orbit data, based on a high-resolution remote sensing image strict geometric imaging model, calculating time sequence precise attitude angular velocity parameters of the earth camera; using star sensor attitude data to calibrate the drift error of the gyro on-orbit, namely constructing a combined attitude determination system state equation and a measurement equation, based on an extended Kalman filter to correct the drift error of the gyro; using the time sequence precise attitude angular velocity parameters of the earth camera and the error measurement model of the gyro, according to the least square principle, to calibrate the gyro installation error and scale factor error on-orbit. The application realizes effective identification and compensation of the gyro system error, realizes high-precision combined attitude determination of the star sensor and the gyro, and provides high-precision exterior azimuth angle elements for high-frequency linear array push-broom image geometric processing of the high-resolution remote sensing satellite.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of remote sensing satellite ground preprocessing, and particularly relates to a gyro error characteristic parameter on-orbit identification and calibration method. BACKGROUND

[0002] In order to guarantee that the effective load of a high-resolution earth observation satellite obtains high-precision images, the satellite attitude determination system is required to be able to provide high-precision satellite attitude information. At present, high-resolution remote sensing satellites widely adopt a satellite attitude determination system composed of a star sensor and a gyro, that is, the star sensor and the gyro are taken as measurement components, Kalman filter (KF) algorithm is adopted to process sensor measurement information in combination with satellite attitude kinematics equations, and high-precision attitude determination of the high-resolution remote sensing satellite is realized. As an important inertial attitude sensor, the gyro is affected by space mechanics environment, thermal environment, magnetic environment and other factors during the on-orbit operation of the satellite, the performance of internal components is degraded, drift, installation error, calibration factor error and other parameters based on ground measurement calibration change, and the measurement error is increased due to the influence of installation error and calibration accuracy and other factors, which reduces the satellite attitude determination accuracy. In the case that the above various errors cannot be measured on orbit, it is necessary to calibrate them on orbit. The existing various gyro calibration algorithms are difficult to comprehensively identify and calibrate various error characteristic parameters of the gyro, so that the high-precision combined attitude determination of the star sensor and the gyro cannot be realized. Therefore, it is necessary to solve the on-orbit identification and calibration of the gyro error characteristic parameters, guarantee the robustness of the high-precision combined attitude determination algorithm of the star sensor and the gyro, and realize high-precision earth observation of the high-resolution remote sensing satellite. SUMMARY

[0003] The application provides an on-orbit identification and calibration method for gyro error characteristic parameters, which is based on a high-precision geometric calibration field.

[0004] The technical scheme adopted by the application comprises the following steps:

[0005] Step 1, a gyro error characteristic parameter measurement model is constructed. The error characteristic parameters of the gyro mainly include gyro installation error, scale factor error, drift error and measurement noise.

[0006] Step 2, high-precision geometric calibration field and high-precision GPS orbit data are used to calculate the time sequence precise attitude angular velocity parameters of the earth camera based on a high-resolution remote sensing image strict geometric imaging model.

[0007] Step 3, the gyro drift error is calibrated on orbit by using the star sensor attitude data without considering the influence of the gyro installation error and the scale factor error, that is, the combined attitude determination system state equation and the measurement equation are constructed, and the gyro drift error is corrected based on the extended Kalman filter.

[0008] Step 4, according to the results of step 3 calibration, using the timing precision angular velocity parameters of the camera and the error measurement model of the gyro, the installation error and scale factor error of the gyro are calibrated in orbit according to the least square principle;

[0009] Moreover, in step 1, according to the gyro measurement principle and various error interference sources, the error measurement model of the gyro is constructed, and the specific implementation form is as follows:

[0010]

[0011]

[0012] Wherein, ω g And ω respectively represent the angular velocity of the satellite body coordinate system relative to the inertial coordinate system and the real angular velocity of the satellite body coordinate system relative to the inertial coordinate system; U, V represent the scale factor error and installation error of the gyro respectively; b,η g respectively represent the gyro drift error and measurement white noise; I is a unit matrix; u x ，u y ，u z is a scale factor error coefficient; Diag is a diagonal matrix; v yz ，v zy ,v xz ,v zx ,v xy ,v yx represents the installation error correction coefficient.

[0013] Moreover, the geometric calibration field in step 2 is the global distribution DEM and DOM data, and the accuracy of the GPS orbit is centimeter level, and the rigorous geometric imaging model of high-resolution image is as follows:

[0014]

[0015] Wherein, X,Y , Z ) WGS84 and x,y respectively represent the positions of the ground object points on the image in the object coordinate system and the image coordinate system respectively; x 0 ,y 0), f ,m and d x ,d y respectively represent the principal point, principal distance, scale factor and lens distortion parameters of the earth-oriented camera; X GPS Y GPS GPS D x D y D z respectively represent the object side coordinates of the earth-oriented camera's photographing center and the GPS eccentric error; respectively represent the rotation matrix from the J2000 coordinate system to the WGS84 coordinate system, the rotation matrix from the satellite body coordinate system to the J2000 coordinate system and the rotation matrix from the camera measurement coordinate system to the satellite body coordinate system.

[0016] Moreover, based on the high-frequency time series of attitude data obtained by scanning the line image of the earth-oriented camera, the geometric calibration field and the GPS precise orbit, the high-frequency time series of attitude angular velocity data is further calculated in step 2, and the specific implementation form is as follows:

[0017] Suppose the attitude quaternion of the earth-oriented camera at time t is t i , t i+1 the attitude quaternion of the earth-oriented camera at time t+Δt is , the sampling period of the line array camera is T = Δt t i+1 -t i According to the conversion relationship between Euler angles and quaternions, we get t i the Euler angle parameters at time t and time t+Δt are t i+1 , The conversion relationship between Euler angles and quaternions is as follows:

[0018]

[0019] wherein, Ψ represents the yaw angle, φ represents the roll angle, θ ​​​​​​​​​Let q0, q1, and q2 represent the pitch angle; q0, q1, and q2 are attitude quaternions, where q0 represents a scalar and q1, q2, and q3 represent vectors. Further, the attitude angular velocity parameters of the ground-based camera are obtained, specifically implemented as follows:

[0020]

[0021] Therefore, based on the above principles, the high-frequency attitude angular velocity parameters of the ground camera are obtained.

[0022] Furthermore, in step 3, the state of the system filter is set to the error quaternion vector and the gyroscope error drift. b x ,b y ,b z The error drift vector is represented here; the system's measurement equations, state equations, and filter design are as follows:

[0023] (1) System measurement equations:

[0024]

[0025]

[0026]

[0027] Where h1 and h2 represent the optical axis pointing functions of the two star sensors, respectively. Z k Represents the observed value; H k For measurement array; lz 1 b lz b 2 lz represents the true value of an orthogonal vector in the body coordinate system. mi 1 lz mi 2 This represents the measured value of an orthogonal vector in an inertial coordinate system. Z k Represents the observation vector. H k Represents the observation matrix. X k Represents state variables, This represents the optimal estimate of the quaternion. This represents the transformation matrix from the body to the inertial frame. V k express t k The noise sequence is observed at all times and satisfies the following relationship:

[0028]

[0029] in, R k Let the variance matrix of the measurement noise sequence be a positive definite matrix; V represents the Dikelical function; k and j represent different times, V j Indicates noise values ​​at different times;

[0030] (2) System state equations:

[0031]

[0032] in, F ( t ), W ( t ) represent the state transition matrix and system noise, respectively; X(t) represents the state variables, with a horizontal line indicating the derivative. 3x3 Represents the identity matrix; η b Indicates drift white noise; This represents the measured angular velocity value. Further discretization of the above equation yields:

[0033]

[0034] In the formula: for t k-1 Time to t k A shift in the formation at any moment; The system noise-driven array; W k-1 X is the system excitation noise sequence. k， X k-1 express t k time, t k-1 The state variable at any given time. Also... W k Satisfy the following equation:

[0035]

[0036] In the formula: Q k Let be the variance matrix of the system noise sequence, assumed to be a non-negative definite matrix; Represents the Dikelic function; W k W j This represents the system noise values ​​at different times;

[0037] (3) System filter design:

[0038]

[0039] wherein, K k denotes n×m a 2n-dimensional symmetric matrix, m ≤ n ); P k / k-1 denotes a 2n-dimensional symmetric matrix, m×n wherein, P k denotes a filter error variance matrix, denotes a state transition matrix, denotes a system noise driving matrix, Q k-1 denotes a system disturbance variance matrix. Furthermore, the calibration model of the gyro installation parameter error and the scale factor error of step 4 is specifically implemented as follows:

[0040]

[0041] wherein, ω the value of the gyro drift error b the value of the gyro drift error

[0042]

[0043] wherein, Z denotes an error angular velocity observation sequence, denotes an installation error and scale factor error coefficient observation matrix, X denotes an installation error and scale factor error coefficient, and V denotes a residual vector, and the optimal estimation value of the installation parameter error and the scale factor error is obtained according to the least square principle:

[0044] .

[0045] Beneficial effects:

[0046] The application provides a gyro error characteristic parameter on-orbit identification and calibration method, realizes effective identification and compensation of the gyro system error, and can effectively compensate the system error of the gyro through the method, realizes high-precision combined attitude determination of the star sensor and the gyro, and provides high-precision outer azimuth angle elements for high-frequency linear array push-broom image geometric processing of a high-resolution remote sensing satellite. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 It is a flow chart of the gyro error characteristic parameter on-orbit identification and calibration method.

[0048] Figure 2 Flow chart of time sequence precise attitude angle velocity parameter inversion of the camera based on geometric calibration field of the application. DETAILED DESCRIPTION

[0049] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0050] As shown in the flow chart of the on-orbit identification and calibration method of the gyro error characteristic parameters of the present application, the present application comprises the following specific steps: Figure 1 Step 1, according to the working principle of the gyro and the on-orbit running environment, a measurement model of the error characteristic parameters of the gyro is constructed, and the specific error characteristic parameters include the drift error of the gyro, the measurement white noise, the installation error of the gyro and the scale factor error. The specific implementation form is as follows:

[0051]

[0052] (1)

[0053] (2)

[0054] wherein, ω g and ω respectively represent the angular velocity of the satellite body coordinate system relative to the inertial coordinate system and the real angular velocity of the satellite body coordinate system relative to the inertial coordinate system; U, V respectively represent the scale factor error and the installation error of the gyro; b,η g respectively represent the gyro drift error and the measurement white noise; I is a unit matrix; u x ，u y ，u z is a scale factor error coefficient; Diag is a diagonal matrix; v yz ，v zy ,v xz ,v zx ,v xy ,v yx represents the installation error correction coefficient.

[0055] ​Step 2, as shown in Figure 2 Based on high-precision geometric calibration field and GPS precise orbit, the timing precise attitude angular velocity parameters of the earth camera are calculated, and the mathematical model used is the rigorous geometric imaging model of high-resolution remote sensing image, and the specific implementation form is as follows:

[0056] (3)

[0057] Wherein, X,Y , Z ) WGS84 and x,y respectively represent the positions of the ground object points on the image in the object coordinate system and the image coordinate system respectively; x ,y , f , m and d x ,d y respectively represent the principal point, principal distance, scale factor and lens distortion parameters of the earth camera; X GPS , Y GPS , GPS ) and D x , D y , D z respectively represent the object coordinate of the photographing center of the earth camera and the GPS eccentric error; , , respectively represent the rotation matrix from J2000 coordinate system to WGS84 coordinate system, the rotation matrix from satellite body coordinate system to J2000 coordinate system and the rotation matrix from camera measurement coordinate system to satellite body coordinate system. Based on the above algorithm principle, the following sub-steps are further included:

[0058] Step 2.1, GPS precise orbit calculation: based on high-precision GPS observation data and orbit dynamics model, using post-precise orbiting mathematical model, the high-precision GPS precise orbit of the earth observation remote sensing satellite can be calculated, and the orbit precision can reach centimeter level.

[0059] Step 2.2, geometric calibration field data adopts Henan Anyang, Songshan geometric calibration field in China and foreign calibration field, so as to realize the inversion of several orbit period precise attitude timing data of the earth camera; geometric calibration field data includes DOM and digital elevation model data.

[0060] ​Step 2.3, high-precision image matching and control point measurement: high-precision dense matching is performed on the image data obtained in step 2.2 and the digital orthographic image map (DOM) data of the geometric calibration field for a single actual shooting of the earth camera, and the object coordinates of the control points are obtained according to the digital elevation model data of the geometric calibration field.

[0061] Step 2.4, calculating the precise attitude of the earth camera in multiple orbits based on the strict geometric imaging model of high-resolution remote sensing images. The inverted precise attitude of the multi-orbit earth camera can reach sub-arcsecond level based on the following conditions: (1) the precision of the high-resolution satellite GPS precise orbit data is centimeter level; (2) the planar precision of the control points of the geometric calibration field is better than 1 meter, and the elevation precision is better than 2 meters; (3) the earth camera is calibrated with high precision on orbit, realizing the 0.3-pixel calibration of the camera pixel pointing angle; (4) solving the time synchronization problem of the earth camera and the attitude orbit measurement and control system, and the time synchronization error is only microsecond level.

[0062] Step 2.5, calculating the time sequence high-precision attitude angular velocity parameters of the earth camera. Assuming t i The attitude quaternion of the earth camera at time t is , t i+1 The attitude quaternion of the earth camera at time t is , and the sampling period of the earth camera is T = t i+1 -t i According to the conversion relationship between Euler angles and quaternions, we get t i The Euler angle parameters of the earth camera at time t and t i+1 The Euler angle parameters of the earth camera at time t and , The conversion relationship between Euler angles and quaternions is as follows:

[0063] (4)

[0064] wherein, Ψ represents the yaw angle (Yaw Angle), represents φ the roll angle (Roll Angle), θ and represents the pitch angle (Pitch Angle); q0, q1, q2, q3 are attitude quaternions, q0 represents a scalar, and q1, q2, q3 represent vectors;

[0065] Further, the attitude angular velocity parameters of the earth camera are obtained, and the specific implementation form is as follows:

[0066] (5)

[0067] According to the above principle, high-precision attitude angular velocity parameters of the camera time sequence relative to the earth are obtained.

[0068] Step 3, without considering the gyro installation error and the scale factor error, based on the gyro error characteristic parameter measurement model constructed in step 1, the optimal estimation of the gyro drift error is realized by using the star sensor and gyro combined attitude determination filter algorithm, and the state of the system filter is set as the error quaternion vector and the gyro error drift x y z , which represents the error drift vector. The specific process includes the following sub-steps:

[0069] Step 3.1, the system measurement equation is constructed based on the star sensor observation value:

[0070] (6)

[0071] (7)

[0072] (8)

[0073] Wherein, h1, h2 represent the optical axis pointing functions of the two star sensors respectively; represents the quaternion vector differential; Z k represents the observation value; H k is the measurement matrix; lz 1 b b 2 represents the real value of the orthogonal vector in the body coordinate system, lz mi 1 mi 2 represents the measured value of the orthogonal vector in the inertial coordinate system Z k represents the observation vector, H k represents the observation matrix, X k represents the state variable, represents the optimal estimation value of the quaternion, represents the conversion matrix from the body to the inertial system, V k represents t k the observation noise sequence at the moment; and satisfies the following relationship:

[0074] ​​​​​

[0075] in, R k The variance matrix of the measurement noise sequence is assumed to be a positive definite matrix; V represents the Dikelical function; k and j represent different times, V j This represents the noise value at different times.

[0076] Step 3.2, construct the system state equations based on the gyroscope measurement model:

[0077] (9)

[0078] in, F ( t ), W ( t ) represent the state transition matrix and system noise, respectively; X(t) represents the state; and the horizontal line represents the derivative. 3X3 Represents the identity matrix; η b Indicates drift white noise; This represents the measured angular velocity value; further discretization of the above equation yields:

[0079] (10)

[0080] In the formula: for t k-1 Time to t k A shift in the formation at any moment; The system noise-driven array; W k-1 X is the system excitation noise sequence. k， X k-1 express t k time, t k-1 The state variable at any given time. Also... Satisfy the following equation:

[0081]

[0082] In the formula: Q k Let be the variance matrix of the system noise sequence, assumed to be a non-negative definite matrix; Represents the Dikelic function; W k W j This represents the system noise value at different times.

[0083] Step 3.3: Design the attitude estimator for the combined attitude determination system to achieve optimal estimation of the gyroscope drift error parameters. The specific implementation is as follows:

[0084] One-step prediction equation for system state:

[0085] (11)

[0086] State estimation equation:

[0087] (12)

[0088] Filter gain equation:

[0089] (13)

[0090] One-step prediction mean square error:

[0091] (14)

[0092] Estimate mean square error:

[0093] (15)

[0094] (16)

[0095] in, K k express n×m 3D gain matrix ( m ≤ n ); P k / k-1 Let Variance be the prediction error variance matrix from time k-1 to time k. m×n 2D symmetric matrix; P k Represents the filter error variance matrix; Represents the state transition matrix. Indicates the system noise-driven array, Q k-1 This represents the system interference variance matrix. Based on the above filtering design, the optimal estimation of gyroscope drift error can be achieved.

[0096] Step 4: Based on the calibration results in Step 3, using the time-series precise angular velocity parameters of the Earth-based camera and the error measurement model of the gyroscope, on-orbit calibration of the gyroscope's installation error and scaling factor error is achieved according to the least squares principle.

[0097] The parameter calibration is based on the earth camera. The calibration model of the gyro installation parameter error and the scale factor error is specifically implemented as follows:

[0098] (17)

[0099] wherein, ω The value is provided by the earth camera, and the gyro drift error b The value is obtained by the calibration result of step 3.

[0100] According to the multi-orbit period time sequence k =1, 2,... N The precise attitude angular velocity parameter of the earth camera, the attitude angular velocity parameter of the gyro output and the gyro drift error parameter obtained by step 3 are jointly constructed into an observation equation, and the specific form is as follows:

[0101] (18)

[0102] wherein, Z The error angular velocity observation sequence is represented by The installation error and the scale factor error coefficient observation matrix are represented by X The installation error and the scale factor error coefficient are represented by V, and the residual error vector is represented by According to the least square principle, the optimal estimation value of the installation parameter error and the scale factor error is obtained:

[0103] (19)

[0104] Based on the above steps, the on-orbit identification and calibration of the gyro error characteristic parameters configured on the satellite are performed, so as to ensure the robustness of the star sensor and the gyro combined attitude determination system.

[0105] The specific examples described in the present text are merely illustrative of the spirit of the present application. Those skilled in the art to which the present application belongs can make various modifications or supplements to the described specific examples or replace them with similar ways, without deviating from the spirit of the present application or exceeding the scope defined by the appended claims.

Claims

1. A method for on-orbit identification and calibration of a gyro error characteristic parameter, characterized in that, The system error of the gyro is compensated, high-precision combined attitude determination of the star sensor and the gyro is realized, high-precision outer azimuth elements are provided for high-frequency linear array push-broom image geometric processing of a high-resolution remote sensing satellite, and the following steps are included. Step 1, an error measurement model of gyro error characteristic parameters is constructed, the gyro error characteristic parameters include: gyro installation error, scale factor error, drift error and measurement noise; Step 2, time sequence precise attitude angular velocity parameters of the earth camera are calculated based on a rigorous geometric imaging model of high-resolution remote sensing images by using a high-precision geometric calibration field and high-precision GPS orbit data; Based on the scanning line images of the earth camera, the high-precision geometric calibration field and the high-precision GPS orbit inversion high-frequency time sequence attitude data, high-frequency time sequence attitude angular velocity parameters of the earth camera are solved, and the specific implementation form is as follows: Assume that the attitude quaternion of the camera to the ground at time t i is , the attitude quaternion of the camera to the ground at time t i+1 is , the sampling period of the camera to the ground is T=t i+1 -t i , the Euler angle parameters of the camera to the ground at time t i and time t i+1 are obtained according to the conversion relationship between Euler angles and quaternions , , and the conversion relationship between Euler angles and quaternions is as follows: ； Wherein, Ψ represents the yaw angle, φ represents the roll angle, and θ represents the pitch angle; q0, q1, q2 and q3 are attitude quaternions, q0 represents a scalar, and q1, q2 and q3 represent vectors; The attitude angular velocity parameters of the camera relative to the ground are calculated: ; Step 3, without considering the influence of the gyro installation error and the scale factor error, the drift error of the gyro is calibrated on orbit by using the attitude data of the star sensor, that is, the combined attitude determination system state equation and the measurement equation are constructed, and the drift error of the gyro is corrected based on the extended Kalman filter; The state of the Kalman filter is set as an error quaternion vector and gyro error drift , b x , b y , b z represents the error drift vector; q1, q2, q3represents the error quaternion vector; the measurement equation, state equation and filter design of the system are as follows: (1) system measurement equation: ； ； ； ； where h1, h2 represent the optical axis pointing function of the two star sensors respectively; represents the quaternion vector differential; Z k represents the observation value; H k is the measurement matrix; lz 1 b , lz b 2 represents the true value of the orthogonal vector in the body coordinate system, lz mi 1 , lz mi 2 represents the measured value of the orthogonal vector in the inertial coordinate system, Z k represents the observation vector, H k represents the observation matrix, X k represents the state variable, represents the optimal estimation value of the quaternion, represents the conversion matrix from the body to the inertial system, V k represents the observation noise sequence at time t k ; and satisfies the following relationship: ； wherein R k is the variance matrix of the measurement noise sequence, assumed to be positive definite; A represents the Dirac function; k, j represent different times; V j represents the noise values at different times; (2) system state equation: ； where F(t), W(t) represent state transition matrix and system noise respectively, X(t) represents state, and a horizontal line above a symbol represents derivative; I 3X3 represents unit matrix; η b represents drift white noise; represents angular velocity measurement value; further discretization of the above formula is obtained: ； In the formula: For t k-1 Time to t k A shift in the formation at any moment; The system noise-driven array; W k-1 Let X be the system excitation noise sequence. k， X k-1 Indicates t k Time, t k-1 The state variable at time, and W at the same time. k Satisfy the following equation: ； where: Q k is the variance matrix of the system noise sequence, which is assumed to be non-negative definite; W k , W j denotes the system noise values at different times; (3) system filter design: ； where K k represents an n x m dimensional gain matrix (m < n); P k / k-1 represents a prediction error variance matrix from k-1 to k, which is an m x n dimensional symmetric matrix; P k represents a filtering error variance matrix; represents a state transition matrix, represents a system noise driving matrix, Q k-1 represents a system disturbance variance matrix; Step 4, according to the on-orbit calibration result of step 3, the time sequence precise attitude angular velocity parameters of the earth camera and the error measurement model of the gyro are used, and the gyro installation error and the scale factor error are calibrated on orbit according to the least square principle; The calibration model of the gyro installation parameter error and the scale factor error has the specific implementation form as follows: ； Wherein, ω is provided by the earth camera, the gyro drift error b is the result of the on-orbit calibration of step 3, and the observation equation is further constructed, and the specific form is as follows: ； where Z represents the error angular velocity observation sequence, represents the installation error and scale factor error coefficient observation matrix, X represents the installation error and scale factor error coefficient, V represents the residual vector, and the optimal estimation value of the installation parameter error and the scale factor error is obtained according to the least square principle: .

2. The method of claim 1, wherein the on-orbit identification and calibration of the gyro error characteristic parameters is characterized by: In step 1, the error measurement model of the gyro is constructed according to the gyro measurement principle and the error interference sources, and the specific implementation form is as follows: ； ； where ω g and ω represent the gyro output of the satellite body coordinate system relative to the inertial coordinate system angular velocity and the real satellite body coordinate system relative to the inertial coordinate system angular velocity, respectively; U, V represent the scale factor error and installation error of the gyro, respectively; b, η g represent the gyro drift error and measurement white noise, respectively; I is a unit matrix; u x , u y , u z is a scale factor error coefficient; Diag is a diagonal matrix; v yz , v zy , v xz , v zx , v xy , v yx represents the installation error correction coefficient.

3. The method of claim 1, wherein the on-orbit identification and calibration of the gyro error characteristic parameters is characterized by: In step 2, the high-precision geometric calibration field is global distribution DEM and DOM data, the accuracy of the high-precision GPS orbit is centimeter level, and the rigorous geometric imaging model of the high-resolution remote sensing image has the specific implementation form as follows: ； wherein (X, Y, Z) WGS84 (x, y) represent the positions of the ground object points on the image in the object coordinate system and the image coordinate system, respectively; (x0, y0), f, m, and (d x ,d y ) represent the principal point, the principal distance, the scale factor, and the lens distortion parameter of the terrestrial camera, respectively; (X GPS ,Y GPS ,Z GPS ) and (D x ,D y ,D z ) represent the object coordinate of the terrestrial camera center and the GPS eccentric error, respectively; , , represent the rotation matrix from the J2000 coordinate system to the WGS84 coordinate system, the rotation matrix from the satellite body coordinate system to the J2000 coordinate system, and the rotation matrix from the camera measurement coordinate system to the satellite body coordinate system, respectively.

Images