Autonomous joint replacement surgical robot navigation and positioning system

Abstract

Provided in the present application is an autonomous joint replacement surgical robot navigation and positioning system, comprising: a three-dimensional import unit, configured to import a three-dimensional bone image of a knee joint; a prosthesis planning unit, configured to plan a knee joint prosthesis on the basis of the three-dimensional bone image; a bone alignment unit, configured to align the three-dimensional bone image with a bone to undergo osteotomy in a real surgery scenario; a path planning unit, configured to plan an osteotomy path by means of an artificial potential field method; and a mechanical arm control unit, configured to control, on the basis of the planned osteotomy path, a mechanical arm to complete automated osteotomy. On the basis of bone alignment, the present application maps osteotomy planes into real surgery scenarios, so as to complete the automatic osteotomy operation, thereby achieving automated operation of the autonomous joint replacement surgical robot navigation and positioning system during surgery processes, reducing human intervention and further reducing errors caused by human intervention, and thus improving the accuracy, stability and efficiency of surgeries.

Description

An autonomous joint replacement surgical robot navigation and positioning system Technical Field

[0001] The present application relates to the field of medical device technology, and in particular to a navigation and positioning system for an autonomous joint replacement surgical robot. Background Art

[0002] Joint replacement surgery is a common and complex surgical procedure used to treat joint diseases and injuries. Traditional joint replacement surgery typically requires surgeons to rely on extensive experience and visual judgment, but this approach has certain limitations and risks. In recent years, with the development of robotics and computer-assisted surgery, joint replacement surgical robots have gradually become an emerging technology, bringing significant changes and advancements to joint replacement surgery.

[0003] Currently, with the introduction of joint replacement surgical robots, they only participate in preoperative planning, and these intraoperative operations are generally performed manually. This surgical method is slow, and due to differences in surgical experience, the surgical level of doctors varies, and the surgical results may be less than expected.

[0004] Summary of the Invention

[0005] The problem that this application solves is that current manual surgery has uneven surgical skills due to individual differences, and the surgical results are not as good as expected.

[0006] To solve the above problems, the present application provides an autonomous joint replacement surgical robot navigation and positioning system, which includes:

[0007] a three-dimensional import unit configured to import a three-dimensional bone image of a knee joint;

[0008] a prosthesis planning unit configured to plan the position, model and osteotomy plane posture information of the knee joint prosthesis based on the three-dimensional bone image;

[0009] a bone registration unit configured to register the three-dimensional bone image with the bone to be osteotomized in the actual surgical scene, and determine osteotomy plane pose information of the bone to be osteotomized;

[0010] A path planning unit is configured to plan an osteotomy path by an artificial potential field method according to the osteotomy plane posture information of the bone to be osteotomized;

[0011] The robotic arm control unit is configured to control the robotic arm to complete automatic osteotomy according to the planned osteotomy path.

[0012] Furthermore, the bone registration unit is specifically configured to:

[0013] Obtain a two-dimensional image from an RGB binocular camera, the two-dimensional image containing a visual marker, and obtain visual marker positioning data from an infrared binocular camera;

[0014] Establishing an initial posture relationship between the probe and the three-dimensional skeleton model according to at least one three-dimensional skeleton key point corresponding to the three-dimensional skeleton image, wherein the relative positions of the probe and the three-dimensional skeleton model are fixed under the initial posture relationship;

[0015] Based on the visual marker, the visual marker positioning data, the two-dimensional image of the RGB binocular camera, and the three-dimensional bone model, digital twin registration is performed on the three-dimensional bone model and the bone to be osteotomized in the real surgical scene;

[0016] Using the probe to touch the surface of the bone to be osteotomized, and determining the touch position coordinates in the infrared binocular camera coordinate system based on the shape data of the probe and the visual marker positioning data of the probe in the infrared binocular camera coordinate system at the time of touch;

[0017] Registering the touch position coordinates with the three-dimensional skeleton model through an iterative closest point algorithm to determine a mapping relationship between the three-dimensional skeleton model and the touch position coordinates;

[0018] Based on the mapping relationship between the three-dimensional bone model and the touch position coordinates, the planned osteotomy plane posture information is mapped to the local coordinate system of the marking and positioning frame of the bone to be osteotomized.

[0019] Furthermore, when performing digital twin registration on the three-dimensional skeleton model and the bone to be osteotomized in the real surgical scene, the skeleton registration unit is specifically configured as follows:

[0020] Identifying coordinate information of a visual marker of the probe in the two-dimensional image, and determining a mapping relationship between a local coordinate system of the probe and a coordinate system of the RGB camera;

[0021] According to the mapping relationship between the local coordinate system of the probe and the RGB camera coordinate system, the three-dimensional bone model is projected into the two-dimensional image in the RGB camera coordinate system, where the two-dimensional image is a real surgical scene image in the RGB camera coordinate system;

[0022] Adjusting the posture of the probe to change the posture of the three-dimensional bone model until the projection of the three-dimensional bone model on the two-dimensional image coincides with the bone to be osteotomized on the two-dimensional image;

[0023] When the projection of the three-dimensional bone model coincides with the bone to be osteotomized, obtaining positioning data of the visual marker of the probe in the coordinate system of the infrared binocular camera;

[0024] Determine a mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized based on the positioning data of the visual marker of the probe in the coordinate system of the infrared binocular camera and the positioning data of the visual marker of the marking and positioning frame of the bone to be osteotomized in the coordinate system of the infrared binocular camera;

[0025] Based on the mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized, the coordinate system of the three-dimensional bone model is converted to the local coordinate system of the marking and positioning frame of the bone to be osteotomized to complete the digital twin alignment.

[0026] Furthermore, in the bone registration unit, when the bone to be osteotomized includes a tibia to be osteotomized and a femur to be osteotomized, the three-dimensional image of the tibia in the three-dimensional bone model and the tibia to be osteotomized in the real surgical scene, and the three-dimensional image of the femur in the three-dimensional bone model and the femur to be osteotomized in the real surgical scene are respectively registered through the digital twin registration process.

[0027] Furthermore, the path planning unit, when planning the osteotomy path by the artificial potential field method according to the osteotomy plane posture information of the bone to be osteotomized, is specifically configured as follows:

[0028] Generate osteotomy plane boundaries and soft tissue protection areas based on the osteotomy plane posture information of the bone to be osteotomized and the three-dimensional bone model;

[0029] Determine the calculation formula for the osteotomy path, set the distal end of the path surface as the target position, and set the soft tissue protection area as the gravitational field;

[0030] The osteotomy plane boundary is set as a strong constraint safety strategy, and a plurality of waypoints are generated based on the strong constraint safety strategy. The combination of the waypoints is the planned osteotomy path.

[0031] Furthermore, the calculation formula of the osteotomy path is: F a =k a (x g -x)

[0032] Among them, F a is the gravitational potential field, F r is the repulsive potential field, m is the potential field constant, k a is the proportional coefficient, x g is the target position, x is the current position, k ris a positive coefficient, x0 is the obstacle position, ρ(x) is the distance from the current position to the obstacle, and ρ0 is the impact distance.

[0033] Furthermore, when generating a plurality of waypoints based on the strong constraint safety policy, the path planning unit is specifically configured to:

[0034] Generate a waypoint based on the gravitational field and repulsive field;

[0035] Determine whether the generated waypoint is within the boundary of the osteotomy plane;

[0036] When the waypoint is outside the boundary of the osteotomy plane, the waypoint is translated into the boundary of the osteotomy plane along the direction toward the centroid of the gravitational field;

[0037] When the waypoint is within the boundary of the osteotomy plane, the next waypoint is generated according to the attraction field and the repulsion field;

[0038] Determine whether the currently generated waypoint reaches the distal end of the osteotomy plane boundary;

[0039] In the case where the waypoint does not reach the distal end of the osteotomy plane boundary, returning to the process of determining whether the generated waypoint is located within the osteotomy plane boundary;

[0040] When the waypoint reaches the distal end of the osteotomy plane boundary, updating the gravitational field, the repulsive field, and the osteotomy plane boundary;

[0041] When the remaining osteotomy area is greater than the preset threshold, the process returns to the process of generating a waypoint according to the gravitational field and the repulsive field.

[0042] Furthermore, the robotic arm control unit is specifically configured to:

[0043] Obtaining a calibrated osteotomy tool coordinate system;

[0044] Mapping the planned osteotomy path into an osteotomy path in a robot arm flange coordinate system based on the osteotomy tool coordinate system;

[0045] Based on the kinematic constraint formula, the osteotomy path in the robot arm flange coordinate system is converted into the joint angle trajectory of each joint;

[0046] Interpolate the joint angle trajectory of each joint;

[0047] The feedforward PID controller outputs the joint angular velocity of each joint angle to control the robot arm;

[0048] Acquire the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized in real time;

[0049] The target joint angle trajectory is updated based on the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized, and the interpolation process of the joint angle trajectory of each joint is returned.

[0050] Furthermore, the PID controller is expressed as:

[0051] in, is the joint angular velocity output by the proportional-integral-derivative controller, q r is the target joint angle trajectory, K p , K i , K d For controller parameters.

[0052] Furthermore, when importing the three-dimensional bone image of the knee joint, the three-dimensional import unit is specifically configured to:

[0053] Acquire CT images of the knee joint of the joint replacement patient;

[0054] Perform image segmentation on the bone part and non-bone part in the knee joint CT image to obtain the segmentation result of the bone part;

[0055] Based on the connected domain analysis, the skeleton is segmented and the skeleton structures of the tibia, femur, patella and fibula are extracted.

[0056] A three-dimensional bone image of the knee joint is created based on the extracted bone structures of the tibia, femur, patella, and fibula. The three-dimensional bone image includes three-dimensional images of the tibia, femur, patella, and fibula.

[0057] In this application, based on high-precision bone alignment before and during surgery, the osteotomy plane posture information can be mapped to the bone to be osteotomized in the real surgical scene, and the robotic arm can be controlled to complete the automatic osteotomy operation, thereby realizing the automated operation of the autonomous joint replacement surgical robot navigation and positioning system during the operation, reducing human intervention, and thus reducing the errors caused by human intervention, thereby improving the accuracy, stability and efficiency of the operation.

[0058] In this application, 3D imaging and robotics can be combined to perform personalized automated osteotomy planning, and the autonomous joint replacement surgical robot navigation and positioning system can be used to automatically complete the osteotomy with high precision. Compared with traditional manual methods, this system can significantly improve the accuracy, efficiency, and safety of osteotomy. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] FIG1 is a hardware schematic diagram of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0060] FIG2 is an architecture diagram of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0061] FIG3 illustrates the processing process of a three-dimensional introduction unit of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0062] FIG4 is a schematic diagram of a Marching Cubes reconstruction process of an autonomous joint replacement surgical robot navigation and positioning system according to an embodiment of the present application;

[0063] FIG5 illustrates the processing process of the prosthesis planning unit of the autonomous joint replacement surgical robot navigation and positioning system according to an embodiment of the present application;

[0064] FIG6 illustrates the processing process of the skeletal registration unit of the autonomous joint replacement surgical robot navigation and positioning system according to an embodiment of the present application;

[0065] FIG7 illustrates a coarse registration process of a skeletal registration unit of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0066] Figure 8 is a schematic diagram of a simulation of a real surgical scene captured by an RGB camera;

[0067] FIG9 is a processing process of a path planning unit of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0068] FIG10 illustrates a specific process of generating multiple waypoints by a path planning unit of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0069] FIG11 is a schematic diagram of a process for generating multiple waypoints by a path planning unit of a navigation and positioning system for an autonomous joint replacement surgical robot according to an embodiment of the present application;

[0070] FIG12 is a diagram illustrating the processing process of the robotic arm control unit of the autonomous joint replacement surgical robot navigation and positioning system according to an embodiment of the present application. DETAILED DESCRIPTION

[0071] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, specific embodiments of the present application are described in detail below with reference to the accompanying drawings. Although the accompanying drawings show exemplary embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0072] It should be noted that, unless otherwise specified, the technical or scientific terms used in this application should have the common meanings understood by those skilled in the art to which this application belongs.

[0073] Joint replacement surgery is a common and complex surgical procedure used to treat joint diseases and injuries. Traditional joint replacement surgery typically requires surgeons to rely on extensive experience and visual judgment, but this approach has certain limitations and risks. In recent years, with the development of robotics and computer-assisted surgery, joint replacement surgical robots have gradually become an emerging technology, bringing significant changes and advancements to joint replacement surgery.

[0074] Currently, with the introduction of joint replacement surgical robots, they only participate in preoperative planning, and these operations during the operation are generally performed manually. This manual osteotomy method has the following problems:

[0075] The accuracy of osteotomy depends on the doctor's experience and there are individual differences, making it difficult to guarantee the osteotomy effect.

[0076] Manual operation causes great strain, is slow and inefficient; osteotomy is repetitive and laborious, and doctors easily become fatigued.

[0077] Manual operation carries the risk of human error such as slipping and distortion, which may damage cartilage or ligaments.

[0078] The above problems will lead to the surgical results being less than expected.

[0079] In response to the above problems, the present application provides a new autonomous joint replacement surgical robot navigation and positioning system solution, which can complete knee osteotomy through preoperative planning, intraoperative alignment, and establishment of an artificial potential field to control the robotic arm, thereby solving the problem that current manual surgery has uneven surgical levels due to individual differences and the surgical results are not as expected.

[0080] For ease of understanding, the following terms that may be used are explained here:

[0081] An embodiment of the present application provides a navigation and positioning system for an autonomous joint replacement surgical robot, the specific scheme of which is shown in Figures 1 to 12.

[0082] FIG1 and FIG2 are schematic diagrams illustrating an autonomous joint replacement surgical robot navigation and positioning system according to an embodiment of the present application. The autonomous joint replacement surgical robot navigation and positioning system includes:

[0083] A three-dimensional importing unit 101 is configured to import a three-dimensional bone image of a knee joint;

[0084] In the present application, the three-dimensional bone image of the knee joint is imported as three-dimensional stereoscopic image data, which is generated by a medical image of the knee joint. The medical image of the knee joint can be a CT image and / or an X-ray image.

[0085] The prosthesis planning unit 102 is configured to plan the knee joint prosthesis position, prosthesis model and osteotomy plane posture information based on the three-dimensional bone image;

[0086] In this application, the model parameters of the knee joint prosthesis can be obtained by measuring the actual knee joint prosthesis or collecting corresponding medical data, or can be obtained from the corresponding medical device company. The specific acquisition method will not be repeated here.

[0087] In this application, after planning the knee prosthesis position, prosthesis model and osteotomy plane posture information, the planned content can be combined with the three-dimensional bone image for display. Based on this display, medical staff can adjust the planning information to achieve a more accurate planning effect.

[0088] The skeleton registration unit 103 is configured to register the three-dimensional skeleton image with the skeleton to be osteotomized in the real surgical scene, and determine the osteotomy plane posture information of the skeleton to be osteotomized;

[0089] The bone registration in this application is to map the osteotomy plane posture information to the bone to be osteotomized in the real surgical scene, so as to perform osteotomy based on the mapped information.

[0090] The path planning unit 104 is configured to plan the osteotomy path by an artificial potential field method according to the osteotomy plane posture information of the bone to be osteotomized;

[0091] The robotic arm control unit 105 is configured to control the robotic arm to complete automatic osteotomy according to the planned osteotomy path.

[0092] As shown in Figure 1, it is a hardware schematic diagram of the autonomous joint replacement surgical robot navigation and positioning system; it shows a robotic arm mounted on a trolley, a visual sensor (infrared binocular camera, RGB binocular camera, etc.) and a display; in addition, the autonomous joint replacement surgical robot navigation and positioning system is also provided with control software for controlling and processing the above-mentioned hardware devices.

[0093] As shown in FIG2 , it is an architecture diagram of the autonomous joint replacement surgical robot navigation and positioning system; the figure shows the various units included in the autonomous joint replacement surgical robot navigation and positioning system.

[0094] It should be noted that the various units included in the autonomous joint replacement surgical robot navigation and positioning system can be considered as the control software of the above-mentioned hardware devices, or can be considered as each unit consisting of the control software of the above-mentioned hardware devices and the hardware controlled by the software.

[0095] In this application, preoperative planning of the prosthesis is performed through a three-dimensional bone image of the knee joint, and the osteotomy plane posture information is mapped to the bone to be osteotomized in the real surgical scene through bone registration, so as to achieve high-precision correspondence between the osteotomy planning and the osteotomy scene, and thus complete the osteotomy of the joint replacement on this basis.

[0096] It should be noted that after completing high-precision osteotomy, the corresponding medical staff can complete the planned prosthesis installation and surgical suturing, thereby completing the entire joint replacement surgery. Based on this, it can be seen that the autonomous joint replacement surgical robot navigation and positioning system in this application is mainly involved in the preoperative planning and intraoperative osteotomy of the entire replacement surgery. The prosthesis installation and surgical suturing after osteotomy are completed by medical staff (this is because the prosthesis needs to be proofread after osteotomy. This proofreading is generally based on the experience of medical staff and the workload is small. Therefore, it is changed to be completed by medical staff, which can not only improve the safety of the entire replacement surgery, but also increase the universality of the autonomous joint replacement surgical robot navigation and positioning system). In this application, the operation content after osteotomy (completed by medical staff) will not be repeated.

[0097] It should be noted that the autonomous joint replacement surgical robot navigation and positioning system is mainly used for preoperative planning and intraoperative osteotomy. The navigation is to guide the osteotomy movement of the osteotomy tool, and the positioning is to determine the position of each position or key point during the operation. On this basis, it can also be identified as an autonomous joint replacement surgical robot navigation and positioning osteotomy system, which does not affect the meaning of the entire application document.

[0098] In this application, based on high-precision bone alignment before and during surgery, the osteotomy plane posture information can be mapped to the bone to be osteotomized in the real surgical scene, and the robotic arm can be controlled to complete the automatic osteotomy operation, thereby realizing the automated operation of the autonomous joint replacement surgical robot navigation and positioning system during the operation, reducing human intervention, and thus reducing the errors caused by human intervention, thereby improving the accuracy, stability and efficiency of the operation.

[0099] In this application, 3D imaging and robotics can be combined to perform personalized automated osteotomy planning, and the autonomous joint replacement surgical robot navigation and positioning system can be used to automatically complete the osteotomy with high precision. Compared with traditional manual methods, this system can significantly improve the accuracy, efficiency, and safety of osteotomy.

[0100] FIG3 shows a processing process of a three-dimensional import unit of a navigation and positioning system for an autonomous joint replacement surgery robot. When importing a three-dimensional bone image of a knee joint, the three-dimensional import unit 101 is specifically configured as follows:

[0101] Acquire CT images of the knee joint of the joint replacement patient;

[0102] Perform image segmentation on the bone part and non-bone part in the knee joint CT image to obtain the segmentation result of the bone part;

[0103] Based on the connected domain analysis, the skeleton is segmented and the skeleton structures of the tibia, femur, patella and fibula are extracted.

[0104] A three-dimensional bone image of the knee joint is created based on the extracted bone structures of the tibia, femur, patella, and fibula. The three-dimensional bone image includes three-dimensional images of the tibia, femur, patella, and fibula.

[0105] In this application, importing three-dimensional bone image information is the basis of osteotomy. First, the patient's knee joint CT scan image is collected. The acquisition method can be to use medical image processing software (such as Mimics, 3D Slicer, etc.) to read the original DICOM format scan image. After this, a threshold value can be set and an image segmentation algorithm (such as based on region growing, cascade classifier, etc.) is used to separate bone from non-bone tissue. After the bone tissue is separated, the bone structure of tibia, femur, and fibula is accurately extracted from the segmentation result through algorithms such as connected domain analysis. The extracted bone is then three-dimensionally reconstructed and converted into a volume data model (three-dimensional bone image model).

[0106] It should be noted that when partial osteotomy is not affected by the fibula and patella, only the bone structures of the tibia and femur can be extracted during bone extraction, thereby reducing the bone information that needs to be extracted and increasing the speed of three-dimensional image generation.

[0107] In the present application, in creating a three-dimensional bone image of the knee joint based on the extracted bone structures of the tibia, femur, patella and fibula, different three-dimensional images are created based on different bone structures. Specifically, a three-dimensional bone image of the tibia of the knee joint is created based on the extracted bone structure of the tibia, a three-dimensional bone image of the femur of the knee joint is created based on the extracted bone structure of the femur, and / or a three-dimensional bone image of the patella of the knee joint is created based on the extracted bone structure of the patella, and / or a three-dimensional bone image of the fibula of the knee joint is created based on the extracted bone structure of the fibula.

[0108] In the present application, the creation of three-dimensional images of different bones is performed separately, but the relative positional relationship between the different bones is still retained (the relative positional relationship can be maintained during the three-dimensional image creation process).

[0109] In this application, a Marching Cubes algorithm is used to create a three-dimensional bone image of the knee joint based on the extracted bone structures of the tibia, femur, patella, and fibula.

[0110] FIG4 is a schematic diagram of the Marching Cubes reconstruction process. The specific process of using the Marching Cubes algorithm to create a three-dimensional skeletal image of the knee joint based on the extracted skeletal structures of the tibia, femur, patella, and fibula is as follows:

[0111] The three-dimensional space is divided into several small squares, and the bone structure data points in each square are interpolated; based on the set isosurface threshold, the intersection of the isosurface in the square and the square edge is determined; based on the combination of intersection points on the square edge and the preset topology table, the isosurface triangle fragments in the square are constructed; the local triangle fragments constructed in each square are spliced ​​together to finally obtain the overall three-dimensional isosurface of the bone structure, which is the three-dimensional bone image.

[0112] FIG5 shows the processing of the prosthesis planning unit 102 of the autonomous joint replacement surgical robot navigation and positioning system. When planning the knee joint prosthesis position, prosthesis model, and osteotomy plane posture information based on the three-dimensional bone image, the prosthesis planning unit 102 is specifically configured as follows:

[0113] Input the 3D skeleton image into the trained point recognition deep reinforcement learning model to obtain the corresponding 3D skeleton key points;

[0114] Calculating medical data of the knee joint based on the three-dimensional skeleton key points;

[0115] determining a prosthesis model of the knee joint based on the medical data of the knee joint;

[0116] Determining, according to the prosthesis signal of the knee joint, the prosthesis position of the knee joint and osteotomy plane posture information, wherein the osteotomy plane posture information is determined based on the overlapping position information of the three-dimensional image of the knee joint and the prosthesis of the knee joint;

[0117] The point recognition deep reinforcement learning model is trained based on three-dimensional skeleton image samples, and the three-dimensional skeleton image samples are marked with key point data.

[0118] In this application, the three-dimensional skeleton key points are identified through the point recognition deep reinforcement learning model. On the one hand, the point recognition deep reinforcement learning model can be trained in the background to reduce the real-time calculation amount; on the other hand, the trained model can be directly used for key point recognition during planning, thereby greatly improving the recognition efficiency.

[0119] It should be noted that the three-dimensional bone key points that need to be identified in this application are determined based on the actual needs when planning the prosthesis. For example, if only the intercondylar fossa key points are needed to complete the prosthesis planning, then the deep reinforcement learning model for identifying this point is the intercondylar fossa identification deep reinforcement learning model.

[0120] In this application, the training and use of the point recognition deep reinforcement learning model are carried out according to actual needs, thereby substantially reducing the amount of calculation and improving the corresponding speed.

[0121] In the present application, the prosthesis position and model are automatically planned by the prosthesis planning unit 102, thereby reducing abnormalities in key information of postoperative evaluation caused by differences in prosthesis installation experience.

[0122] In the present application, the medical data of the knee joint are medical data that can determine the position and model of the prosthesis, such as the lateral femoral extension gap: the vertical distance from the lateral distal end of the femur to the osteotomy surface; the lateral femoral flexion gap: the distance from the lateral posterior femoral condyle point to the posterior condyle osteotomy surface; the medial femoral extension gap: the vertical distance from the medial distal end of the femur to the osteotomy surface; the medial femoral flexion gap: the distance from the medial posterior femoral condyle point to the posterior condyle osteotomy surface; the medial tibial extension gap: the distance from the medial point of the tibial plateau to the osteotomy surface; the medial tibial flexion gap: the distance from the medial point of the tibial plateau to the osteotomy surface, etc. The specific medical data used can be determined according to actual needs.

[0123] In this application, the osteotomy plane posture information is determined based on the overlapping position information of the three-dimensional image of the knee joint and the prosthesis of the knee joint, that is, the overlapping position of the three-dimensional image of the knee joint and the prosthesis of the knee joint is used as the part that needs to be cut off, and the boundary of the overlapping position is the osteotomy plane posture information.

[0124] FIG6 shows a processing process of the skeletal registration unit 103 of the autonomous joint replacement surgical robot navigation and positioning system. The skeletal registration unit 103 is specifically configured to register the three-dimensional skeletal image with the bone to be osteotomized in the actual surgical scene and determine the osteotomy plane pose information of the bone to be osteotomized.

[0125] Obtain a two-dimensional image from an RGB binocular camera, the two-dimensional image containing a visual marker, and obtain visual marker positioning data from an infrared binocular camera;

[0126] Establishing an initial posture relationship between the probe and the three-dimensional skeleton model according to at least one three-dimensional skeleton key point corresponding to the three-dimensional skeleton image, wherein the relative positions of the probe and the three-dimensional skeleton model are fixed under the initial posture relationship;

[0127] Based on the visual marker, the visual marker positioning data, the two-dimensional image of the RGB binocular camera, and the three-dimensional bone model, digital twin registration is performed on the three-dimensional bone model and the bone to be osteotomized in the real surgical scene;

[0128] Using the probe to touch the surface of the bone to be osteotomized, and determining the touch position coordinates in the infrared binocular camera coordinate system based on the shape data of the probe and the visual marker positioning data of the probe in the infrared binocular camera coordinate system at the time of touch;

[0129] Registering the touch position coordinates with the three-dimensional skeleton model through an iterative closest point algorithm to determine a mapping relationship between the three-dimensional skeleton model and the touch position coordinates;

[0130] Based on the mapping relationship between the three-dimensional bone model and the touch position coordinates, the planned osteotomy plane posture information is mapped to the local coordinate system of the marking and positioning frame of the bone to be osteotomized.

[0131] In this application, the RGB binocular camera and the infrared binocular camera are both oriented toward a real surgical scene, and can acquire information about the bone to be osteotomized in that scene. It should be noted that visual markers are installed on the bone to be osteotomized, the end of the surgical robot's robotic arm, and the probe in the real surgical scene, and based on these visual markers, the positional information of the bone to be osteotomized, the end of the surgical robot's robotic arm, and the probe can be determined.

[0132] It should be noted that in a real surgical scenario, the bone to be osteotomized, the end of the robotic arm of the surgical robot, and the visual marker installed on the probe have a determined and known relative position relationship with the visual marker (the specific determination method will not be repeated in this application). Based on this relative position relationship, the position of the bone to be osteotomized, the end of the robotic arm of the surgical robot, and the probe can be directly determined; and based on this relative position relationship, after the visual marker is mapped to the coordinate system, the position information of the bone to be osteotomized, the end of the robotic arm of the surgical robot, and the probe can also be synchronously mapped to the new coordinate system.

[0133] It should be noted that although the infrared binocular camera is facing the real surgical scene, it only obtains the position data of each visual marker in the real surgical scene, and does not process the remaining data in the real surgical scene (regardless of whether it is obtained or not). In this way, the amount of data that needs to be processed by the infrared binocular camera can be greatly reduced.

[0134] In this application, the infrared binocular camera is an NDI camera.

[0135] In the present application, the shape data of the probe is predetermined. Based on the shape data of the probe, the needle tip position coordinates of the probe, ie, the coordinates of the touch position, can be determined when the visual marker positioning data of the probe is known.

[0136] In this application, coarse registration is completed by an RGB camera, and fine registration is completed by an NDI camera, so that when the amount of calculation is large in the coarse registration stage, the accuracy is reduced and the transmission rate is increased; in the fine registration stage, the actual surgical scene data obtained is reduced to reduce the amount of calculation and increase the calculation speed / reaction speed.

[0137] In this application, the purpose of skeleton registration is to establish a rigid transformation relationship between the image coordinate system and the real space.

[0138] During the fine registration process, the data from the probe probing the bone surface is used to perform iterative point cloud calculations based on the traditional iterative closest point (ICP) algorithm until the error converges to a preset threshold.

[0139] FIG7 shows a coarse registration process of the skeletal registration unit 103 of the autonomous joint replacement surgical robot navigation and positioning system. The skeletal registration unit 103 is specifically configured to perform digital twin registration of the three-dimensional skeletal model and the bone to be osteotomized in the real surgical scene, as shown in FIG7 .

[0140] Identifying coordinate information of a visual marker of the probe in the two-dimensional image, and determining a mapping relationship between a local coordinate system of the probe and a coordinate system of the RGB camera;

[0141] According to the mapping relationship between the local coordinate system of the probe and the RGB camera coordinate system, the three-dimensional bone model is projected into the two-dimensional image in the RGB camera coordinate system, where the two-dimensional image is a real surgical scene image in the RGB camera coordinate system;

[0142] Adjusting the posture of the probe to change the posture of the three-dimensional bone model until the projection of the three-dimensional bone model on the two-dimensional image coincides with the bone to be osteotomized on the two-dimensional image;

[0143] When the projection of the three-dimensional bone model coincides with the bone to be osteotomized, obtaining positioning data of the visual marker of the probe in the coordinate system of the infrared binocular camera;

[0144] Determine a mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized based on the positioning data of the visual marker of the probe in the coordinate system of the infrared binocular camera and the positioning data of the visual marker of the marking and positioning frame of the bone to be osteotomized in the coordinate system of the infrared binocular camera;

[0145] Based on the mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized, the coordinate system of the three-dimensional bone model is converted to the local coordinate system of the marking and positioning frame of the bone to be osteotomized to complete the digital twin alignment.

[0146] As shown in Figure 8, it is a simulated schematic diagram of a real surgical scene obtained by an RGB camera, which includes a simulation of the position of the bone to be osteotomized in the real surgical scene, a simulation of the visual marker of the probe, and a simulation of the planned three-dimensional bone model; wherein, the bone drawn with a solid line is a schematic diagram of the bone to be osteotomized in the real surgical scene, and its position is the position of the bone to be osteotomized in the current two-dimensional image (the bone to be osteotomized in the real surgical scene will not move after being fixed), and the bone drawn with a dotted line is a schematic diagram of the three-dimensional bone model, and its relative position with the probe is fixed. The four concentric circles on the probe are used to represent the visual markers. By manually moving the probe, the visual markers of the probe in the schematic diagram and the planned three-dimensional bone model will also move synchronously; when the three-dimensional bone model in the schematic diagram coincides with the position of the bone to be osteotomized, the rough alignment is completed.

[0147] It should be noted that the diagram shows the projection of the three-dimensional skeleton model on the two-dimensional image, but the projection will change as the posture of the three-dimensional skeleton model changes.

[0148] In the present application, the probe itself has a local coordinate system, in which the visual markers of the probe have their own coordinates. At the same time, since the probe and the three-dimensional skeleton model have a fixed relative relationship, it can be considered that the coordinates of the three-dimensional skeleton model at this time are also the coordinates of the probe in the local coordinate system (that is, whether it is, or not, it can be unified when establishing the initial posture relationship between the probe and the three-dimensional skeleton model. The unification can be to unify the coordinates of the probe's visual markers to the coordinate system of the three-dimensional skeleton model, or to unify the three-dimensional skeleton model to the coordinate system of the probe's visual markers, but after unification, they are all called the probe's local coordinate system); when identifying the coordinate information of the probe's visual marker in the two-dimensional image, the two-dimensional image is the coordinate in the RGB camera coordinate system, then the coordinates of the probe's visual marker therein are also the coordinates in the RGB camera coordinate system; at this time, a mapping relationship is established based on the coordinates of the two coordinate systems of the probe's visual marker; based on the mapping relationship, the coordinates of the three-dimensional skeleton model can be converted into a coordinate system, and the projection is calculated based on the plane where the two-dimensional image in the coordinate system is located, and the projection is displayed. At this time, if the probe is moved manually, the projection will move synchronously. Medical staff can move the probe based on the displayed projection and two-dimensional image until the projection coincides with the corresponding bone in the two-dimensional image. The state at this time is the coarse alignment state.

[0149] In this application, the digital twin registration is a coarse registration process. After the probe is fixedly associated with the three-dimensional bone model, the probe is moved manually, and the three-dimensional bone model will be moved synchronously until the three-dimensional bone model coincides with the position of the bone to be osteotomized in the real surgical scene, and the coarse registration is completed; this state is recorded, and based on the mapping relationship between the local coordinate system of the probe in the infrared binocular camera coordinate system and the local coordinate system of the marking and positioning frame of the bone to be osteotomized, the specific data of the three-dimensional bone model can be mapped to the bone to be osteotomized.

[0150] In this application, by displaying the real surgical scene and projecting the three-dimensional bone model, rough registration is completed with human participation. On the one hand, rough registration can be completed quickly without a lot of calculations; on the other hand, the accuracy of rough registration can be greatly increased, the speed of subsequent fine registration can be improved, and the amount of calculation for fine registration can be reduced.

[0151] In one embodiment, in the bone registration unit 103, when the bone to be osteotomized includes a tibia to be osteotomized and a femur to be osteotomized, the three-dimensional image of the tibia in the three-dimensional bone model and the tibia to be osteotomized in the real surgical scene, and the three-dimensional image of the femur in the three-dimensional bone model and the femur to be osteotomized in the real surgical scene are respectively registered through the digital twin registration process.

[0152] It should be noted that the relative positions of the various bone structures in the human skeleton 3D model are the preoperative shooting positions, which are different from the surgical positions during the operation. Therefore, during the registration, the tibia and femur need to be registered separately.

[0153] FIG9 shows a processing process of the path planning unit 104 of the autonomous joint replacement surgical robot navigation and positioning system. The path planning unit 104 is specifically configured to plan the osteotomy path using the artificial potential field method based on the osteotomy plane posture information of the bone to be osteotomized:

[0154] Generate osteotomy plane boundaries and soft tissue protection areas based on the osteotomy plane posture information of the bone to be osteotomized and the three-dimensional bone model;

[0155] Determine the calculation formula for the osteotomy path, set the distal end of the path surface as the target position, and set the soft tissue protection area as the gravitational field;

[0156] The osteotomy plane boundary is set as a strong constraint safety strategy, and a plurality of waypoints are generated based on the strong constraint safety strategy. The combination of the waypoints is the planned osteotomy path.

[0157] In this application, the osteotomy plane boundary is set as a strong constraint safety strategy, thereby separating the osteotomy area from the soft tissue protection area, and executing path planning under only the gravitational field (the osteotomy plane boundary can also be regarded as a special repulsive field).

[0158] In this application, based on the osteotomy plane posture information of the bone to be osteotomized and the three-dimensional bone model (or a CT image of the corresponding position, or a non-bone segmentation result when the bone part and the non-bone part are segmented), a soft tissue protection area is obtained, and the soft tissue protection area is located outside the osteotomy plane. The specific determination method is not limited in this application.

[0159] In this application, the artificial potential field method (APF) is abbreviated as APF. The basic idea is to construct an obstacle repulsive potential field (Obstacle Potential Field) around the obstacle and a gravitational potential field (Goal Potential Field) around the target point. The controlled object is simultaneously subjected to repulsive and gravitational forces in the composite field composed of these two potential fields. The direction of the combined force of repulsion and gravitation (potential field gradient) guides the movement of the controlled object, generating a collision-free path.

[0160] In one embodiment, the calculation formula of the osteotomy path is: F a =k a (x g -x)

[0161] Among them, F a is the gravitational potential field, F r is the repulsive potential field, m is the potential field constant, k a is the proportional coefficient, x g is the target position, x is the current position, k r is a positive coefficient, x0 is the obstacle position, ρ(x) is the distance from the current position to the obstacle, and ρ0 is the impact distance.

[0162] It should be noted that, in the present application, when no repulsive field is set, the repulsive potential field is reset to zero.

[0163] In one embodiment, as shown in FIG. 10 , which illustrates a specific process for generating multiple waypoints, the path planning unit 104 is specifically configured to:

[0164] Generate a waypoint based on the gravitational field and repulsive field;

[0165] Determine whether the generated waypoint is within the boundary of the osteotomy plane;

[0166] When the waypoint is outside the boundary of the osteotomy plane, the waypoint is translated into the boundary of the osteotomy plane along the direction toward the centroid of the gravitational field;

[0167] When the waypoint is within the boundary of the osteotomy plane, the next waypoint is generated according to the attraction field and the repulsion field;

[0168] Determine whether the currently generated waypoint reaches the distal end of the osteotomy plane boundary;

[0169] In the case where the waypoint does not reach the distal end of the osteotomy plane boundary, returning to the process of determining whether the generated waypoint is located within the osteotomy plane boundary;

[0170] When the waypoint reaches the distal end of the osteotomy plane boundary, updating the gravitational field, the repulsive field, and the osteotomy plane boundary;

[0171] When the remaining osteotomy area is greater than the preset threshold, the process returns to the process of generating a waypoint according to the gravitational field and the repulsive field.

[0172] As shown in Figure 11, it is a schematic diagram of the process of generating multiple waypoints. When observed horizontally from left to right, it can be seen that the osteotomy plane boundary is generated in the first figure, the gravitational field of the soft tissue protection area and the target position is generated in the second figure, and five waypoints of a path are generated, and the path reaches the distal end of the osteotomy plane boundary; in the fourth figure, based on the path of the five waypoints that have been generated, the path is used as the new osteotomy plane boundary, and the gravitational field, the repulsive field and the osteotomy plane boundary are updated; the fifth figure shows the waypoints of all the generated osteotomy paths. At this time, the remaining osteotomy area is less than the preset threshold, and the osteotomy planning is completed.

[0173] In this application, the soft tissue area is the obstacle gravitational field, and the centroid of the gravitational field is the center of the gravitational field.

[0174] In this application, the strong constraint of the gravitational field attracting waypoints in the soft tissue area is the boundary of the osteotomy plane. The purpose of setting the soft tissue protection area as the gravitational field is to make the path close to the soft tissue area, but after using the plane boundary to constrain it, it cannot cross the boundary, forming a safety strategy.

[0175] In this application, when a waypoint in the path appears within the non-osteotomy plane boundary, the waypoint is placed within the non-osteotomy plane boundary. When the waypoint approaches the target position, the path generation is completed and the gravitational field boundary is updated using the generated path until the residual area is less than the threshold.

[0176] FIG12 shows a processing process of the robot control unit 105. The robot control unit 105 is specifically configured to:

[0177] Obtaining a calibrated osteotomy tool coordinate system;

[0178] Mapping the planned osteotomy path into an osteotomy path in a robot arm flange coordinate system based on the osteotomy tool coordinate system;

[0179] Based on the kinematic constraint formula, the osteotomy path in the robot arm flange coordinate system is converted into the joint angle trajectory of each joint;

[0180] Interpolate the joint angle trajectory of each joint;

[0181] The feedforward PID controller outputs the joint angular velocity of each joint angle to control the robot arm;

[0182] Acquire the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized in real time;

[0183] The target joint angle trajectory is updated based on the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized, and the interpolation process of the joint angle trajectory of each joint is returned.

[0184] In this application, the kinematic constraint formula is a conventional kinematic formula and will not be described in detail here.

[0185] In this application, a PID controller (Proportion Integration Differentiation) is composed of a proportional unit (P), an integral unit (I) and a differential unit (D).

[0186] In this application, base coordinate system: the coordinate system with the base of the robot as the origin

[0187] In this application, the robot arm flange coordinate system is the end joint coordinate system: generally, the end joint of the robot arm is covered with a layer of flange, so the robot arm end joint coordinate system is also the flange end coordinate system.

[0188] During specific execution, the position and posture of the end joint coordinate system in the base coordinate system is the position and posture of the robotic arm.

[0189] Tool End Coordinate System: During the operation of the robotic arm, the end of the robotic arm will carry a tool. During actual execution, the tool moves to a certain position, rather than the end of the robotic arm moving to a certain position. In this case, we need to establish a coordinate system for the tool, which is also known as the Tool End Coordinate System. In this application, the osteotomy tool coordinate system is the Tool End Coordinate System.

[0190] In the present application, when interpolating the joint angle trajectory of each joint, the interpolation number of the joint angle of each joint is different, and the specific interpolation data is determined based on the rotation of the joint angle.

[0191] In the present application, the osteotomy path in the robot arm flange coordinate system includes multiple waypoints, and each waypoint has a corresponding trajectory point in the joint angle trajectory of each joint.

[0192] In this application, the joint angle trajectory of each joint is interpolated, that is, interpolation is performed between adjacent trajectory points.

[0193] In this application, the real-time positions of the osteotomy tool and the bone to be osteotomized are determined by acquiring the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized in real time, and the PID controller is adjusted based on the real-time tracked position to achieve timely feedback and accurate execution.

[0194] In one embodiment, the PID controller is expressed as:

[0195] in, is the joint angular velocity output by the proportional-integral-derivative controller, q r is the target joint angle trajectory, K p , K i , K d For controller parameters.

[0196] In this application, closed-loop control is completed by forming a dynamically adjusted planning trajectory and an error model of the current robotic arm joint angle based on the alignment results and image planning results combined with the robotic arm tool positioning frame acquired by the infrared binocular.

[0197] It should be noted that, in the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this description.

[0198] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, system, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, system, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a..." does not preclude the presence of other identical elements in the process, system, commodity, or apparatus that includes the element.

[0199] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. An autonomous joint replacement surgical robot navigation and positioning system, comprising: A three-dimensional import unit configured to import a three-dimensional bone image of a knee joint; A prosthesis planning unit is configured to plan the position of the knee joint prosthesis, the prosthesis model and the osteotomy plane posture information based on the three-dimensional bone image; A bone registration unit is configured to register the three-dimensional bone image with the bone to be osteotomized in the real surgical scene, and determine the osteotomy plane posture information of the bone to be osteotomized; A path planning unit is configured to plan an osteotomy path by an artificial potential field method according to the osteotomy plane posture information of the bone to be osteotomized; The robot arm control unit is configured to control the robot arm to complete automatic osteotomy according to the planned osteotomy path.

2. According to the autonomous joint replacement surgical robot navigation and positioning system of claim 1, the bone registration unit, when registering the three-dimensional bone image with the bone to be osteotomized in the real surgical scene to determine the osteotomy plane posture information of the bone to be osteotomized, is specifically configured as follows: Obtain a two-dimensional image of an RGB binocular camera, the two-dimensional image containing a visual marker, and obtain visual marker positioning data of an infrared binocular camera; According to at least one three-dimensional skeleton key point corresponding to the three-dimensional skeleton image, an initial posture relationship between the probe and the three-dimensional skeleton model is established, under which the relative position of the probe and the three-dimensional skeleton model is fixed; Based on the visual marker, the visual marker positioning data, the two-dimensional image of the RGB binocular camera, and the three-dimensional bone model, digital twin registration is performed on the three-dimensional bone model and the bone to be osteotomized in the real surgical scene; Using the probe to touch the surface of the bone to be osteotomized, based on the shape data of the probe and the visual marker positioning data of the probe in the coordinate system of the infrared binocular camera when touching, determine the coordinates of the touching position in the coordinate system of the infrared binocular camera; By iterating the closest point algorithm, the touch position coordinates are registered with the three-dimensional skeleton model to determine the mapping relationship between the three-dimensional skeleton model and the touch position coordinates; Based on the mapping relationship between the three-dimensional bone model and the touch position coordinates, the planned osteotomy plane posture information is mapped to the local coordinate system of the marking and positioning frame of the bone to be osteotomized.

3. According to the autonomous joint replacement surgical robot navigation and positioning system of claim 2, the bone registration unit, when performing digital twin registration of the three-dimensional bone model and the bone to be osteotomized in the real surgical scene, is specifically configured as follows: Identify coordinate information of a visual marker of the probe in the two-dimensional image, and determine a mapping relationship between a local coordinate system of the probe and a coordinate system of the RGB camera; According to the mapping relationship between the local coordinate system of the probe and the RGB camera coordinate system, the three-dimensional bone model is projected into the two-dimensional image in the RGB camera coordinate system, and the two-dimensional image is the real surgical field in the RGB camera coordinate system. Scenery image; Adjusting the position and posture of the probe to change the position and posture of the three-dimensional bone model until the projection of the three-dimensional bone model on the two-dimensional image coincides with the bone to be osteotomized on the two-dimensional image; When the projection of the three-dimensional bone model coincides with the bone to be osteotomized, obtaining positioning data of the visual marker of the probe in the coordinate system of the infrared binocular camera; Based on the positioning data of the visual marker of the probe in the infrared binocular camera coordinate system and the positioning data of the visual marker of the marking and positioning frame of the bone to be osteotomized in the infrared binocular camera coordinate system, determine the mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized; Based on the mapping relationship between the local coordinate system of the probe and the local coordinate system of the marking and positioning frame of the bone to be osteotomized, the coordinate system of the three-dimensional bone model is converted to the local coordinate system of the marking and positioning frame of the bone to be osteotomized to complete the digital twin alignment.

4. The autonomous joint replacement surgical robot navigation and positioning system according to claim 2, wherein: In the bone registration unit, when the bones to be osteotomized include a tibia to be osteotomized and a femur to be osteotomized, the three-dimensional image of the tibia in the three-dimensional bone model and the tibia to be osteotomized in the real surgical scene, and the three-dimensional image of the femur in the three-dimensional bone model and the femur to be osteotomized in the real surgical scene are respectively registered through the digital twin registration process.

5. The autonomous joint replacement surgical robot navigation and positioning system according to any one of claims 1 to 4, wherein the path planning unit, when planning the osteotomy path by an artificial potential field method according to the osteotomy plane posture information of the bone to be osteotomized, is specifically configured as follows: Generate osteotomy plane boundaries and soft tissue protection areas based on osteotomy plane posture information of the bone to be osteotomized and a three-dimensional bone model; Determine the calculation formula of the osteotomy path, set the distal end of the path surface as the target position, and set the soft tissue protection area as the gravitational field; The osteotomy plane boundary is set as a strong constraint safety strategy, and a plurality of waypoints are generated based on the strong constraint safety strategy, and a combination of the waypoints is a planned osteotomy path.

6. The autonomous joint replacement surgical robot navigation and positioning system according to claim 5, wherein: The calculation formula of the osteotomy path is: F a =k a (x g -x) Among them, F a is the gravitational potential field, F r is the repulsive potential field, m is the potential field constant, k a is the positive proportionality coefficient, x g is the target position, x is the current position, k r is a positive coefficient, x0 is the obstacle position, ρ(x) is the distance from the current position to the obstacle, and ρ0 is Impact distance.

7. The autonomous joint replacement surgical robot navigation and positioning system according to claim 5, wherein: The path planning unit, when generating a plurality of waypoints based on the strong constraint safety policy, is specifically configured as follows: Generate a waypoint based on the gravitational field and repulsive field; Determine whether the generated waypoint is within the boundary of the osteotomy plane; When the waypoint is outside the boundary of the osteotomy plane, the waypoint is translated into the boundary of the osteotomy plane along the direction toward the centroid of the gravitational field; When the waypoint is within the boundary of the osteotomy plane, the next waypoint is generated according to the attraction field and the repulsion field; Determine whether the currently generated waypoint reaches the distal end of the osteotomy plane boundary; In the case where the waypoint does not reach the far end of the osteotomy plane boundary, returning to the process of determining whether the generated waypoint is located within the osteotomy plane boundary; When the waypoint reaches the distal end of the osteotomy plane boundary, updating the gravitational field, the repulsive field and the osteotomy plane boundary; When the remaining osteotomy area is greater than the preset threshold, the process returns to the process of generating a waypoint according to the gravitational field and the repulsive field.

8. The autonomous joint replacement surgical robot navigation and positioning system according to any one of claims 1 to 4, wherein: The robot arm control unit is specifically configured to: Obtaining a calibrated osteotomy tool coordinate system; Mapping the planned osteotomy path to an osteotomy path in a robot arm flange coordinate system based on an osteotomy tool coordinate system; Based on the kinematic constraint formula, the osteotomy path in the robot arm flange coordinate system is converted into the joint angle trajectory of each joint; Interpolate the joint angle trajectory of each joint; The feedforward PID controller outputs the joint angular velocity of each joint angle to control the robot arm; Acquire the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized in real time; The target joint angle trajectory is updated based on the positioning data of the osteotomy tool positioning frame and the positioning data of the marking positioning frame of the bone to be osteotomized, and the interpolation process of the joint angle trajectory of each joint is returned.

9. The autonomous joint replacement surgical robot navigation and positioning system according to claim 8, wherein: The PID controller is expressed as: in, is the joint angular velocity output by the proportional-integral-derivative controller, q r is the target joint angle trajectory, K p , K i , K d For controller parameters.

10. The autonomous joint replacement surgical robot navigation and positioning system according to any one of claims 1 to 4, wherein: The three-dimensional import unit, when importing the three-dimensional bone image of the knee joint, is specifically configured as follows: Acquire CT images of knee joints of joint replacement subjects; Perform image segmentation on the bone part and the non-bone part in the knee joint CT image to obtain the segmentation result of the bone part; Based on the connected domain analysis, the bone segmentation results are extracted to extract the bone structures of the tibia, femur, patella and fibula. A three-dimensional bone image of the knee joint is created based on the extracted bone structures of the tibia, femur, patella and fibula, and the three-dimensional bone image includes three-dimensional images of the tibia, femur, patella and fibula.

Images