Bronchoscope robot simulation platform and method based on VTK

Abstract

The invention relates to the technical field of robot simulation and control, in particular to a bronchoscope robot simulation control platform and method based on VTK, and is suitable for structural modeling, motion control, virtual simulation and the like of continuum robots (such as bronchoscope robots). The method can be widely applied to the fields of medical robot simulation, operation training, robot control algorithm verification and the like. Firstly, the platform adopts a voxelization modeling technology to carry out modeling on a bronchoscope robot, and forms and behaviors of different continuum robots can be simulated only by setting structural parameters and freedom of motion of the robot. The center line of the bronchial tree is rapidly extracted by using a vmtk library, and a motion control algorithm based on the center line of the bronchial tree is designed, so that the simulation platform can effectively simulate interaction between the robot and the bronchial tree. Meanwhile, the collision condition of the robot and the bronchial tree is calculated in real time through a collision detection interface built in the vtk, and real-time accurate feedback is conducted in the simulation process.

Description

Technical Field

[0001] The present invention relates to the technical field of robot simulation and control, and particularly relates to a bronchoscope robot simulation control platform and method based on VTK (Visualization Toolkit), which is applicable to the structural modeling, motion control, virtual simulation, etc. of continuum robots (such as bronchoscope robots), and can be widely applied to fields such as medical robot simulation, surgical training, and robot control algorithm verification. Background Art

[0002] Bronchoscope surgery, as an important means of minimally invasive diagnosis and treatment, is widely used in the diagnosis and treatment of lung diseases, such as early screening of lung cancer, treatment of bronchial stenosis, etc. A bronchoscope robot is a typical continuum robot, and its flexible structure and multi-degree-of-freedom motion characteristics enable it to flexibly navigate in the complex bronchial tree structure and complete operations such as lesion localization and tissue sampling. However, due to the complex structural environment of the human bronchial tree, the motion control and simulation technology of bronchoscope robots still face many challenges.

[0003] Firstly, the complexity and individual differences of the bronchial tree structure make it extremely difficult to plan the path and control the motion of the robot. The branch structure of the bronchial tree is narrow and variable, and the robot needs to complete precise motions in a limited space while avoiding collisions with the tube wall. Secondly, the kinematic modeling of continuum robots itself is highly non-linear. Especially in the case of multi-degree-of-freedom motion, the inverse kinematic solution is computationally complex and difficult to control in real time. In addition, the bronchoscope robot needs to interact with soft tissues during actual operation, and the simulation and feedback of this interaction force are one of the difficulties of existing simulation technologies. Existing simulation platforms often have difficulty accurately simulating the dynamic interaction behavior between the robot and tissues, resulting in a large deviation between the simulation results and actual operations.

[0004] In the prior art, the simulation and navigation of bronchoscope surgery mainly rely on the scheme of preoperative static model + intraoperative real-time position. Specifically, a static three-dimensional model of the bronchial tree is reconstructed through medical images (such as CT), and the real-time navigation of the robot is realized by combining electromagnetic positioning or optical tracking technology during the operation. However, this scheme has obvious limitations: firstly, the static model cannot reflect the influence of respiratory motion or tissue deformation on the surgical path; secondly, due to the lack of preoperative simulation function, doctors cannot pre-act and evaluate the motion path, collision detection, and operation feasibility of the robot before the operation, which limits the safety and efficiency of the operation. For doctors, the static model cannot provide the opportunity for preoperative simulation operations, and doctors can only rely on experience and real-time images for adjustment during the operation, increasing the surgical risk and learning curve. And physical learning resources (such as cadaver models or animal experiments) are extremely costly, consume a large amount, and are difficult to meet the large-scale training needs.

[0005] At present, although there are some surgical robot simulation platforms, there is no dedicated bronchoscope robot simulation platform. For general simulation tools, their visualization effects and interaction performance are limited, making it difficult to meet the requirements for high precision, real-time performance, and immersion in medical simulation. Especially in aspects such as tissue interaction simulation, the existing technologies often cannot provide sufficient precision and real-time performance, resulting in a large gap between the simulation results and actual operations.

[0006] "A VR bronchoscopy and treatment teaching platform": This patent relates to a VR bronchoscopy and treatment teaching platform, which includes a VR handle, VR glasses, a storage unit, a scoring unit, and a system unit. Through the immersive VR system, during conventional training and teaching processes, based on the bronchial model and bronchoscope positioning points, this platform helps users observe and learn about bronchial structure recognition and lesion treatment. The system can select random treatment points and can adjust the shape of the bronchial model as needed to achieve a more realistic teaching process and immersive operation experience.

[0007] "Interventional surgical robot simulation system": This patent relates to an interventional surgical robot simulation system, which includes a control module, a state simulation module, and a robotic arm simulation module. The control module communicates with the state simulation module and the robotic arm simulation module and is responsible for sending control commands to them. The state simulation module receives the commands and simulates and adjusts the states of peripheral devices, while the robotic arm simulation module obtains motion parameters according to the commands and simulates and executes corresponding surgical actions. By simulating the complete working process, this system can debug the entire process without the need for hardware.

[0008] The existing technologies have multiple limitations in the field of bronchoscope robot simulation, and these problems significantly affect the efficiency and practicality of simulation. First of all, there is a lack of a dedicated simulation platform for bronchoscope robots. The existing simulation platforms are usually based on traditional laparoscopic robot robotic arms or other types of robots and cannot fully adapt to the characteristics of continuum robots. Especially for bronchoscope robots, traditional platforms require manual modeling and setting of kinematic parameters, which not only involves a large amount of work but also has extremely limited flexibility and scalability for complex continuum robots. In addition, traditional simulation systems usually have difficulty handling the interaction between the robot and the bronchial tree, such as the lack of collision detection and motion feedback functions, which makes the simulation platform unable to truly reproduce the motion situation and operation accuracy of the robot in a complex airway environment, affecting the simulation effect and reliability of actual operations. Summary of the Invention

[0009] The technical problem to be solved by the present invention is: To overcome the deficiencies of the existing technologies, a bronchoscope robot simulation platform and method based on VTK are proposed.

[0010] The technical solution of the present invention is:

[0011] A bronchoscope robot simulation platform based on VTK. The simulation platform includes a three-dimensional reconstruction module, a centerline extraction module, a voxelization modeling module, a data reading module, a scene arrangement module, a motion processing module, and a rendering and refreshing module;

[0012] The three-dimensional reconstruction module is used to receive the lung CT scan data given by the user, perform three-dimensional reconstruction on the received CT scan data to obtain a three-dimensional model, and then output the obtained three-dimensional model to the centerline extraction module and the scene arrangement module;

[0013] The centerline extraction module is used to receive the three-dimensional model output by the three-dimensional reconstruction module, perform centerline extraction on the received three-dimensional model to obtain centerline data, and then output the obtained centerline data to the scene arrangement module;

[0014] The voxelization modeling module is used to receive the robot parameters set by the user, establish a voxelized robot model according to the received robot parameters, and then output the voxelized robot model to the scene arrangement module;

[0015] The data reading module is used to receive the real-time control data of the external handle, perform normalization processing on the received real-time control data to obtain standard external input data, and then output the obtained standard external input data to the scene arrangement module;

[0016] The scene arrangement module is used to receive the voxelized robot model output by the voxelization modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the three-dimensional model output by the three-dimensional reconstruction module, and use the VTK library to build a simulation scene according to the voxelized robot model, the three-dimensional model, and the centerline data; The scene arrangement module is also used to initialize the virtual camera of the robot; The scene arrangement module is also used to process the received standard external input data to obtain a robot motion signal, and then output the obtained robot motion signal to the motion processing module;

[0017] The motion processing module is used to receive the robot motion signal output by the scene arrangement module, process the received robot motion signal to obtain a VTK transformation matrix, and then output the obtained VTK transformation matrix to the rendering and refreshing module;

[0018] The rendering and refreshing module is used to receive the VTK transformation matrix output by the motion processing module, update the spatial position of the robot using the VTK transformation matrix, and refresh the virtual camera and the simulation scene of the robot according to the updated spatial position.

[0019] The method by which the motion processing module processes the received robot motion signal to obtain a VTK transformation matrix is:

[0020] Step S1: Obtain the robot motion signal H and the pose state T of the robot at this time.

[0021] Step S2: Use the obb-based collision detection algorithm built into VTK to detect the coincidence situation S between the robot and the bronchial tree model at this time.

[0022] Step S3: According to the coincidence situation S, quickly calculate the relative position information between each part of the robot and the central axis of the bronchial tree through the KD tree algorithm, and use the robot motion signal H to deduce the pose change T0 of the robot at this moment.

[0023] Step S4: Perform an operation on the pose state T and the pose change T0 to obtain the pose state T' at the next moment, and process T' into the standard matrix format, which is the VTK transformation matrix.

[0024] The method for the voxelization modeling module to establish a voxelized robot model according to the received robot parameters is as follows:

[0025] Step S1: Obtain the input robot parameters R and L; R is the radius of the robot, and L is the length of the central axis of the robot.

[0026] Step S2: Create an empty matrix list.

[0027] Step S3: Create an empty boolean matrix, calculate the coordinates of the points on the robot axis in the created empty boolean matrix according to the equal-curvature bending hypothesis when bending at a certain angle, and set the boolean value at the calculated coordinate position to 1.

[0028] Step S4: Traverse all the points within the spherical region with a radius of R around each point on the robot axis, and set the boolean value at its position to 1 to obtain a boolean matrix corresponding to the robot model.

[0029] Step S5: Repeat steps S3 - S4 with different angles, and store the boolean matrices corresponding to the robot models bent at different angles into the empty matrix list created in step S2.

[0030] Step S6: Add the transformation matrices of rotation and displacement to the robot in the matrix obtained in step S5 to obtain a new matrix, and the robot bends, rotates, or displaces according to the obtained new matrix.

[0031] The method for the centerline extraction module to extract the centerline from a three-dimensional model is as follows: Use the VMTK library, select the initial point and the end point, and calculate the centerline data.

[0032] A bronchoscope robot simulation method based on VTK, the steps of this method include:

[0033] First step, the 3D reconstruction module receives the lung CT scan data given by the user, performs 3D reconstruction on the received CT scan data to obtain a 3D model, and then outputs the obtained 3D model to the centerline extraction module and the scene arrangement module;

[0034] Second step, the centerline extraction module receives the 3D model output by the 3D reconstruction module, performs centerline extraction on the received 3D model to obtain centerline data, and then outputs the obtained centerline data to the scene arrangement module;

[0035] Third step, the voxelization modeling module receives the robot parameters set by the user, builds a voxelized robot model based on the received robot parameters, and then outputs the voxelized robot model to the scene arrangement module;

[0036] Fourth step, the data reading module receives the real-time control data of the external handle, performs normalization processing on the received real-time control data to obtain standard external input data, and then outputs the obtained standard external input data to the scene arrangement module;

[0037] Fifth step, the scene arrangement module receives the voxelized robot model output by the voxelization modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the 3D model output by the 3D reconstruction module, and uses the VTK library to build a simulation scene based on the voxelized robot model, 3D model, and centerline data;

[0038] Sixth step, the scene arrangement module initializes the virtual camera of the robot;

[0039] Seventh step, the scene arrangement module processes the received standard external input data to obtain a robot motion signal, and then outputs the obtained robot motion signal to the motion processing module;

[0040] Eighth step, the motion processing module receives the robot motion signal output by the scene arrangement module, processes the received robot motion signal to obtain a VTK transformation matrix, and then outputs the obtained VTK transformation matrix to the rendering and refreshing module;

[0041] Ninth step, the rendering and refreshing module receives the VTK transformation matrix output by the motion processing module, uses the VTK transformation matrix to update the spatial position of the robot, and refreshes the virtual camera and the simulation scene of the robot according to the updated spatial position.

[0042] Beneficial effects

[0043] In view of the deficiencies of the prior art, an innovative bronchoscope robot simulation platform is proposed. Developed based on the vtk library, it can solve the deficiencies of traditional simulation platforms to a certain extent. First, the platform uses voxelization modeling technology to model the bronchoscope robot. By simply setting the structural parameters and degrees of freedom of motion of the robot, it can simulate the morphology and behavior of different continuum robots. The vmtk library is used to quickly extract the centerline of the bronchial tree, and a motion control algorithm based on the centerline of the bronchial tree is designed, enabling the simulation platform to effectively simulate the interaction between the robot and the bronchial tree. At the same time, by using the built-in collision detection interface of vtk, the collision situation between the robot and the bronchial tree is calculated in real time and a real-time and accurate feedback is given during the simulation process. In addition, a mapping algorithm for controlling the robot with a handle is designed for the platform. Finally, all functions are connected in series to build a complete simulation system, which can well meet the situations of different robot configurations and bronchial tree environments for surgical simulation.

[0044] In the present invention, the robot adopts the voxelization modeling method, which can adapt to different types of continuum robot models, significantly improve the calculation efficiency and reduce the simulation calculation time, ensuring that the simulation system has high efficiency and flexible adaptability. The present invention provides an integrated bronchoscope motion simulation platform, which can support functions such as preoperative teaching, path planning, collision detection, and operation testing, providing more accurate preoperative simulation for doctors, optimizing the surgical path, and improving safety and efficiency. The designed platform provides the bronchoscope robot motion simulation function, which can be used for operations such as preoperative teaching, path planning, and testing and verification of control methods. By real-time simulating the motion of the bronchoscope in the complex bronchial tree, the platform can help doctors evaluate the motion path of the robot and the feasibility of surgical operations in advance. The present invention adopts the voxelization modeling method, and the platform can flexibly adapt to various types of continuum robot models. By representing the robot as an arrangement combination of voxel grids, not only can the modeling process of complex morphologies be effectively simplified, but also the calculation efficiency can be significantly improved. The voxelization model has a faster calculation speed and can quickly process collision detection and motion feedback with the environment during the simulation process, thereby reducing the simulation calculation time and providing more timely feedback for actual operations. This will help improve the planning accuracy before surgery, help doctors familiarize themselves with the operation process, reduce the risks in actual surgeries, and improve the efficiency and safety of surgeries. This simulation platform is developed based on vtk (Visualization Toolkit), has the characteristics of small volume and high performance, and is very suitable for the needs of the medical simulation field. vtk provides powerful visualization functions, and the platform is also convenient for subsequent expansion and function integration. Brief Description of the Drawings

[0045] Figure 1 It is a schematic diagram of the system composition of the present invention;

[0046] Figure 2It is a schematic diagram of a simulation scenario. Detailed implementation manner

[0047] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0048] Embodiment

[0049] As Figure 1 shown, a bronchoscope robot simulation platform based on VTK, the simulation platform includes a three-dimensional reconstruction module, a centerline extraction module, a voxelization modeling module, a data reading module, a scene arrangement module, a motion processing module, and a rendering and refreshing module;

[0050] The three-dimensional reconstruction module is used to receive the lung CT scan data given by the user, perform three-dimensional reconstruction on the received CT scan data to obtain a three-dimensional model, and then output the obtained three-dimensional model to the centerline extraction module and the scene arrangement module;

[0051] The centerline extraction module is used to receive the three-dimensional model output by the three-dimensional reconstruction module, perform centerline extraction on the received three-dimensional model to obtain centerline data, and then output the obtained centerline data to the scene arrangement module;

[0052] The voxelization modeling module is used to receive the robot parameters set by the user R = 1.5 mm and the central axis length L = 35 mm, and establish a voxelized robot model according to the received robot parameters, and then output the voxelized robot model to the scene arrangement module;

[0053] The data reading module is used to receive the real-time control data of the external handle, the data includes the left joystick coordinates and the right joystick coordinates, perform normalization processing on the received real-time control data to obtain standard external input data, and then output the obtained standard external input data to the scene arrangement module;

[0054] The scene arrangement module is used to receive the voxelized robot model output by the voxelization modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the three-dimensional model output by the three-dimensional reconstruction module, and use the VTK library to build a simulation scene according to the voxelized robot model, the three-dimensional model, and the centerline data, as Figure 2As shown in the figure; the voxelized robot end is set at the origin in space, and the orientation is the positive direction of the X-axis. The entry point of the three-dimensional model is set at the origin in space, and the orientation is the same as the initial movement direction of the voxelized robot. The scene arrangement module is also used to initialize the virtual camera of the robot, use the VTK library camera module and bind the camera to the head point position of the voxelized robot; the scene arrangement module is also used to process the received standard external input data to obtain the robot motion signal, corresponding to the three degrees of freedom of rotation, bending, and advancing and retreating of the voxelized robot, and then output the obtained robot motion signal to the motion processing module;

[0055] The motion processing module is used to receive the robot motion signal output by the scene arrangement module, process the received robot motion signal to obtain the VTK transformation matrix, and then output the obtained VTK transformation matrix to the rendering and refreshing module;

[0056] The rendering and refreshing module is used to receive the VTK transformation matrix output by the motion processing module, and use the VTK transformation matrix to update the spatial position of the robot, and refresh the virtual camera and the simulation scene of the robot according to the updated spatial position.

[0057] The method for the motion processing module to process the received robot motion signal to obtain the VTK transformation matrix is as follows:

[0058] Step S1, obtain the robot motion signal H and the pose state T of the robot at this time;

[0059] Step S2, use the VTK built-in obb-based collision detection algorithm to detect the coincidence situation S between the robot and the bronchial tree model at this time;

[0060] Step S3, according to the coincidence situation S, quickly calculate the relative position information between each part of the robot and the central axis of the bronchial tree through the KD tree algorithm, and use the robot motion signal H to deduce the pose change T0 of the robot at this moment;

[0061] Step S4, perform an operation on the pose state T and the pose change T0 to obtain the pose state T' at the next moment, and process T' into the standard matrix format, which is the VTK transformation matrix.

[0062] The method for the voxelization modeling module to establish the voxelized robot model according to the received robot parameters is as follows:

[0063] Step S1, obtain the input robot parameters R, L; R is the radius of the robot, and L is the length of the central axis of the robot;

[0064] Step S2, create an empty matrix list;

[0065] Step S3: Create an empty boolean matrix. Calculate the coordinates of the points on the robot's axis when bent at a certain angle according to the equal-curvature bending hypothesis in the created empty boolean matrix, and set the boolean value at the calculated coordinate position to 1.

[0066] Step S4: Traverse all the points within a spherical region with a radius of R around each point on the robot's axis, and set the boolean value at their positions to 1 to obtain a boolean matrix corresponding to the robot model.

[0067] Step S5: Repeat Steps S3 - S4 at different angles, and store the boolean matrices corresponding to the robot models bent at different angles into the empty matrix list created in Step S2.

[0068] Step S6: Add a transformation matrix for rotation and displacement to the robot in the matrix obtained in Step S5 to get a new matrix, and the robot bends, rotates, or displaces according to the obtained new matrix.

[0069] The method for the centerline extraction module to extract the centerline of the 3D model is as follows: Use the VMTK library, select the initial point and the end point, and calculate the centerline data.

[0070] A bronchoscope robot simulation method based on VTK, the steps of this method include:

[0071] The first step: The 3D reconstruction module receives the lung CT scan data given by the user, performs 3D reconstruction on the received CT scan data to obtain a 3D model, and then outputs the obtained 3D model to the centerline extraction module and the scene arrangement module.

[0072] The second step: The centerline extraction module receives the 3D model output by the 3D reconstruction module, extracts the centerline of the received 3D model to obtain the centerline data, and then outputs the obtained centerline data to the scene arrangement module.

[0073] The third step: The voxelization modeling module receives the robot parameters set by the user, and establishes a voxelized robot model according to the received robot parameters, and then outputs the voxelized robot model to the scene arrangement module.

[0074] The fourth step: The data reading module receives the real-time control data of the external handle, performs normalization processing on the received real-time control data to obtain the standard external input data, and then outputs the obtained standard external input data to the scene arrangement module.

[0075] In the fifth step, the scene arrangement module receives the voxelized robot model output by the voxelization modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the 3D model output by the 3D reconstruction module, and uses the VTK library to build a simulation scene based on the voxelized robot model, the 3D model, and the centerline data;

[0076] In the sixth step, the scene arrangement module initializes the virtual camera of the robot;

[0077] In the seventh step, the scene arrangement module processes the received standard external input data to obtain a robot motion signal, and then outputs the obtained robot motion signal to the motion processing module;

[0078] In the eighth step, the motion processing module receives the robot motion signal output by the scene arrangement module, processes the received robot motion signal to obtain a VTK transformation matrix, and then outputs the obtained VTK transformation matrix to the rendering and refreshing module;

[0079] In the ninth step, the rendering and refreshing module receives the VTK transformation matrix output by the motion processing module, uses the VTK transformation matrix to update the spatial position of the robot, and refreshes the virtual camera and the simulation scene of the robot according to the updated spatial position.

[0080] In summary, the above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A VTK-based bronchoscopy robot simulation platform, characterized by: The simulation platform includes a 3D reconstruction module, a centerline extraction module, a voxel modeling module, a data reading module, a scene arrangement module, a motion processing module, and a rendering refresh module; The three-dimensional reconstruction module is used to receive lung CT scan data provided by the user, and perform three-dimensional reconstruction on the received CT scan data to obtain a three-dimensional model, and then output the obtained three-dimensional model to the centerline extraction module and the scene arrangement module; The centerline extraction module is used to receive the three-dimensional model output by the three-dimensional reconstruction module, and extract the centerline of the received three-dimensional model to obtain centerline data, and then output the obtained centerline data to the scene arrangement module; The voxel modeling module is used to receive robot parameters set by the user, and to establish a voxel robot model according to the received robot parameters, and then output the voxel robot model to the scene arrangement module; The data reading module is used to receive the real-time control data of the external handle, and normalize the received real-time control data to obtain standard external input data, and then output the obtained standard external input data to the scene arrangement module; The scene arrangement module is used to receive the voxelized robot model output by the voxelized modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the three-dimensional model output by the three-dimensional reconstruction module, and to build a simulation scene using the VTK library according to the voxelized robot model, the three-dimensional model, and the centerline data; the scene arrangement module is also used to initialize the virtual camera of the robot; the scene arrangement module is also used to process the received standard external input data to obtain a robot motion signal, and then output the obtained robot motion signal to the motion processing module; The motion processing module is used to receive the robot motion signal output by the scene arrangement module, process the received robot motion signal to obtain a VTK transformation matrix, and then output the obtained VTK transformation matrix to the rendering refresh module; The rendering refresh module is used to receive the VTK transformation matrix output by the motion processing module, and use the VTK transformation matrix to update the robot's spatial position, and refresh the robot's virtual camera and simulation scene according to the updated spatial position.

2. The VTK-based bronchoscopy robot simulation platform according to claim 1, characterized in that: The method by which the motion processing module processes the received robot motion signal to obtain the VTK transformation matrix is: Step S1, obtaining the robot motion signal H and the robot's posture state T at this time; Step S2, using the built-in obb-based collision detection algorithm of VTK to detect the overlap S between the robot and the bronchial tree model at this time; Step S3, according to the overlap situation S, the relative position information between each part of the robot and the central axis of the bronchial tree is quickly calculated by the KD tree algorithm, and the robot motion signal H is used to deduce the posture change T0 of the robot at this moment; Step S4, calculate the posture state T and the posture change T0 to obtain the posture state T' at the next moment, and process T' into a standard matrix format, that is, the VTK transformation matrix.

3. The VTK-based bronchoscopy robot simulation platform according to claim 1, characterized in that: The method for the voxel modeling module to establish a voxel robot model according to the received robot parameters is: Step S1, obtaining input robot parameters R, L; R is the robot radius, L is the robot centerline length; Step S2, create an empty matrix list; Step S3, creating an empty Boolean matrix, calculating the coordinates of the points on the robot axis in the created empty Boolean matrix when bending at a certain angle according to the equal curvature bending assumption, and setting the Boolean values ​​of the calculated coordinate positions to 1; Step S4, traverse all points in a spherical area with a radius of R around each point on the robot axis, and set the Boolean value of its position to 1, to obtain a Boolean matrix corresponding to the robot model; Step S5, repeating steps S3-S4 with different angles, and storing the Boolean matrices corresponding to the robot models bent at different angles into the empty matrix list created in step S2; Step S6, adding the transformation matrix of rotation and displacement to the robot in the matrix obtained in step S5 to obtain a new matrix, and the robot bends, rotates or displaces according to the obtained new matrix.

4. The VTK-based bronchoscopy robot simulation platform according to claim 1, characterized in that: The method for extracting the centerline of the three-dimensional model by the centerline extraction module is: using the VMTK library, selecting the initial point and the end point, and calculating the centerline data.

5. A VTK-based bronchoscopic robot simulation method, characterized in that The steps of the method include: In the first step, the 3D reconstruction module receives the lung CT scan data provided by the user, and performs 3D reconstruction on the received CT scan data to obtain a 3D model, and then outputs the obtained 3D model to the centerline extraction module and the scene arrangement module; In the second step, the centerline extraction module receives the 3D model output by the 3D reconstruction module, extracts the centerline of the received 3D model, obtains the centerline data, and then outputs the obtained centerline data to the scene arrangement module; In the third step, the voxel modeling module receives the robot parameters set by the user, builds a voxel robot model according to the received robot parameters, and then outputs the voxel robot model to the scene arrangement module; Step 4: The data reading module receives the real-time control data of the external handle, normalizes the received real-time control data, obtains standard external input data, and then outputs the obtained standard external input data to the scene arrangement module; In the fifth step, the scene arrangement module receives the voxelized robot model output by the voxelized modeling module, the standard external input data output by the data reading module, the centerline data input by the centerline extraction module, and the 3D model output by the 3D reconstruction module, and uses the VTK library to build a simulation scene based on the voxelized robot model, the 3D model, and the centerline data; Step 6: The scene arrangement module initializes the robot's virtual camera; Step 7: The scene arrangement module processes the received standard external input data to obtain a robot motion signal, and then outputs the obtained robot motion signal to the motion processing module; In the eighth step, the motion processing module receives the robot motion signal output by the scene arrangement module, processes the received robot motion signal to obtain a VTK transformation matrix, and then outputs the obtained VTK transformation matrix to the rendering refresh module; In the ninth step, the rendering refresh module receives the VTK transformation matrix output by the motion processing module, and uses the VTK transformation matrix to update the robot's spatial position, and refreshes the robot's virtual camera and simulation scene according to the updated spatial position.

Images