Adaptive calibration method and system for external parameters of camera with eyes outside hand

Abstract

The invention discloses a self-adaptive calibration method and system for external parameters of a camera with eyes outside the hand, and the method comprises the steps: controlling a mechanical arm tail end tool to move in the visual field of the camera, and photographing the mechanical arm tail end tool through the camera when the tail end tool reaches a preset position point, thereby obtaining an image; according to the image information, coordinates of a mechanical arm tail end tool relative to a camera coordinate system are obtained, and meanwhile coordinates of the tail end tool relative to a mechanical arm base coordinate system are obtained; according to the coordinate of the mechanical arm tail end tool relative to the camera coordinate system and the coordinate of the tail end tool relative to the mechanical arm base coordinate system, a coordinate transformation relational expression related to the external parameters is obtained; constructing a first-order derivative expression of the auxiliary matrix according to the coordinate transformation relational expression related to the external parameters, and solving to obtain the auxiliary matrix; and according to the auxiliary matrix, constructing an error variable related to the estimation error, and designing an adaptive law driven by the error variable. According to the method for designing the adaptive law through the estimation error, convergence to a true value can be more accurately achieved, and the accuracy of external parameter calibration is remarkably improved.

Description

Technical Field

[0001] The present invention relates to a method and system for adaptively calibrating the external parameters of a camera with the eye outside the hand, belonging to the technical field of robot vision. Background Art

[0002] In many application fields of automation and robotics, cameras have been widely used in industrial inspection, navigation, and robot vision systems. Especially in robot operations (such as assembly, handling, and fine operations), cameras are often used to achieve precise positioning of target objects in the environment. However, the effective application of a camera depends on accurate calibration of its external parameters with respect to the robot - that is, determining the position and orientation of the camera in the robot coordinate system.

[0003] Traditional methods for calibrating the external parameters of a camera usually involve using a checkerboard calibration board and complex calculation processes, and require offline calibration, that is, the data acquisition process and the calibration solution process are separated. This is not only time - consuming but also difficult to implement in some complex scenarios. In addition, once there are minor changes in the working environment of the robot or the installation position of the camera, recalibration is required, which significantly increases the complexity and usage cost of the system.

[0004] As an excellent online estimation method, adaptive parameter estimation has been widely used in fields such as calibration. However, existing adaptive parameter estimation algorithms often do not consider the convergence of parameter estimation errors. Therefore, designing a simple, efficient, and robust online external parameter calibration method is of great practical significance for the robotic arm vision system. Summary of the Invention

[0005] The present invention provides a method and system for adaptively calibrating the external parameters of a camera with the eye outside the hand to solve the problem of time - consuming calibration of camera external parameters and further solve the problem of inability to perform online calibration.

[0006] The technical solution of the present invention is as follows:

[0007] According to a first aspect of the present invention, there is provided a method for adaptively calibrating the external parameters of a camera with the eye outside the hand, comprising the following steps:

[0008] Step S1, controlling the end - effector of the robotic arm to move within the camera's field of view. When the end - effector reaches a preset position point, taking a picture of the end - effector of the robotic arm through the camera to obtain an image;

[0009] Step S2, obtaining the coordinates of the end - effector of the robotic arm relative to the camera coordinate system according to the image information, and simultaneously obtaining the coordinates of the end - effector relative to the robotic arm base coordinate system;

[0010] Step S3: Obtain a coordinate transformation relation related to the external parameters based on the coordinates of the end - effector tool of the robotic arm in the camera coordinate system and the coordinates of the end - effector tool in the robotic arm base coordinate system.

[0011] Step S4: Construct a first - derivative expression of the auxiliary matrix according to the coordinate transformation relation related to the external parameters, and solve it to obtain the auxiliary matrix.

[0012] Step S5: Construct an error variable related to the estimation error based on the auxiliary matrix, and design an adaptive law driven by the error variable.

[0013] Furthermore, it further includes Step S6, and specifically, Step S6 is: Collect images of other preset position points in real - time, update the auxiliary matrices F and M according to S2 - S4, and perform integration according to the adaptive law established in S5 to update the estimated value of the external parameters until convergence.

[0014] Furthermore, before taking the image of the end - effector tool of the robotic arm, it further includes:

[0015] Calibrate the internal parameters of the camera based on the calibration board.

[0016] Fix a calibration ball with a known radius at the end of the robotic arm as the end - effector tool, and calibrate the tool coordinate system to obtain the homogeneous transformation matrix of the tool relative to the end of the robotic arm.

[0017] Furthermore, the obtaining of the coordinates of the end - effector tool of the robotic arm in the camera coordinate system according to the image information includes: Calculate the depth of the center point of the end - effector tool relative to the camera based on the image and the camera internal parameter matrix information; Calculate the coordinates of the end - effector tool of the robotic arm in the camera coordinate system based on the calculated depth of the center point of the end - effector tool relative to the camera.

[0018] Furthermore, the obtaining of the coordinates of the end - effector tool in the robotic arm base coordinate system is specifically: Read the homogeneous transformation matrix from the end of the robotic arm to the base, and solve the homogeneous coordinate form of the coordinates of the end - effector tool in the robotic arm base coordinate system from the homogeneous transformation matrix from the end of the robotic arm to the base. The expression is:

[0019] b X = b T e e T t t X

[0020] Wherein, b T e is the homogeneous transformation matrix from the end of the robotic arm to the base, which is determined by the kinematics of the robotic arm; e T t is the homogeneous transformation matrix of the tool relative to the end of the robotic arm; The coordinates of the tool relative to the base coordinate system of the robotic arm b The homogeneous coordinate form of x; The coordinates of the tool in the tool coordinate system t The homogeneous coordinate form of x.

[0021] Furthermore, the coordinate transformation relationship related to the external parameters obtained based on the coordinates of the end effector of the robotic arm relative to the camera coordinate system and the coordinates of the end effector relative to the base coordinate system of the robotic arm is:

[0022] b X = b T c c X;

[0023] Wherein, b T c Is the homogeneous transformation matrix from the camera to the base coordinate system of the robotic arm; Are the coordinates of the tool relative to the base coordinate system of the robotic arm b The homogeneous coordinate form of x; Are the coordinates of the end effector of the robotic arm relative to the camera coordinate system c The homogeneous coordinate form of x.

[0024] Furthermore, the S4 is specifically:

[0025] Construct the first derivative of the first auxiliary matrix F; construct the first derivative of the second auxiliary matrix M according to the coordinate transformation relationship related to the external parameters; the first derivatives of the first auxiliary matrix F and the second auxiliary matrix M are:

[0026]

[0027] Solve the first derivative to obtain F and M as:

[0028]

[0029] Wherein, l represents the forgetting factor and is a positive constant;

[0030] Furthermore, the S5 is specifically:

[0031] Construct an error variable E related to the estimation error according to the auxiliary matrix:

[0032]

[0033] Wherein, Is the estimated external parameter, F is the first auxiliary matrix, M is the second auxiliary matrix, Is the estimation error of the external parameter;

[0034] Therefore, the adaptive law designed with the error variable is as follows:

[0035]

[0036] where c is a constant gain, which is the designed adaptive law.

[0037] According to the second aspect of the present invention, an external parameter adaptive calibration system for a camera with the eye outside the hand is provided, including the module of the external parameter adaptive calibration method for a camera with the eye outside the hand described in any one of the above.

[0038] According to the third aspect of the present invention, a processor is provided, and the processor is used to run a program, wherein when the program runs, it executes the external parameter adaptive calibration method for a camera with the eye outside the hand described in any one of the above.

[0039] The beneficial effects of the present invention are as follows: Compared with ordinary adaptive algorithms, the method of designing an adaptive with the estimated error proposed in this application can converge to the true value more precisely, significantly improving the accuracy of external parameter calibration; in addition, the process of adaptive external parameter calibration is completed online, integrating the data acquisition and the external parameter calibration process, improving the efficiency, and having simple operation and strong real-time performance, with important practical application value. In addition, this method can also be applied to the external parameter calibration of the eye-out-of-hand system of binocular cameras. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions covered by the present application, the following provides a detailed description of the technical solutions in conjunction with the drawings of related embodiments of the present application. It should be emphasized that the presented embodiments only represent one of the possible forms that the present application may achieve, rather than all possible forms.

[0041] Figure 1 is a schematic diagram of external parameter calibration of a camera with the eye outside the hand in the related art;

[0042] Figure 2 is a schematic diagram of the end of the robotic arm with a calibration ball;

[0043] Figure 3 is a schematic flowchart of an external parameter adaptive calibration method for a camera with the eye outside the hand provided by the embodiment;

[0044] Figure 4 is a schematic diagram of eye-out-of-hand calibration provided by the embodiment;

[0045] Figure 5 is a schematic diagram of external parameter estimation provided by the embodiment;

[0046] Figure 6 is a schematic diagram of the convergence effect of the external parameter estimation error provided by the embodiment;

[0047] Figure 7 Internal functional module relationship diagram of an external parameter calibration device provided for an example;

[0048] Each label in the figure is: 1 - camera, 2 - robotic arm, 3 - calibration board, 4 - base, 5 - calibration ball. Detailed implementation manners

[0049] To make the objectives, 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 with reference to the accompanying drawings. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be arbitrarily combined with each other.

[0050] Traditional camera external parameter calibration methods usually involve using a checkerboard calibration board, as Figure 1 is a schematic diagram of external parameter calibration with the eye outside the hand in the related art. It can be seen from the figure that the camera 1 is fixed on the shelf outside the robotic arm 2, the calibration board 3 is installed at the end of the robotic arm 2. By controlling the movement of the robotic arm, different robotic arm pose information is collected, the camera captures images of the calibration board in different poses, and then an optimization algorithm is used for iterative solution of the camera external parameters. However, this method requires a special fixture to be designed for the fixed calibration board. In addition, the calibration board is difficult to fix, and it is inevitable that there will be a relative position offset with the robotic arm during movement, affecting the accuracy of camera external parameter calibration. Moreover, the optimization iteration process is complex, and the solution result is greatly affected by the accuracy of calibration board corner point acquisition. To simplify the calibration process and improve the calibration accuracy, the present invention proposes a new method and system for adaptive calibration of camera external parameters with the eye outside the hand.

[0051] As Figures 2 - 7 shown, according to the first aspect of the embodiments of the present invention, there is provided a method for adaptive calibration of camera external parameters with the eye outside the hand, including the following steps:

[0052] Step S1: Control the end effector of the robotic arm to move within the camera's field of view. When the end effector reaches a preset position point, capture the end effector of the robotic arm through the camera to obtain an RGB image;

[0053] Step S2: Obtain the coordinates of the end effector of the robotic arm relative to the camera coordinate system according to the RGB image information, and at the same time obtain the coordinates of the end effector relative to the robotic arm base coordinate system;

[0054] Step S3: Based on the coordinates of the end effector of the robotic arm relative to the camera coordinate system and the coordinates of the end effector relative to the robotic arm base coordinate system, obtain a coordinate transformation relationship formula related to the external parameters;

[0055] Step S4: According to the coordinate transformation relation related to the external parameters, construct the first derivative expression of the auxiliary matrix and solve it to obtain the auxiliary matrix.

[0056] Step S5: According to the auxiliary matrix, construct an error variable related to the estimation error and design an adaptive law driven by the error variable.

[0057] Furthermore, it further includes Step S6: Real-time collect the RGB images of other preset position points, update the auxiliary matrices F and M according to S2 - S4, and perform integration according to the adaptive law established in S5 to update the external parameter estimation value until convergence.

[0058] Furthermore, before taking the RGB image of the end - effector of the robotic arm, it is also necessary to calibrate the internal parameters of the camera based on the calibration board in advance; in addition, a calibration ball 5 with a known radius is fixedly connected to the end of the robotic arm as the end - effector, and the tool coordinate system is calibrated. Figure 2 FIG. is a schematic diagram of the end of the robotic arm with a calibration ball. The calibration ball 5 is aligned with the center of the end of the robotic arm and is installed on the flange of the robotic arm, which can be regarded as the end - effector of the robotic arm.

[0059] Exemplarily, a Hikvision industrial camera is adopted, and the camera is calibrated based on the calibration board to obtain the camera internal parameter K; further, a calibration ball 5 is installed at the end of the Elite robotic arm, and the end - effector is calibrated based on the traditional calibration method to obtain the homogeneous transformation matrix e T t ; The camera internal parameter K and the homogeneous transformation matrix of the tool relative to the end of the robotic arm e T t The expressions are as follows:

[0060]

[0061]

[0062] In the embodiment of the present invention, the Zhang - Zhengyou calibration method is used for the camera, and the nine - point method is used for the end - effector. It should be noted that this is only an example here, and other calibration methods can also be used according to needs.

[0063] Furthermore, the obtaining of the coordinates of the end - effector of the robotic arm relative to the camera coordinate system according to the RGB image information includes:

[0064] According to the RGB image and the camera internal parameter matrix information, calculate the depth of the center point of the end - effector relative to the camera:

[0065]

[0066] Wherein, cz is the depth of the end - effector center point relative to the camera; R is the radius of the calibration sphere; f x is the focal length of the camera in the horizontal direction, obtained from the intrinsic matrix; r img is the projected radius of the calibration sphere in the RGB image.

[0067] According to the calculated depth of the end - effector center point relative to the camera, calculate the coordinates of the end - effector of the robotic arm relative to the camera coordinate system:

[0068] c x = c zK -1 p;

[0069] where, is the coordinate of the end - effector of the robotic arm relative to the camera coordinate system, that is, the spatial position of the end - effector in the camera coordinate system, represents a three - dimensional vector; is the intrinsic parameter of the camera, represents a 3×3 matrix; p is the pixel coordinate of the end - effector center point.

[0070] Furthermore, the obtaining of the coordinates of the end - effector relative to the robotic arm base coordinate system is specifically: read the homogeneous transformation matrix from the robotic arm end to the base 4 from the robotic arm controller, and solve the homogeneous coordinate form of the coordinates of the end - effector relative to the robotic arm base coordinate system from the homogeneous transformation matrix from the robotic arm end to the base 4. The expression is:

[0071] b X = b T e e T t t X

[0072] where, b T e is the homogeneous transformation matrix from the robotic arm end to the base 4, determined by the robotic arm kinematics; e T t is the homogeneous transformation matrix of the tool relative to the robotic arm end; is the coordinate of the tool relative to the robotic arm base coordinate system b in the homogeneous coordinate form of x; is the coordinate of the tool in the tool coordinate system t in the homogeneous coordinate form of x; for a calibration sphere, is the origin of the tool coordinate system; represents the transpose.

[0073] Further, the coordinate transformation relationship related to the external parameters obtained based on the coordinates of the end - effector tool of the robotic arm in the camera coordinate system and the coordinates of the end - effector tool in the robotic arm base coordinate system is as follows:

[0074] b X= b T c c X;

[0075] Among them, b T c is the homogeneous transformation matrix from the camera to the robotic arm base coordinate system, that is, the external camera parameters to be calibrated; is the coordinate of the tool in the robotic arm base coordinate system b in the homogeneous coordinate form of x; is the coordinate of the end - effector tool of the robotic arm in the camera coordinate system c in the homogeneous coordinate form of x.

[0076] Further, the S4 is specifically as follows:

[0077] Construct the first - order derivative of the first auxiliary matrix F; construct the first - order derivative of the second auxiliary matrix M according to the coordinate transformation relationship related to the external parameters; the first - order derivatives of the first auxiliary matrix F and the second auxiliary matrix M are:

[0078]

[0079] Solve the first - order derivative to obtain F and M as:

[0080]

[0081] Among them, l represents the forgetting factor, which is a positive constant;

[0082] According to the coordinate transformation relationship related to the external parameters, the equivalent relationship

[0083] Further, based on the auxiliary matrix, construct an error variable related to the estimation error, and design an adaptive law driven by the error variable, specifically as follows:

[0084] Construct an error variable E related to the estimation error based on the first and second auxiliary matrices:

[0085]

[0086] Among them, is the estimated external parameter; it can be known from the equivalent relationship of M that Among them, is the estimation error of the external parameter;

[0087] Therefore, the adaptive law designed with the error variable is as follows:

[0088]

[0089] where c is a constant gain, which is the designed adaptive law. Integrating the adaptive law can obtain the estimated value of the external parameters of the camera.

[0090] Figure 3 FIG. is a schematic flow chart of the calibration method described in the embodiment. In the embodiment, the present invention online calibrates the external parameters of the camera with an adaptive law driven by an error variable. And this application builds a kinematic model of a 6-degree-of-freedom robotic manipulator in MATLAB / Simulink, and realizes the adaptive estimation of the external parameters of the eye-to-hand camera by controlling the movement of the end trajectory of the manipulator. Figure 4 This is a schematic diagram of the simulation of the external parameter calibration. Figure 5 It shows the convergence of the parameter estimation of the external parameters of the camera, Figure 6 and shows the convergence of the estimation error. In the simulation, in order to verify the effectiveness of the algorithm, the given external parameters of the camera are:

[0091]

[0092] The estimated results of the external parameters by adaptive parameter estimation are:

[0093]

[0094] From the simulation results, the estimated values of the external parameters all converge close to the true values. At the same time, the method provided in this application can converge in about 5 s. From the convergence results of the error variable, the error can converge to near zero. This shows that the method of the present invention has the advantages of fast convergence and high precision. At the same time, the external parameter calibration method of the present invention does not need to use a calibration board, which simplifies the calibration process. This method can estimate the external parameters online, so it can be used in some occasions where the camera needs to be moved frequently, and can also be used for the external parameter calibration of binocular cameras.

[0095] Since the adaptive parameter estimation method needs to satisfy the persistent excitation condition, that is, I is the identity matrix, and δ is a constant greater than 0. To satisfy the persistent excitation condition, it is necessary to control the end of the manipulator to move continuously within the camera's field of view until the error converges. When the error converges, the adaptive law will stop updating, which means that the calibration work has been completed and the adaptive algorithm has made the estimated value converge to the true value.

[0096] According to the second aspect of the embodiment of the present invention, an adaptive calibration system for the external parameters of an eye-to-hand camera is provided. Figure 7It shows the functional module relationship diagram of the system. It includes: an acquisition module, which is used to control the movement of the end tool of the robotic arm within the camera's field of view. When the end tool reaches the preset position point, the end tool of the robotic arm is photographed by the camera to obtain an image; a first acquisition module, which is used to obtain the coordinates of the end tool of the robotic arm relative to the camera coordinate system according to the image information; a second acquisition module, which is used to obtain the coordinates of the end tool relative to the robotic arm base coordinate system; a first obtaining module, which is used to obtain the coordinate transformation relationship formula related to the external parameters based on the coordinates of the end tool of the robotic arm relative to the camera coordinate system and the coordinates of the end tool relative to the robotic arm base coordinate system; a second obtaining module, which is used to construct the first-order derivative expression of the auxiliary matrix according to the coordinate transformation relationship formula related to the external parameters and solve it to obtain the auxiliary matrix; an external parameter estimation module, which is used to construct an error variable related to the estimation error based on the auxiliary matrix and design an adaptive law driven by the error variable to estimate the external parameters.

[0097] According to the third aspect of the embodiments of the present invention, a processor is provided. The processor is used to run a program. When the program runs, it executes the method for adaptive calibration of the external parameters of the camera outside the hand-eye as described in any one of the above.

[0098] Optionally, a computer device is also provided, including a memory, a processor, and an interface for collecting camera and robotic arm data. An executable program is stored in the memory, and when the executable program runs, it will implement the method described in any one of the above.

[0099] Applying the above technical solutions, the present invention proposes a method for adaptive calibration of the external parameters of the camera outside the hand-eye in view of the complex problem of the camera calibration process outside the hand-eye. This method does not require the use of a calibration board, can achieve online calibration, perform the two processes of data acquisition and calibration simultaneously, and can judge whether the calibration result is accurate according to the parameter convergence situation. It has the advantages of simple operation, high precision, and fast convergence speed, and can further apply the calibration result to the visual servo of the robotic arm.

[0100] The specific embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those of ordinary skill in the art, various changes can be made without departing from the gist of the present invention.

Claims

1. A camera extrinsic parameter adaptive calibration method with eye outside hand, characterized in that: The following steps are involved: Step S1, controlling the end tool of the robot arm to move within the field of view of the camera, and when the end tool reaches a preset position, the end tool of the robot arm is photographed by the camera to obtain an image; Step S2, obtaining the coordinates of the end tool of the robot arm relative to the camera coordinate system according to the image information, and simultaneously obtaining the coordinates of the end tool relative to the robot arm base coordinate system; Step S3, obtaining a coordinate transformation relation related to the external parameter according to the coordinates of the end tool of the robot arm relative to the camera coordinate system and the coordinates of the end tool relative to the robot arm base coordinate system; Step S4, constructing a first-order derivative expression of the auxiliary matrix according to a coordinate transformation relation related to the external parameter, and solving it to obtain the auxiliary matrix; Step S5, constructing an error variable related to the estimation error based on the auxiliary matrix, and designing an adaptive law driven by the error variable.

2. The method for adaptive calibration of camera extrinsics with eye outside hand according to claim 1, characterized in that: The method further includes step S6, which specifically includes: acquiring images of other preset position points in real time, updating auxiliary matrices F and M according to S2-S4, and integrating according to the adaptive law established in S5 to update the external parameter estimation value until convergence.

3. The method for adaptive calibration of camera extrinsics with eye outside hand according to claim 1, characterized in that: Before taking an image of the end-of-arm tooling, it also includes: Performing an internal parameter calibration on the camera based on a calibration plate; A calibration sphere with a known radius is fixed to the end of the robot arm as the end tool, and the tool coordinate system is calibrated to obtain the homogeneous transformation matrix of the tool relative to the end of the robot arm.

4. The method for adaptive calibration of camera extrinsics with eye outside hand according to claim 1, characterized in that: The method of obtaining the coordinates of the end tool of the robot arm relative to the camera coordinate system based on the image information includes: calculating the depth of the center point of the end tool relative to the camera based on the image and the camera intrinsic parameter matrix information; calculating the coordinates of the end tool of the robot arm relative to the camera coordinate system based on the calculated depth of the center point of the end tool relative to the camera.

5. The method for adaptive calibration of camera extrinsic parameters when the eye is outside the hand according to claim 1, characterized in that: The acquisition of the coordinates of the end tool relative to the robot base coordinate system is specifically as follows: reading the homogeneous transformation matrix from the end of the robot to the base, and solving the homogeneous coordinate form of the coordinates of the end tool relative to the robot base coordinate system from the homogeneous transformation matrix from the end of the robot to the base, and the expression is: b X= b T e e T t t X in, b T e is the homogeneous transformation matrix from the end of the robot to the base, which is determined by the kinematics of the robot; e T t is the homogeneous transformation matrix of the tool relative to the end of the robot arm; is the coordinate of the tool relative to the robot base coordinate system b The homogeneous coordinate form of x; is the coordinate of the tool in the tool coordinate system t The homogeneous coordinate form of x.

6. The method for adaptive calibration of camera extrinsics with eye outside hand according to claim 1, characterized in that: According to the coordinates of the end tool of the robot arm relative to the camera coordinate system and the coordinates of the end tool relative to the robot arm base coordinate system, the coordinate transformation relationship related to the external parameter is obtained as follows: b X= b T c c X; in, b T c is the homogeneous transformation matrix from the camera to the robot base coordinate system; is the coordinate of the tool relative to the robot base coordinate system b The homogeneous coordinate form of x; is the coordinate of the tool at the end of the robot arm relative to the camera coordinate system c The homogeneous coordinate form of x.

7. The method for adaptive calibration of camera extrinsic parameters when the eye is outside the hand according to claim 1, characterized in that: The S4 is specifically: Construct the first derivative of the first auxiliary matrix F; construct the first derivative of the second auxiliary matrix M according to the coordinate transformation relationship related to the external parameter; the first derivatives of the first auxiliary matrix F and the second auxiliary matrix M are: Solving the first-order derivative, we get F and M as: Among them, l represents the forgetting factor, which is a positive constant; 8. The method for adaptively calibrating camera extrinsics with eye outside hand according to claim 1, characterized in that: The S5 is specifically: According to the auxiliary matrix, construct the error variable E related to the estimation error: in, is the estimated external parameter, F is the first auxiliary matrix, M is the second auxiliary matrix, is the estimation error of the external parameter; Therefore, the adaptive law driven by the error variable is designed as: Where c is the constant gain, This is the designed adaptive law.

9. An eye-outside-hand camera extrinsic adaptive calibration system, characterized in that: A module comprising the method for adaptively calibrating camera extrinsic parameters when the eye is outside the hand as described in any one of claims 1-8.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the method for adaptively calibrating camera extrinsic parameters when the eye is outside the hand as described in any one of claims 1-8.

Images