Two-dimensional turntable on-orbit calibration method based on multi-source error fusion

Abstract

The invention relates to the technical field of aerospace, in particular to a multi-source error fusion two-dimensional turntable on-orbit calibration method. The embodiment of the invention provides a multi-source error fusion two-dimensional turntable on-orbit calibration method. The method comprises the following steps: establishing an error model about a camera optical axis actual value, a full-link error and a measured value; defining generalized error parameters based on the error model, and performing linearization processing on the error model according to the defined generalized error parameters to obtain a linear model; on the basis of a set error parameter initial value, the linear model carries out recursive operation through a recursive algorithm, and an error parameter estimated value is obtained through recursive operation in each sampling period. The embodiment of the invention provides a multi-source error fusion two-dimensional turntable on-orbit calibration method, which ensures the pointing precision of the visual axis of a camera during on-orbit operation.

Description

Technical Field

[0001] The invention relates to the field of aerospace technology, and in particular to an on-orbit calibration method for a two-dimensional turntable with multi-source error fusion. Background Art

[0002] The two-dimensional turntable is increasingly widely used in the rapid discovery, tracking and identification of moving targets in the active and mid-stage of space due to its high precision and high agility. Satellites equipped with a two-dimensional turntable can maintain the independence of the attitude control of the satellite while performing tracking tasks to meet the working requirements of other payloads on the satellite. In order to achieve accurate tracking of the target, the camera's line of sight is often required to have a high pointing accuracy, and the system error link between the camera and the turntable, and the turntable and the satellite coordinate system will directly affect the accuracy. Although the installation parameters are precisely measured and calibrated on the ground before the satellite enters orbit, the turntable and the camera will be subjected to huge vibration and impact during the launch of the spacecraft. At the same time, the on-orbit space environment and the ground environment are quite different in terms of gravity, which will inevitably cause changes in the system structure parameters. Summary of the invention

[0003] The embodiment of the present invention provides a two-dimensional turntable on-orbit calibration method with multi-source error fusion to ensure the pointing accuracy of the camera's visual axis when it is running on-orbit.

[0004] The embodiment of the present invention provides a two-dimensional turntable on-orbit calibration method with multi-source error fusion, comprising: Establish an error model about the actual value of the camera optical axis, the error of the entire link, and the measured value; Defining generalized error parameters based on the error model, and linearizing the error model according to the defined generalized error parameters to obtain a linear model; Based on the set initial value of the error parameter, the linear model performs recursive operation through a recursive algorithm, and an error parameter estimation value is obtained through the recursive operation in each sampling period.

[0005] In a possible design, the step of establishing an error model regarding a camera optical axis pre-adjustment value, an error value, and a measurement value includes: Establish the pointing vector formula of the camera optical axis; Decomposing the image plane position error into a first azimuth error and a first pitch error, and obtaining a plane transformation matrix according to the first azimuth error and the first pitch error; Decomposing the camera optical axis pointing error into a second azimuth error and a second pitch error, and obtaining a camera optical axis transformation matrix according to the second azimuth error and the second pitch error; Decomposing the optical axis verticality error into a third position error and a third pitch error, and obtaining a verticality transformation matrix according to the third position error and the third pitch error; A pitch transformation matrix is ​​established according to the actual value of the pitch angle, the pitch axis measurement value, the pitch axis encoding zero position error, the pitch axis encoder measurement error, the verticality errors of the two pitch axis systems and the rotation errors of the two pitch axis systems; An azimuth transformation matrix is ​​established according to the actual value of the azimuth angle, the measured value of the azimuth axis, the zero position error of the azimuth axis encoder, the measured error of the azimuth axis encoder, the verticality error of the two azimuth axis systems and the rotation error of the two azimuth axis systems; Decomposing the attitude error into a fourth azimuth error, a fourth pitch error and a roll error, and obtaining an attitude transformation matrix according to the fourth azimuth error, the fourth pitch error and the roll error; The error model is obtained according to the vector formula, the plane transformation matrix, the camera optical axis transformation matrix, the verticality transformation matrix, the pitch transformation matrix, the azimuth transformation matrix and the attitude transformation matrix.

[0006] In one possible design, the vector formula is as follows: is the actual pointing vector of the camera optical axis. is the initial vector of the camera optical axis, is the transformation matrix, the transformation matrix , It represents the transformation matrix between coordinate systems, b0 is the star system, b is the actual star system, A0 is the turntable azimuth coordinate system, A is the azimuth motion coordinate system, B0 is the pitch frame coordinate system, B is the pitch motion coordinate system, P0 is the ideal optical axis pointing coordinate system, P is the actual optical axis pointing coordinate system, Cam is the actual pointing system in the image plane coordinate system, I It is the inertial system of J2000 satellite.

[0007] In one possible design, the plane transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the first azimuth error and the first pitch error respectively; The camera optical axis transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the second azimuth error and the second pitch error respectively; The verticality transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and They are the third position error and the third pitch error respectively; The pitch transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the pitch angle, where is the pitch axis measurement value, is the zero position error of the pitch axis encoder, is the encoder measurement error, , is the verticality error of the pitch axis system, , is the pitch axis rotation error; The orientation transformation matrix is ​​as follows: In the formula, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the azimuth angle, is the azimuth axis measurement value, is the azimuth axis rotation error, where Azimuth zero error, is the encoder measurement error, , is the verticality error of the azimuth axis system, , is the azimuth axis rotation error; The posture transformation matrix is ​​as follows: in, represents the transformation matrix obtained after rotating around the m-axis by an angle n, and the fourth azimuth error, the fourth pitch error and the roll error are respectively , and .

[0008] In one possible design, the error model is as follows: in, is the camera optical axis measurement value, express The i-th dimension element of , i=1, 2, 3.

[0009] In one possible design, the generalized error parameter is as follows: .

[0010] In one possible design, the linear model is as follows: in, is a recursive matrix consisting only of the turntable measurement angles.

[0011] In a possible design, based on the set initial value of the error parameter, the linear model is recursively calculated by a recursive algorithm, and an error parameter estimation value is obtained by recursive calculation in each sampling period, including: Define variables: , is the camera optical axis measurement value, and Represents the measured value y and recursive matrix of the kth sampling period , use the following recursive algorithm to calculate the error parameters of each sampling period: in, is the estimated value of the error parameter θ, the initial value of which is determined based on the system ground precision measurement parameters combined with experiments, , , c and is an adjustable constant.

[0012] Compared with the prior art, the present invention has at least the following beneficial effects: Compared with the prior art, the technical method adopted in this application proposes an online autonomous calibration method of multi-source error fusion, which fully exploits the error dynamic characteristics of the camera-two-dimensional turntable-satellite full link, fuses the multi-source error into the system's generalized parameter linearization error, and uses the recursive gradient method to identify the generalized error parameters online. The entire algorithm is simple in design, with a small workload for parameter debugging, no need to add additional data input, and simple calculation. The algorithm can be adapted to a large class of military and civilian satellite systems with similar on-orbit calibration requirements, and has strong engineering practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 is a turntable angle curve diagram provided by an embodiment of the present invention; Figure 2 is an error online calibration result diagram provided by an embodiment of the present invention; Figure 3 It is a comparison diagram of an online error calibration result and a real error provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, 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 part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0016] In the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance; unless otherwise specified or explained, the term "multiple" refers to two or more; the terms "connected" and "fixed" should be understood in a broad sense, for example, "connected" can be a fixed connection, a detachable connection, an integral connection, or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0017] In the description of this specification, it should be understood that the directional words such as "upper" and "lower" described in the embodiments of the present invention are described at the angles shown in the drawings and should not be understood as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is connected to another element "upper" or "lower", it can not only be directly connected to another element "upper" or "lower", but also indirectly connected to another element "upper" or "lower" through an intermediate element.

[0018] The embodiment of the present invention provides a two-dimensional turntable on-orbit calibration method with multi-source error fusion, comprising: Establish an error model about the actual value of the camera optical axis, the error of the entire link, and the measured value; Defining generalized error parameters based on the error model, and linearizing the error model according to the defined generalized error parameters to obtain a linear model; Based on the set initial value of the error parameter, the linear model performs recursive operation through a recursive algorithm, and an error parameter estimation value is obtained through the recursive operation in each sampling period.

[0019] Compared with the prior art, the technical method adopted in this application proposes an online autonomous calibration method of multi-source error fusion, which fully exploits the error dynamic characteristics of the camera-two-dimensional turntable-satellite full link, fuses the multi-source error into the system's generalized parameter linearization error, and uses the recursive gradient method to identify the generalized error parameters online. The entire algorithm is simple in design, with a small workload for parameter debugging, no need to add additional data input, and simple calculation. The algorithm can be adapted to a large class of military and civilian satellite systems with similar on-orbit calibration requirements, and has strong engineering practicality.

[0020] In some embodiments of the present invention, the step of establishing an error model regarding the camera optical axis pre-adjustment value, the error value, and the measured value includes: Establish the pointing vector formula of the camera optical axis; Decomposing the image plane position error into a first azimuth error and a first pitch error, and obtaining a plane transformation matrix according to the first azimuth error and the first pitch error; Decomposing the camera optical axis pointing error into a second azimuth error and a second pitch error, and obtaining a camera optical axis transformation matrix according to the second azimuth error and the second pitch error; Decomposing the optical axis verticality error into a third position error and a third pitch error, and obtaining a verticality transformation matrix according to the third position error and the third pitch error; A pitch transformation matrix is ​​established according to the actual value of the pitch angle, the pitch axis measurement value, the pitch axis encoding zero position error, the pitch axis encoder measurement error, the verticality errors of the two pitch axis systems and the rotation errors of the two pitch axis systems; An azimuth transformation matrix is ​​established according to the actual value of the azimuth angle, the measured value of the azimuth axis, the zero position error of the azimuth axis encoder, the measured error of the azimuth axis encoder, the verticality error of the two azimuth axis systems and the rotation error of the two azimuth axis systems; Decomposing the attitude error into a fourth azimuth error, a fourth pitch error and a roll error, and obtaining an attitude transformation matrix according to the fourth azimuth error, the fourth pitch error and the roll error; The error model is obtained according to the vector formula, the plane transformation matrix, the camera optical axis transformation matrix, the verticality transformation matrix, the pitch transformation matrix, the azimuth transformation matrix and the attitude transformation matrix.

[0021] In some embodiments of the present invention, the vector formula is as follows: is the actual pointing vector of the camera optical axis. is the initial vector of the camera optical axis, is the transformation matrix, the transformation matrix , It represents the transformation matrix between coordinate systems, b0 is the star system, b is the actual star system, A0 is the turntable azimuth coordinate system, A is the azimuth motion coordinate system, B0 is the pitch frame coordinate system, B is the pitch motion coordinate system, P0 is the ideal optical axis pointing coordinate system, P is the actual optical axis pointing coordinate system, Cam is the actual pointing system in the image plane coordinate system, I It is the inertial system of J2000 satellite.

[0022] In this embodiment, the initial direction of the camera optical axis can be set to , the above vector formula is the actual pointing vector of the sensor after azimuth and pitch rotation under the influence of system error.

[0023] In this embodiment, the specific directions of the coordinate axes in each coordinate system are as follows: (1) Star system (b0), including: The axis is along the satellite's structural longitudinal axis and points to the direction of the satellite's flight. The axis is perpendicular to the longitudinal axis and points to the center of the earth. Axis and axis, The axis forms a right-hand system; the actual star system (b system) The actual direction of the axis coincides with the longitudinal axis of the satellite structure.

[0024] (2) Turntable azimuth coordinate system (A0), The actual direction of the axis coincides with the azimuth axis. The axis is along the zero direction of the azimuth encoder; the azimuth motion coordinate system (A system) is fixedly connected to the azimuth frame and is used to describe the azimuth axis motion. The axis is along the instantaneous axis direction of the azimuth encoder.

[0025] (3) Turntable pitch coordinate system, including: pitch frame coordinate system (B0 system), The axis coincides with the actual direction of the pitch axis after azimuth rotation, and the axis is along the zero position direction of the pitch encoder; the pitch motion coordinate system (B system) is fixedly connected to the pitch frame and is used to describe the motion of the pitch axis, and the axis is along the instantaneous axis direction of the pitch encoder.

[0026] (4) Optical axis coordinate system, including: ideal optical axis pointing coordinate system (P0 system), The axis coincides with the ideal direction of the optical axis after azimuth and elevation rotation, and coincides with the F system when there is no optical axis verticality error. The actual optical axis points to the coordinate system (P system), The axis actually coincides with the camera optical axis.

[0027] (5) Image plane coordinate system, including: actual pointing system (Cam system), The axis points to the actual target and coincides with the optical axis coordinate system in the absence of image plane position error.

[0028] In some embodiments of the present invention, the plane transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the first azimuth error and the first pitch error respectively; The camera optical axis transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the second azimuth error and the second pitch error respectively; Since the turntable adopts a two-degree-of-freedom structure, the camera optical axis system only has an optical axis verticality error relative to the turntable pitch axis system. The verticality transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and They are the third position error and the third pitch error respectively; The pitch transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the pitch angle, where is the pitch axis measurement value, is the zero position error of the pitch axis encoder, is the encoder measurement error, , is the verticality error of the pitch axis system, , is the pitch axis rotation error; The orientation transformation matrix is ​​as follows: In the formula, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the azimuth angle, is the azimuth axis measurement value, is the azimuth axis rotation error, where Azimuth zero error, is the encoder measurement error, , is the verticality error of the azimuth axis system, , is the azimuth axis rotation error; Since there is an error between the actual attitude of the satellite and the theoretical attitude, the actual initial pointing of the camera also has a deviation from the ideal initial pointing, and there is an orientation error between them. , pitch error and roll error , the posture transformation matrix is ​​as follows: in, represents the transformation matrix obtained after rotating around the m-axis by an angle n, and the fourth azimuth error, the fourth pitch error and the roll error are respectively , and .

[0029] In some embodiments of the present invention, the error model is as follows: in, is the camera optical axis measurement value, express The i-th dimension element of , i=1, 2, 3.

[0030] In some embodiments of the present invention, the generalized error parameter is as follows: .

[0031] In some embodiments of the invention, the linear model is as follows: in, is a recursive matrix consisting only of the turntable measurement angles.

[0032] In some embodiments of the present invention, the linear model is recursively calculated by a recursive algorithm based on the set initial value of the error parameter, and an error parameter estimation value is obtained by recursive calculation in each sampling period, including: Define variables: , is the camera optical axis measurement value, and Represents the measured value y and recursive matrix of the kth sampling period , use the following recursive algorithm to calculate the error parameters of each sampling period: in, is the estimated value of the error parameter θ, the initial value of which is determined based on the system ground precision measurement parameters combined with experiments, , , c and is an adjustable constant.

[0033] In order to more clearly illustrate the effect of the present application, a simulation verification is performed according to the method provided in the present application. The initial mode of the satellite is set to the normal ground-to-earth mode, that is, the three-axis zero attitude of the satellite to the ground, and the rotation angle curve of the turntable is as follows: Figure 1 As shown, the online calibration structure is as follows Figure 2~Figure 3 As shown in the figure, it can be seen that after the recursive error online identification of the present application, the estimated value can be made close to the actual error.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A two-dimensional turntable on-orbit calibration method with multi-source error fusion, characterized in that: include: Establish an error model about the actual value of the camera optical axis, the error of the entire link, and the measured value; Defining generalized error parameters based on the error model, and linearizing the error model according to the defined generalized error parameters to obtain a linear model; Based on the set initial value of the error parameter, the linear model performs recursive operation through a recursive algorithm, and an error parameter estimation value is obtained through the recursive operation in each sampling period.

2. The calibration method according to claim 1, characterized in that: The step of establishing an error model regarding the camera optical axis pre-adjustment value, error value, and measurement value includes: Establish the pointing vector formula of the camera optical axis; Decomposing the image plane position error into a first azimuth error and a first pitch error, and obtaining a plane transformation matrix according to the first azimuth error and the first pitch error; Decomposing the camera optical axis pointing error into a second azimuth error and a second pitch error, and obtaining a camera optical axis transformation matrix according to the second azimuth error and the second pitch error; Decomposing the optical axis verticality error into a third position error and a third pitch error, and obtaining a verticality transformation matrix according to the third position error and the third pitch error; A pitch transformation matrix is ​​established according to the actual value of the pitch angle, the pitch axis measurement value, the pitch axis encoding zero position error, the pitch axis encoder measurement error, the verticality errors of the two pitch axis systems and the rotation errors of the two pitch axis systems; An azimuth transformation matrix is ​​established according to the actual value of the azimuth angle, the measured value of the azimuth axis, the zero position error of the azimuth axis encoder, the measured error of the azimuth axis encoder, the verticality error of the two azimuth axis systems and the rotation error of the two azimuth axis systems; Decomposing the attitude error into a fourth azimuth error, a fourth pitch error and a roll error, and obtaining an attitude transformation matrix according to the fourth azimuth error, the fourth pitch error and the roll error; The error model is obtained according to the vector formula, the plane transformation matrix, the camera optical axis transformation matrix, the verticality transformation matrix, the pitch transformation matrix, the azimuth transformation matrix and the attitude transformation matrix.

3. The calibration method according to claim 2, characterized in that: The vector formula is as follows: is the actual pointing vector of the camera optical axis. is the initial vector of the camera optical axis, is the transformation matrix, the transformation matrix , It represents the transformation matrix between coordinate systems, b0 is the star system, b is the actual star system, A0 is the turntable azimuth coordinate system, A is the azimuth motion coordinate system, B0 is the pitch frame coordinate system, B is the pitch motion coordinate system, P0 is the ideal optical axis pointing coordinate system, P is the actual optical axis pointing coordinate system, Cam is the actual pointing system in the image plane coordinate system, I It is the inertial system of J2000 satellite.

4. The calibration method according to claim 3, characterized in that: The plane transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the first azimuth error and the first pitch error respectively; The camera optical axis transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and are the second azimuth error and the second pitch error respectively; The verticality transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. and They are the third position error and the third pitch error respectively; The pitch transformation matrix is ​​as follows: in, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the pitch angle, where is the pitch axis measurement value, is the zero position error of the pitch axis encoder, is the encoder measurement error, , is the verticality error of the pitch axis system, , is the pitch axis rotation error; The orientation transformation matrix is ​​as follows: In the formula, Represents the transformation matrix obtained after rotating around the m axis by angle n. is the actual value of the azimuth angle, is the azimuth axis measurement value, is the azimuth axis rotation error, where Azimuth zero error, is the encoder measurement error, , is the verticality error of the azimuth axis system, , is the azimuth axis rotation error; The posture transformation matrix is ​​as follows: in, represents the transformation matrix obtained after rotating around the m-axis by an angle n, and the fourth azimuth error, the fourth pitch error and the roll error are respectively , and .

5. The calibration method according to claim 1, characterized in that: The error model is as follows: in, is the camera optical axis measurement value, express The i-th dimension element of , i=1, 2, 3.

6. The calibration method according to claim 5, characterized in that: The generalized error parameter is as follows: 。 7. The calibration method according to claim 6, characterized in that: The linear model is as follows: in, is a recursive matrix consisting only of the turntable measurement angles.

8. The calibration method according to claim 7, characterized in that: Based on the set initial value of the error parameter, the linear model performs recursive calculations through a recursive algorithm, and obtains an error parameter estimation value through recursive calculations in each sampling period, including: Define variables: , is the camera optical axis measurement value, and Represents the measured value y and recursive matrix of the kth sampling period , use the following recursive algorithm to calculate the error parameters of each sampling period: in, is the estimated value of the error parameter θ, the initial value of which is determined based on the system ground precision measurement parameters combined with experiments, , , c and is an adjustable constant.

Images