Endoscope active following motion control method

Abstract

The invention discloses an endoscope active following motion control method, which comprises the following steps of: 1, carrying out kinematics modeling on a continuum endoscope, obtaining a pose transformation matrix of the continuum endoscope so as to obtain a working space of the continuum endoscope, and verifying a reachable range of the working space in an eyeball; step 2, constructing a visual servo following controller based on the endoscope image, collecting the endoscope image and identifying the tip of the surgical instrument in the endoscope image based on an image processing algorithm of a YOLO neural network architecture; and extracting the deviation between the pixel space coordinates of the feature points and the expected features in real time, and controlling the camera at the tail end of the continuum endoscope to move based on the deviation of the feature points until the image features reach a target state. The multi-view intraoperative image can be provided in the continuous bending movement process of the continuum endoscope, the tail end of a surgical instrument is followed in real time in the operation, the posture of the surgical instrument in the operation is observed in real time, and the automation degree of the continuum endoscope is improved.

Description

Technical Field

[0001] This invention relates to the field of endoscope control technology, and in particular to an active motion control method for endoscopes. Background Technology

[0002] Retinal microsurgery is one of the most precise procedures in ophthalmology, and its success heavily relies on the surgeon's exceptional skill and stable intraoperative field of vision. In traditional manual surgery, the surgeon directly manipulates instruments, facing numerous inherent challenges: first, the surgeon's physiological tremors directly affect the precision of the procedure; second, the limited field of vision restricts the surgeon's ability to observe the entire surgical area; and third, hand fatigue from prolonged surgery has a cumulative effect, further increasing surgical risks. Against this backdrop, robot-assisted ophthalmic surgery systems have emerged, demonstrating significant potential to improve surgical stability and precision while reducing the surgeon's workload.

[0003] Intraoperative image clarity and flexibility play a crucial role in surgical procedures. Microscopes and OCT (Optical Coherence Tomography) devices can provide images from the outside of the eye through the pupil to the retina at a fixed viewing angle. However, when patients have corneal opacity or pupillary constriction, clear intraoperative images cannot be obtained. Furthermore, the inability to dilate the pupil, resulting in insufficient light penetration and direct visualization, limits the further use of extraocular imaging equipment. Endoscopes, inserted through a scleral incision, avoid visual field obstruction caused by these issues and can adjust the visual field to some extent, providing an ideal solution for improving surgical illumination and direct visualization of fundus lesions.

[0004] However, the handheld rigid ophthalmic endoscopes commonly used in clinical practice still lack sufficient flexibility. Within the confined space of the eye, it is difficult to flexibly and quickly adjust the viewing angle to track surgical instruments in real time or observe lesions from different angles, limiting their ability to provide multi-angle, comprehensive intraoperative imaging. Furthermore, the operation of the endoscope relies entirely on manual manipulation by the surgeon, which not only increases the surgeon's workload but also enhances the complexity of the surgery. Simultaneously, rigid endoscopes also suffer from problems such as easy loss of field of vision and heavy reliance on the operator's surgical experience.

[0005] In recent years, the integration of visual servo control systems with endoscopes has attracted attention. This technology promises to enable automatic adjustment of the endoscope's posture or field of view, thereby freeing doctors from the arduous task of endoscopic manipulation and reducing their workload. Simultaneously, the visual servo system, through feedback control and coupling with the endoscope, can actively compensate for eye or instrument movements to a certain extent, helping to reduce the possibility of visual field loss and maintain a stable and constant intraoperative visual field. Summary of the Invention

[0006] In view of the above problems, the present invention provides an active motion control method for endoscopes. Compared with rigid endoscopes, the present invention can provide multi-view intraoperative images during the continuous bending motion of a continuous endoscope, and observe the posture of surgical instruments in real time during the operation.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for controlling the active motion tracking of an endoscope, comprising the following steps: Step 1: Perform kinematic modeling on the continuous endoscope to obtain the pose transformation matrix of the continuous endoscope, thereby obtaining the workspace of the continuous endoscope, and verifying the reachability of the workspace within the eyeball. Step 2: Construct a visual servo follower controller based on endoscopic images. Collect endoscopic images and use an image processing algorithm based on the YOLO neural network architecture to identify the surgical instrument tips in the endoscopic images. Extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of the continuous endoscope based on the deviation of the feature points until the image features reach the target state.

[0008] In one embodiment of the present invention, the kinematic modeling and pose transformation matrix of the continuous endoscope in step one includes: The homogeneous transformation of the continuous endoscope's motion in the XZ plane is decomposed into offset parameters of the X and Z axes of the connecting rods, using the chord length of the arc formed by the continuous endoscope. The homogeneous transformation matrix for the bending angle of the continuous endoscope is shown in equation (1): (1) The continuous endoscope as a whole has rotational degrees of freedom along the central axis, allowing the motion of the continuous endoscope in the two-dimensional plane to cover the three-dimensional space, generating a continuous set of workspace points. Its coordinate transformation relationship is shown in equation (2): (2) In the formula, The angle represents the rotational degree of freedom of the continuous endoscope; let the origin of this rotational degree of freedom coordinate system coincide with the telecentric fixed point of the initial continuous endoscope, and let this point have a fixed straight-line distance L relative to the root of the curved part of the continuous endoscope. This continuous endoscope is integrated into a three-degree-of-freedom (RCM) mechanism. The joint coordinate system of the RCM mechanism is located at the telecentric fixed point of the RCM mechanism. This coordinate system consists of three pairwise orthogonal coordinate axes: x-axis, y-axis, and z-axis. The RCM mechanism has three degrees of freedom, which, through pitch, roll, and feed, respectively, allow the end-effector to reach the target position to complete the operation. The transformation matrix between adjacent joints is... Referring to the Denavit-Hartenberg parametric method, the pose transformation matrix of the telecentric fixed point of the RCM mechanism is... As shown in equation (4): (3) (4) in, This represents the rotation angle of the RCM mechanism about the z-axis of the joint coordinate system. This represents the difference in the length of the common normal of the RCM mechanism along the z-axis of the joint coordinate system. This represents the distance between the common normals of two adjacent z-axis in the joint coordinate system of the RCM mechanism. This indicates that the joint coordinate system of the RCM mechanism is perpendicular to... The angle between the two axes in the plane; The pose transformation matrix of the continuous endoscope is shown in equation (5): (5).

[0009] In one embodiment of the present invention, when the continuous endoscope performs a following task, the velocity equation is described as follows: (6) In the formula, Represents the linear and angular velocities of the endoscope tip in the base coordinate system; matrix This represents the Jacobian matrix of a continuous endoscope. This indicates the driving speed of a five-degree-of-freedom vehicle.

[0010] In one embodiment of the present invention, step two specifically includes: The continuous endoscope transforms the three-dimensional coordinates of the target point in the world coordinate system to the camera coordinate system and projects them onto the camera's pixel plane; assuming This represents the position of the instrument's end point in the camera coordinate system. This represents the position of a two-dimensional point projected onto the image plane in the camera coordinate system. These are the coordinates of the feature point in the pixel plane. According to the principle of pinhole imaging, the transformation relationship between the three is shown in equation (7): (7) In the formula, , This represents the actual width and length corresponding to each pixel value. The reference coordinates representing the pixel coordinate system. The focal length of the camera; In image-based visual servoing, it is necessary to extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of the continuous endoscope based on the deviation of the feature points until the image features reach the target state; the image features that change over time are defined as... The target image features are At different times The characteristic error is defined by equation (8): (8) When the end-camera of a continuous endoscope has a spatial velocity that varies with time... At that time, the features of the corresponding pixel plane The time derivative and its relationship are described by equation (9): (9) in, The image Jacobian matrix is ​​composed of information about individual image feature points and the position information of the corresponding spatial points in the camera frame: (10) According to formulas (8) to (10), let the error converge exponentially, and the visual servo control law is expressed as follows: (11) In the formula, Represents the pseudo-inverse of a matrix. The convergence coefficient is . , This represents the rate of change of target image features over time; due to the inherent error tolerance of visual servoing, the image Jacobian matrix... The value is set to a constant; combined with the kinematic model of the continuous endoscope itself, a visual servo follower controller from the drive space to the image space is obtained: (12) (13) In the formula, This represents the pseudo-inverse of the Jacobian matrix of a continuous endoscope.

[0011] In one embodiment of the present invention, the continuous endoscope has spin degree of freedom and yaw degree of freedom, and its effective working space is a spatial curved surface.

[0012] In one embodiment of the present invention, the continuous endoscope includes a distal fixed section, an intermediate continuous flexible section, and an end camera connected together, wherein the intermediate continuous flexible section is flexible.

[0013] In one embodiment of the present invention, the continuous endoscope is mounted on an RCM mechanism and moves flexibly within the entire working space inside the eyeball to acquire intraoperative images in different end-effector postures.

[0014] In one embodiment of the present invention, the bending angle range of the continuous endoscope is: The spin angle range is .

[0015] In a second aspect, the present invention provides a computer-readable storage medium storing computer instructions which are executed by a processor using the method described above.

[0016] Thirdly, the present invention provides a computer program product, the computer program product storing computer instructions, the computer instructions being executed by a processor using the method described above.

[0017] The beneficial effects achieved by this invention are as follows: This invention provides an active tracking motion control method for an endoscope. Based on visual servo tracking control of endoscopic images, it enables a continuous endoscope to follow the surgical instrument tip in real time during surgery, thus providing medical professionals with real-time intraoperative images. This invention uses the YOLO object detection algorithm to obtain the feature coordinates of the continuous endoscope tip in the endoscopic image plane. The continuous endoscope and active tracking motion control method were experimentally validated on an ophthalmic surgical robot (ESR) using an eye model. Experimental results demonstrate that the continuous endoscope can stably track the instrument tip inside the eye, causing it to converge to the desired coordinates, with a motion error of approximately 25 pixels.

[0018] Compared to the uncertainty caused by the need for medical staff to manually adjust the position of traditional rigid endoscopes, the active motion control method for endoscopes provided in this invention has the following advantages: 1. Avoid problems such as secondary eye wounds and loss of the endoscope position that may be caused by human factors such as physiological tremors, fatigue accumulation, and external interference, which may lead to instability in the endoscope position; 2. It improves the automation level of continuous endoscopes, enabling them to automatically align with fundus lesions and follow the end of surgical instruments in real time during the operation, giving them significant advantages in lesion identification, posture maintenance, and actuator guidance. 3. It allows the endoscope to flexibly acquire fundus images from multiple perspectives (including lateral and emphyseal views) and multiple images. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the drawings without creative effort.

[0020] Figure 1 (a) is a schematic diagram of the continuous endoscope of the present invention; Figure 1 (b) is a schematic diagram of the motion space of the continuous endoscope of the present invention; Figure 1 (c) in the figure is a schematic diagram of the motion of the two-dimensional plane of the continuous endoscope of the present invention; Figure 1 (d) in the figure is a schematic diagram of the motion space of the RCM mechanism of the present invention.

[0021] Figure 2 This is a schematic diagram of the camera coordinate system of the present invention.

[0022] Figure 3 This is a control block diagram of the present invention. Detailed Implementation

[0023] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0024] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the terms “comprising” and “having”, and any variations thereof, in the specification, claims and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0025] In the description of the embodiments of this invention, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this invention, "multiple" means two or more, unless otherwise explicitly defined.

[0026] In this invention, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least some embodiments of the invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this invention can be combined with other embodiments.

[0027] In the description of the embodiments of this invention, the term "and / or" is merely a description of the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this invention, the character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0028] like Figures 1 to 3 As shown, this embodiment provides an active motion control method for an endoscope. The endoscope is a continuous endoscope (FEA), which, compared to a rigid endoscope, can provide multi-view intraoperative images during continuous bending motion, allowing real-time observation of the surgical instrument's posture. The structure of this continuous endoscope can be referenced in Chinese Patent Document CN115969300A, which discloses a small-scale multi-cavity endoscopic continuous robotic arm structure and manufacturing method, or in Chinese Patent Document CN112545435B, which discloses a modular multi-wire driven continuous lens arm based on a fixed pulley. The continuous endoscope includes a connected distal fixed section, a middle continuous flexible section, and an end-effector camera. The middle continuous flexible section is flexible. The continuous endoscope is mounted on the telecentric motion (RCM) mechanism of an ophthalmic surgical robot. In conjunction with the RCM mechanism of the ophthalmic surgical robot, it can flexibly move within the entire working space inside the eyeball to acquire intraoperative images at different end-effector postures.

[0029] This embodiment provides an active motion control method for endoscopes, comprising the following steps: Step 1: Perform kinematic modeling on the continuous endoscope to obtain the pose transformation matrix of the continuous endoscope, thereby obtaining the workspace of the continuous endoscope, and verifying the reachability of the workspace within the eyeball. Step 2: Construct a visual servo follower controller based on endoscopic images. Collect endoscopic images and identify the surgical instrument tips in the endoscopic images using an image processing algorithm based on the YOLO neural network architecture. Extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of the continuous endoscope based on the deviation of the feature points until the image features reach the target state.

[0030] Optionally, the kinematic modeling and pose transformation matrix of the continuous endoscope in step one includes: The homogeneous transformation of the continuous endoscope motion in the XZ plane varies at any given distance. After setting the value, the position of the continuous endoscope can be calculated using an equivalent DH parameter method. The chord length of the arc formed by the continuous endoscope can be decomposed into offset parameters of the connecting rod along the X and Z axes. The bending angle of the continuous endoscope is shown. In this transformation, unlike the DH parameters of the traditional rigid linkage structure, the coordinate system movement and attitude change occur simultaneously when the continuous endoscope moves, and the motion parameters are coupled with each other.

[0031] The homogeneous transformation matrix is ​​shown in equation (1): (1) The continuous endoscope as a whole has a rotational degree of freedom along the central axis, which allows the motion of the continuous endoscope in the two-dimensional plane to cover the three-dimensional space, generating a continuous set of workspace points. Its coordinate transformation relationship is shown in Equation (2).

[0032] (2) In the formula, The angle represents the rotational degree of freedom of the continuous endoscope. To facilitate subsequent trajectory control and end-effector pose calculation, the origin of this rotational degree of freedom coordinate system is aligned with the initial telecentric fixed point of the continuous endoscope. This point is a fixed straight-line distance L relative to the root of the curved portion of the continuous endoscope. The telecentric fixed point of the continuous endoscope is the point around which the endoscope only performs translational and rotational movements when an external driving torque is applied. This telecentric fixed point is fixed at the incision site where the endoscope enters the body, thus enabling minimally invasive surgery via the telecentric fixed point.

[0033] This continuous endoscope is integrated into a three-degree-of-freedom (RCM) mechanism. The joint coordinate system of the RCM mechanism is located at the telecentric fixed point of the RCM mechanism. This coordinate system consists of three pairwise orthogonal axes: x-axis, y-axis, and z-axis. The RCM mechanism has three degrees of freedom, which, through pitch, roll, and feed, respectively, allow the end-effector to reach the target position to complete the operation. The transformation matrix between adjacent joints is... Referring to the Denavit-Hartenberg parametric method (DH parametric method), the pose transformation matrix of the distal fixed point of the RCM mechanism is then determined. As shown in equation (4): (3) (4) The corresponding DH parameters are shown in Table 1, where, This represents the rotation angle of the RCM mechanism about the z-axis of the joint coordinate system. This represents the difference in the length of the common normal of the RCM mechanism along the z-axis of the joint coordinate system. This represents the distance between the common normals of two adjacent z-axis in the joint coordinate system of the RCM mechanism. This indicates that the joint coordinate system of the RCM mechanism is perpendicular to... The angle between the two axes in the plane.

[0034] Table 1. Theoretical DH parameters of RCM mechanism

[0035] The pose transformation matrix of the continuous endoscope is shown in equation (5): (5) In summary, the kinematic equations for the combined RCM mechanism and the continuous endoscope were obtained. These equations demonstrate that the continuous endoscope, compared to a rigid endoscope, possesses the ability to actively adjust its own posture, thus meeting the requirements for surgical lateral imaging from different perspectives. The velocity equation of the continuous endoscope during the tracking task is described as follows: (6) In the formula, Represents the linear and angular velocities of the endoscope tip in the base coordinate system; matrix This represents the Jacobian matrix of a continuous endoscope. This indicates the driving speed of a five-degree-of-freedom vehicle.

[0036] This continuum endoscope has one spin degree of freedom and one yaw degree of freedom, and its effective workspace is a spatial curved surface. The coverage range depends on the maximum spin angle and the continuum bending parameters. The bending angle range designed in this embodiment is [insert range here]. The spin angle range is Based on the existing RCM mechanism, a workspace diagram is drawn, such as... Figure 1 As shown.

[0037] In this invention, an image processing algorithm based on the YOLO neural network architecture is used to identify the tips of surgical instruments in endoscopic images. The single-stage detection network model in the YOLO series has been widely used in most object detection applications. The steps for training a YOLO model to identify needle tips in endoscopic images are: collecting data, labeling images to divide the dataset, setting training parameters, and training and validating the model's performance.

[0038] Five hundred images of randomly placed needle tips under different lighting conditions were collected using an endoscope (Octha). These images were then augmented to 2000 using data augmentation and image rotation techniques. A labeled dataset was generated using the Labelme tool, and the training, test, and validation sets were allocated at 70%, 15%, and 15% respectively. A YOLO model was trained using the PyTorch deep learning library, and the Adam optimization algorithm was used to accelerate the neural network training process. NVIDIA CUDA components were used to improve training and inference speed. The training results demonstrate that the model can identify the two-dimensional coordinates of needle tips from complex backgrounds.

[0039] Optionally, step two specifically includes: In visual servo control, a visual camera is typically used to acquire images and extract target features to perform servo tasks. An endoscope transforms the 3D coordinates of the target point in the world coordinate system to the camera coordinate system and projects it onto the camera's pixel plane. Let... This represents the position of the instrument's end point in the camera coordinate system. This represents the position of a two-dimensional point projected onto the image plane in the camera coordinate system. These are the coordinates of the feature point in the pixel plane. According to the principle of pinhole imaging, the transformation relationship between the three is shown in equation (7): (7) In the formula, , This represents the actual width and length corresponding to each pixel value. The reference coordinates representing the pixel coordinate system. This refers to the camera's focal length.

[0040] In image-based visual servoing, it is necessary to extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of a continuous endoscope based on the deviation of the feature points until the image features reach the target state. The image features that change over time are defined as... The target image features are At different times The characteristic error is defined as follows (8): (8) When the end-camera of a continuous endoscope has a spatial velocity that varies with time... At that time, the features of the corresponding pixel plane The time derivative and its relationship are described by equation (9): (9) in, The image Jacobian matrix is ​​composed of information about individual image feature points and the position information of the corresponding spatial points in the camera frame: (10) According to formulas (8) to (10), let the error converge exponentially, and the visual servo control law is expressed as follows: (11) In the formula, Represents the pseudo-inverse of a matrix. The convergence coefficient is . , This represents the rate of change of target image features over time; due to the inherent error tolerance of visual servoing, the image Jacobian matrix... The value can be set to a constant; combined with the kinematic model of the continuous endoscope itself, a visual servo follower controller from the drive space to the image space can be obtained: (12) (13) In the formula, This represents the pseudo-inverse of the Jacobian matrix of a continuous endoscope.

[0041] The above describes the image-driven autonomous motion control of a continuum endoscope. Compared to traditional rigid endoscopes, the flexible structure of the continuum enables agile movement within the eye, which is significant for the application of control methods. The control method based on endoscopic images adjusts the camera's spatial velocity according to characteristic errors to achieve target tracking.

[0042] In some embodiments, the present invention provides a computer-readable storage medium storing computer instructions that are executed by a processor as in any of the above embodiments, a method for controlling active endoscopic following motion.

[0043] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples (not an exhaustive list) of readable storage media may include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0044] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0045] Embodiments of the present invention may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in an endoscope active following motion control method according to various embodiments of the present invention as described in the "Exemplary Methods" section above.

[0046] The steps of the method of the present invention are not limited to the specific order described above, unless otherwise specifically stated. Furthermore, in some embodiments, the invention may also be implemented as a program recorded on a recording medium, the program comprising machine-readable instructions for implementing the method according to the invention. Therefore, the invention also covers recording media storing programs for performing the method according to the invention.

[0047] Although the invention has been described with reference to preferred embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, the technical features mentioned in the various embodiments can be combined in any manner as long as there is no structural conflict. The invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for controlling the active motion of an endoscope, characterized in that, Includes the following steps: Step 1: Perform kinematic modeling on the continuous endoscope to obtain the pose transformation matrix of the continuous endoscope, thereby obtaining the workspace of the continuous endoscope, and verifying the reachability of the workspace within the eyeball. Step 2: Construct a visual servo follower controller based on endoscopic images. Collect endoscopic images and use an image processing algorithm based on the YOLO neural network architecture to identify the surgical instrument tips in the endoscopic images. Extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of the continuous endoscope based on the deviation of the feature points until the image features reach the target state.

2. The endoscope active tracking motion control method according to claim 1, characterized in that, The kinematic modeling and pose transformation matrix of the continuous endoscope in step one includes: The homogeneous transformation of the continuous endoscope's motion in the XZ plane is decomposed into offset parameters of the X and Z axes of the connecting rods, using the chord length of the arc formed by the continuous endoscope. The homogeneous transformation matrix for the bending angle of the continuous endoscope is shown in equation (1): （1） The continuous endoscope as a whole has rotational degrees of freedom along the central axis, allowing the motion of the continuous endoscope in the two-dimensional plane to cover the three-dimensional space, generating a continuous set of workspace points. Its coordinate transformation relationship is shown in equation (2): （2） In the formula, The angle represents the rotational degree of freedom of the continuous endoscope; let the origin of this rotational degree of freedom coordinate system coincide with the telecentric fixed point of the initial continuous endoscope, and let this point have a fixed straight-line distance L relative to the root of the curved part of the continuous endoscope. This continuous endoscope is integrated into a three-degree-of-freedom (RCM) mechanism. The joint coordinate system of the RCM mechanism is located at the telecentric fixed point of the RCM mechanism. This coordinate system consists of three pairwise orthogonal coordinate axes: x-axis, y-axis, and z-axis. The RCM mechanism has three degrees of freedom, which, through pitch, roll, and feed, respectively, allow the end-effector to reach the target position to complete the operation. The transformation matrix between adjacent joints is... Referring to the Denavit-Hartenberg parametric method, the pose transformation matrix of the telecentric fixed point of the RCM mechanism is... As shown in equation (4): （3） （4） in, This represents the rotation angle of the RCM mechanism about the z-axis of the joint coordinate system. This represents the difference in the length of the common normal of the RCM mechanism along the z-axis of the joint coordinate system. This represents the distance between the common normals of two adjacent z-axis in the joint coordinate system of the RCM mechanism. This indicates that the joint coordinate system of the RCM mechanism is perpendicular to... The angle between the two axes in the plane; The pose transformation matrix of the continuous endoscope is shown in equation (5): （5）。 3. The endoscope active tracking motion control method according to claim 2, characterized in that, When the continuous endoscope performs a following task, the velocity equation is described as follows: （6） In the formula, Represents the linear and angular velocities of the endoscope tip in the base coordinate system; matrix This represents the Jacobian matrix of a continuous endoscope. This indicates the driving speed of a five-degree-of-freedom vehicle.

4. The endoscope active tracking motion control method according to claim 2, characterized in that, Step two specifically includes: The continuous endoscope transforms the three-dimensional coordinates of the target point in the world coordinate system to the camera coordinate system and projects them onto the camera's pixel plane; assuming This represents the position of the instrument's end point in the camera coordinate system. This represents the position of a two-dimensional point projected onto the image plane in the camera coordinate system. These are the coordinates of the feature point in the pixel plane. According to the principle of pinhole imaging, the transformation relationship between the three is shown in equation (7): （7） In the formula, , This represents the actual width and length corresponding to each pixel value. The reference coordinates representing the pixel coordinate system. The focal length of the camera; In image-based visual servoing, it is necessary to extract the deviation between the pixel spatial coordinates of feature points and the desired features in real time, and control the camera movement at the end of the continuous endoscope based on the deviation of the feature points until the image features reach the target state; the image features that change over time are defined as... The target image features are At different times The characteristic error is defined by equation (8): （8） When the end-camera of a continuous endoscope has a spatial velocity that varies with time... At that time, the features of the corresponding pixel plane The time derivative and its relationship are described by equation (9): （9） in, The image Jacobian matrix is ​​composed of information about individual image feature points and the position information of the corresponding spatial points in the camera frame: （10） According to formulas (8) to (10), let the error converge exponentially, and the visual servo control law is expressed as follows: （11） In the formula, Represents the pseudo-inverse of a matrix. The convergence coefficient is . , This represents the rate of change of target image features over time; due to the inherent error tolerance of visual servoing, the image Jacobian matrix... The value is set to a constant; combined with the kinematic model of the continuous endoscope itself, a visual servo follower controller from the drive space to the image space is obtained: （12） （13） In the formula, This represents the pseudo-inverse of the Jacobian matrix of a continuous endoscope.

5. The endoscope active tracking motion control method according to claim 4, characterized in that, The continuous endoscope has spin and yaw degrees of freedom, and its effective working space is a spatial curved surface.

6. The endoscope active tracking motion control method according to claim 5, characterized in that, The continuous endoscope includes a connected distal fixed section, an intermediate continuous flexible section, and an end camera, wherein the intermediate continuous flexible section is flexible.

7. The endoscope active tracking motion control method according to claim 6, characterized in that, The continuous endoscope is mounted on the RCM mechanism and moves flexibly within the entire working space inside the eyeball to acquire intraoperative images in different end-effector postures.

8. The endoscope active tracking motion control method according to claim 7, characterized in that, The bending angle range of the continuous endoscope is: The spin angle range is .

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are executed by a processor according to any one of claims 1 to 8.

10. A computer program product, characterized in that, The computer program product stores computer instructions, which are executed by a processor using the method as described in any one of claims 1 to 8.

Images