Surgical robot control method, device, equipment, medium and product

Abstract

The embodiment of the invention discloses a surgical robot control method, device, equipment, system, medium and product, and the method comprises the steps: obtaining the sequence data of the joint angle, the joint angular acceleration, the tail end force and the bone mineral density of an osteotomy area of a mechanical arm of a surgical robot in the operation process; inputting the sequence data into a pre-trained control parameter analysis model to obtain a stiffness parameter and a terminal current integral prediction value; constraint conditions of the control parameter analysis model in the training process comprise a dynamic constraint relation between the tail end force and the joint torque, a preset rigidity model and a preset current integral threshold value; and control parameters of the mechanical arm are updated according to the rigidity parameters and the terminal current integral predicted value, and the operation process of the mechanical arm is controlled based on the updated control parameters. According to the technical scheme, the power parameters of the mechanical arm can be determined under the constraint of the preset current integral threshold value, the probability of an overcurrent event is reduced, and the operation stability of the surgical robot in an operation is optimized.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular to a surgical robot control method, device, equipment, system, medium, and product. Background Art

[0002] Total knee arthroplasty (TKA) is an effective means of treating serious knee joint diseases. The application of surgical robots in TKA and the precise control of the robotic arms of surgical robots can significantly improve the quality of TKA surgery.

[0003] However, when the robotic arm of the surgical robot is operating during surgery, an overcurrent alarm event may occur due to the maximum continuous current exceeding the accumulation time, which will interrupt the operation of the robotic arm and cause the surgery to be interrupted, affecting the progress of the surgery. Summary of the Invention

[0004] Embodiments of the present invention provide a surgical robot control method, device, equipment, medium and product, which can determine the power parameters of the robotic arm under the constraint of a preset current integration threshold, reduce the probability of overcurrent events, and optimize the operating stability of the surgical robot during surgery.

[0005] In a first aspect, an embodiment of the present invention provides a surgical robot control method, the method comprising:

[0006] Obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window;

[0007] The sequence data is input into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values. The constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold.

[0008] The control parameters of the manipulator are updated according to the stiffness parameters and the terminal current integral prediction value, and the operation process of the manipulator is controlled based on the updated control parameters.

[0009] In a second aspect, an embodiment of the present invention provides a surgical robot control device, the device comprising:

[0010] A data acquisition module is used to obtain sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window;

[0011] A data analysis module is used to input the sequence data into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values. The constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold.

[0012] The robot arm control module is used to update the control parameters of the robot arm according to the stiffness parameters and the terminal current integral prediction value, and control the operation process of the robot arm based on the updated control parameters.

[0013] In a third aspect, an embodiment of the present invention further provides a computer device, comprising:

[0014] one or more processors;

[0015] a memory for storing one or more programs;

[0016] When the above one or more programs are executed by one or more processors, the above one or more processors implement the surgical robot control method provided by any embodiment of the present invention.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a surgical robot control method as provided in any embodiment of the present invention.

[0018] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the surgical robot control method provided by any embodiment of the present invention.

[0019] The embodiments of the above invention have the following advantages or beneficial effects:

[0020] According to an embodiment of the present invention, the sequence data of the joint angle, joint angular acceleration, terminal force and bone density of the osteotomy area of the surgical robot's manipulator arm during operation is obtained in a preset time window; the sequence data is input into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values; wherein, the constraints of the control parameter analysis model during the training process include the dynamic constraint relationship between the terminal force and the joint torque, a preset stiffness model and a preset current integral threshold; the control parameters of the manipulator arm are updated according to the stiffness parameters and the terminal current integral prediction value, and the operation process of the manipulator arm is controlled based on the updated control parameters. The technical solution of the embodiment of the present invention solves the problem of overcurrent affecting the progress of the operation during the operation of the surgical robot. The power parameters of the manipulator arm can be determined under the constraint of the preset current integral threshold, thereby reducing the probability of overcurrent events and optimizing the operation stability of the surgical robot during the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart of a surgical robot control method provided by an embodiment of the present invention;

[0022] Figure 2 is a flowchart of another surgical robot control method provided by an embodiment of the present invention;

[0023] Figure 3 is a flowchart of another surgical robot control method provided by an embodiment of the present invention;

[0024] Figure 4 1 is a schematic structural diagram of a surgical robot control device provided by an embodiment of the present invention;

[0025] Figure 5 It is a structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0027] During TKA surgery, the maximum continuous current of the surgical robot may exceed the accumulation time, triggering an arm overcurrent alarm. This can interrupt the procedure for nearly two minutes, impacting the progress and negatively impacting the surgeon. This overcurrent phenomenon is primarily caused by the high cutting speed of the oscillating saw, which places excessive reaction force on the robot's arm. This overcurrent event can be avoided by reducing the oscillating saw speed or lowering the current limit.

[0028] However, simply reducing the saw's speed would directly prolong the surgery and affect the surgeon's rhythm. Lowering the current limit directly at the algorithm level would reduce the robotic arm's control stiffness, making it impossible to maintain precision, leading to surgeon dissatisfaction and even surgical failure.

[0029] This creates an impossible triangle: ensuring the robotic arm doesn't interrupt surgery due to overcurrent, maintaining high cutting efficiency, and ensuring high rigidity and precision. Further complicating matters is the varying tolerances each doctor has for each of these three conditions. Some doctors can provide excellent robotic arm assistance and also desire high precision and cutting efficiency, the ability to properly arrange heat dissipation gaps, and even the ability to tolerate occasional overcurrent. Others prefer a more measured approach, sacrificing efficiency for high precision. Still others prefer a swift completion of the surgery, tolerating even a slight decrease in precision. This becomes a complex physics problem involving human factors engineering, with parts that are easy to mathematically model and parts that are difficult to model digitally. Traditional optimization methods are subject to significant limitations. Through exploration, it has been discovered that physics-informed neural networks (PINNs) can better incorporate physics knowledge that facilitates mathematical modeling while also possessing the fuzzy reasoning capabilities of artificial intelligence (AI). The following examples describe the detailed process for solving the aforementioned technical problems.

[0030] Figure 1 This is a flowchart of a surgical robot control method provided in an embodiment of the present invention. This embodiment is applicable to scenarios where a surgical robot performs surgical tasks during TKA surgery. This method can be executed by a surgical robot control device, which can be implemented using software and / or hardware and integrated into a computer device with application development capabilities.

[0031] like Figure 1 As shown, the surgical robot control method of this embodiment includes the following steps:

[0032] S110 , obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window.

[0033] Among them, the sequence data of joint angle, joint angular acceleration, end force and bone density in the osteotomy area are the operating status data of the surgical robot during TKA surgery. By collecting the above status data multiple times in the preset time window, the dynamic changes of the operating status of the surgical robot in the preset time window can be obtained.

[0034] The robotic arm of a surgical robot may include one or more joints depending on its structural configuration. The end of the robotic arm is a surgical instrument, such as an oscillating saw. The joint angle can be determined by the sensing value of a sensor such as an encoder or potentiometer installed at the corresponding joint. The joint angular acceleration can be measured by an accelerometer installed at the joint, or it can be calculated based on parameters such as the joint angle. The end force can be determined by the sensing data of a force sensor installed at the end, or it can be determined based on the dynamic model of the surgical robot. According to the dynamic model of the surgical robot, combined with information such as the joint angle and joint angular acceleration, the force applied to the end can be calculated using a dynamic equation. The bone density of the osteotomy area can be determined based on pre-collected knee bone imaging data. The bone density of bones in different locations may be different. The bone density data of the corresponding osteotomy area can be determined in the corresponding imaging information based on the coordinate information of the osteotomy area. In some scenarios, sensor technology can also be used to monitor changes in bone density in real time during surgery.

[0035] S120 , inputting the sequence data into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values.

[0036] Among them, the pre-trained control parameter analysis model performs data analysis based on the input data, and can output the corresponding stiffness parameters and terminal current integral prediction values to provide a reference for the power control of the surgical robot's robotic arm.

[0037] The constraints imposed during training on the control parameter analysis model include the dynamic relationship between end-point force and joint torque, a preset stiffness model, and a preset current integration threshold. The control parameter analysis model is a physical information neural network whose outputs satisfy these physical constraints. Controlling the robotic arm based on these outputs reduces the likelihood of end-point overcurrent failures. Leveraging physical prior knowledge and using data-driven methods can also improve control precision, resulting in greater accuracy and robustness during surgical operation.

[0038] The dynamic constraint relationship can be a constraint relationship between the joint torque and end force of the robotic arm established based on the Newton-Euler dynamic equations. The preset stiffness model can be expressed as K = α*ρ*F, where ρ is the bone density of the osteotomy area, α is the proportionality coefficient, and K is the joint stiffness. It is understood that the adjustment or constraint of the stiffness is strongly correlated with the bone density of the osteotomy area, so that the final control parameters are more closely aligned with the osteotomy area. The preset current integration threshold must ensure that the oscillating saw operates within a safe current range to avoid damage to the equipment or safety hazards during surgery due to excessive current. The upper limit of the threshold should be lower than the maximum safe current integration value that the oscillating saw can withstand. An initial preset current integration threshold should be determined by referring to the technical specifications of the oscillating saw, past experimental data, or experience with similar equipment. During model training and testing, the value of the preset current integration threshold can be gradually increased or decreased, and the model performance can be observed to maintain a relatively optimal level of performance. For example, the threshold can be increased or decreased by a certain percentage (e.g., 5%-10%) each time, and then the model can be retrained and evaluated.

[0039] S130 , updating the control parameters of the robotic arm according to the stiffness parameter and the terminal current integral prediction value, and controlling the operation process of the robotic arm based on the updated control parameters.

[0040] Among them, updating the control parameters of the robotic arm according to the stiffness parameters and the terminal current integral prediction value can be determined according to the mapping relationship between the predetermined stiffness parameters and / or the terminal current integral prediction value and the robotic arm control parameters (such as joint angle, speed, acceleration, etc.). Specifically, the mapping relationship can be determined by a theoretical model, experimental data fitting or a machine learning algorithm. Based on the above mapping relationship, the stiffness parameters and the terminal current integral prediction value output by the model in S120 are used to update the control parameters of the robotic arm. Furthermore, the updated control parameters can be sent to the controller of the robotic arm to drive the robotic arm to operate according to the new parameters.

[0041] In addition, if the predicted value of the terminal current integral exceeds the preset current integral threshold, it can trigger a protective pause of the surgical robot.

[0042] The technical solution of this embodiment is to obtain the sequence data of the joint angle, joint angular acceleration, terminal force and bone density of the osteotomy area of the surgical robot's manipulator arm during operation in a preset time window; input the sequence data into a pre-trained control parameter analysis model to obtain the stiffness parameters and the terminal current integral prediction value; wherein, the constraints of the control parameter analysis model during the training process include the dynamic constraint relationship between the terminal force and the joint torque, the preset stiffness model and the preset current integral threshold; update the control parameters of the manipulator arm according to the stiffness parameters and the terminal current integral prediction value, and control the operation process of the manipulator arm based on the updated control parameters. The technical solution of the embodiment of the present invention solves the problem of overcurrent affecting the progress of the operation during the operation of the surgical robot. The power parameters of the manipulator arm can be determined under the constraint of the preset current integral threshold, thereby reducing the probability of overcurrent events and optimizing the operation stability of the surgical robot during the operation.

[0043] Figure 2 This is a flowchart of a surgical robot control method provided in an embodiment of the present invention. This embodiment, which shares the same inventive concept as the surgical robot control method described in the previous embodiment, further describes the operation of a control parameter analysis model. This method can be executed by a surgical robot control device, which can be implemented using software and / or hardware and integrated into a computer device with application development capabilities.

[0044] like Figure 2 As shown, the surgical robot control method of this embodiment includes the following steps:

[0045] S210 , obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window.

[0046] S220 , inputting the sequence data into the input layer of the control parameter analysis model, and performing feature extraction through the shared hidden layer of the control parameter analysis model to obtain a data feature extraction result.

[0047] The control parameter analysis model can be a multilayer perceptron, consisting of an input layer, hidden layers, and an output layer. Since the model outputs two data points: stiffness parameters and predicted terminal current integrals, the output layer consists of two output branches. Each output branch is somewhat independent, allowing it to learn and adjust based on its own objectives. However, since they are all based on the same input data, there is a certain degree of correlation. The output of one branch may constrain or influence the decision-making of another branch.

[0048] For both output branches, the hidden layer is a shared hidden layer. The shared hidden layer can perform preliminary feature extraction and processing on the input data to extract a common feature representation. Specifically, the sequence data acquired in S210 is input into the shared hidden layer for feature extraction, obtaining corresponding data feature extraction results.

[0049] The shared hidden layer can include a bidirectional long short-term memory network and / or a Transformer architecture. This captures the dependencies between sequential data within a preset time window. For example, it can extract the past motion and force characteristics of an oscillating saw to predict the optimal stiffness adjustment value at the current moment to adapt to changing working conditions.

[0050] S230 , inputting the data feature extraction results into the first branch hidden layer and the second branch hidden layer of the control parameter analysis model respectively, to obtain corresponding first features and second features respectively.

[0051] After the shared hidden layer, the corresponding output layer network is divided into two independent branches. Each branch can have its own independent hidden layer to further process and transform the features extracted by the shared hidden layer to adapt to its own output task.

[0052] S240: Input the first feature into the first output layer corresponding to the first branch hidden layer to obtain a stiffness parameter.

[0053] Among them, the first branch hidden layer uses the preset stiffness model as an intermediate constraint and performs calculations based on this constraint directly in the neuron calculation. The preset stiffness model can be represented by a function, such as K = ραF, where K is the stiffness, F is the end force, ρ is the bone density in the osteotomy area, and α is the preset coefficient. In the neural network, this constraint condition needs to be incorporated into the calculation of the first branch hidden layer. The result of the stiffness parameter output is sensitive to the osteotomy density, which can enhance the adaptability of the trained model to the surgical scenario and can be used in osteotomy scenarios with different bone densities.

[0054] S250 , inputting the second feature and the stiffness parameter into the second output layer corresponding to the second branch hidden layer to obtain a terminal current integral prediction value.

[0055] In the forward operation of the model, the second feature and the stiffness parameter are concatenated along the dimension to form a new input data. This concatenated input data is input to the second branch hidden layer for feature transformation, and then the terminal current integral prediction value is obtained through the output layer.

[0056] S260 , updating the control parameters of the robotic arm according to the stiffness parameter and the terminal current integral prediction value, and controlling the operation process of the robotic arm based on the updated control parameters.

[0057] The technical solution of this embodiment is to obtain sequence data of the joint angle, joint angular acceleration, terminal force and bone density in the osteotomy area of the surgical robot's robotic arm during operation in a preset time window; input the sequence data into the input layer of the control parameter analysis model, and perform feature extraction through the shared hidden layer of the control parameter analysis model to obtain data feature extraction results; input the data feature extraction results into the first branch hidden layer and the second branch hidden layer of the control parameter analysis model respectively to obtain corresponding first features and second features respectively; input the first feature into the first output layer corresponding to the first branch hidden layer to obtain the stiffness parameter; wherein the first branch hidden layer uses the preset stiffness model as the intermediate constraint; input the second feature and the stiffness parameter into the second output layer corresponding to the second branch hidden layer to obtain the terminal current integral prediction value; update the control parameters of the robotic arm according to the stiffness parameters and the terminal current integral prediction value, and control the operation process of the robotic arm based on the updated control parameters. The technical solution of the embodiment of the present invention solves the problem of overcurrent during the operation of the surgical robot affecting the progress of the operation. Through a pre-trained physical information neural network, the power parameters of the robotic arm can be determined under the constraint of a preset current integration threshold, thereby reducing the probability of overcurrent events and optimizing the operating stability of the surgical robot during surgery.

[0058] Figure 3 This is a flowchart of a surgical robot control method provided in an embodiment of the present invention. This embodiment, which shares the same inventive concept as the surgical robot control method described in the previous embodiment, further describes the training process of a control parameter analysis model. This method can be executed by a surgical robot control device, which can be implemented using software and / or hardware and integrated into a computer device with application development capabilities.

[0059] like Figure 3 As shown, the surgical robot control method of this embodiment includes the following steps:

[0060] S310. Obtain model training sample data, wherein the model training sample data includes control parameters, joint angles, joint angular accelerations, end forces, bone density in the osteotomy area, and corresponding stiffness data labels and current integral value labels of the robotic arm during operation.

[0061] S320 , inputting the model training sample data into the control parameter analysis model to be trained, and obtaining the corresponding first stiffness parameter and first terminal current integral prediction value.

[0062] Among them, the hidden layer in the control parameter analysis model that needs to be trained uses the preset stiffness model as intermediate constraint information.

[0063] S330. Calculate the first learning loss corresponding to the first stiffness parameter and the corresponding stiffness data label, calculate the second learning loss corresponding to the first terminal current integral prediction value and the corresponding current integral numerical label, and calculate the third learning loss of the relationship between the robotic arm control parameter and the dynamic constraint corresponding to the first stiffness parameter.

[0064] The learning loss can be calculated using cross entropy loss, mean square error loss, or other loss calculation methods. Different learning losses can be calculated using different methods or the same method.

[0065] S340: Reversely update the parameters in the control parameter analysis model that need to be trained based on the first learning loss, the second learning loss, and the third learning loss to complete the model training process.

[0066] The first learning loss, L1, measures the difference between the first stiffness parameter and the corresponding stiffness data label. The second learning loss, L2, measures the difference between the predicted value of the first terminal current integral and the corresponding current integral value label. The third learning loss, L3, calculates the loss of the relationship between the manipulator control parameters and the dynamic constraints corresponding to the first stiffness parameter.

[0067] The total loss L can be the accumulation of the first learning loss, the second learning loss, and the third learning loss, or a weighted accumulation. The parameters of the model to be trained can be updated based on the total loss backpropagation.

[0068] In addition, during the training process of the control parameter analysis model, the weight parameters corresponding to the first learning loss, the second learning loss, and the third learning loss may also be updated.

[0069] During the model training process, physical constraints can reduce dependence on labeled data, while safety control can be achieved through current integral warning. At the same time, the model training target is adjustable. By adjusting the threshold adaptive value, higher stiffness or lower overcurrent probability can be selected.

[0070] The trained control parameter analysis model can be applied in the operation of the surgical robot.

[0071] S350: Obtain sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window.

[0072] S360: Input the sequence data into the control parameter analysis model pre-trained in the above steps to obtain the stiffness parameter and the terminal current integral prediction value.

[0073] Among them, the constraints of the control parameter analysis model during the training process include the dynamic constraint relationship between the end force and the joint torque, the preset stiffness model and the preset current integration threshold.

[0074] S370. Update the control parameters of the robotic arm according to the stiffness parameters and the terminal current integral prediction value, and control the operation process of the robotic arm based on the updated control parameters.

[0075] The technical solution of this embodiment is to obtain model training sample data, wherein the model training sample data includes the control parameters, joint angles, joint angular accelerations, terminal forces, bone density in the osteotomy area, and corresponding stiffness data labels and current integral numerical labels of the robotic arm during operation; input the model training sample data into the control parameter analysis model to be trained to obtain the corresponding first stiffness parameter and the first terminal current integral prediction value; wherein the hidden layer in the control parameter analysis model to be trained uses the preset stiffness model as the intermediate constraint information; calculate the first stiffness parameter and the first learning loss corresponding to the corresponding stiffness data label, calculate the first terminal current integral prediction value and the second learning loss corresponding to the corresponding current integral numerical label, and calculate the robotic arm control parameter and dynamics corresponding to the first stiffness parameter The third learning loss of the constraint relationship; based on the first learning loss, the second learning loss and the third learning loss, the parameters in the control parameter analysis model that need to be trained are reversely updated to complete the model training process; the sequence data of the joint angle, joint angular acceleration, end force and bone density of the osteotomy area of the surgical robot's manipulator arm during operation in a preset time window are obtained; the sequence data is input into the control parameter analysis model pre-trained in the above steps to obtain the stiffness parameter and the end current integral prediction value; wherein, the constraint conditions of the control parameter analysis model during the training process include the dynamic constraint relationship between the end force and the joint torque, the preset stiffness model and the preset current integral threshold; the control parameters of the manipulator arm are updated according to the stiffness parameter and the end current integral prediction value, and the operation process of the manipulator arm is controlled based on the updated control parameters. The technical solution of the embodiment of the present invention solves the problem of overcurrent affecting the progress of the operation of the surgical robot during operation. The power parameters of the manipulator arm can be determined under the constraint of the preset current integral threshold through a pre-trained physical information neural network, thereby reducing the probability of overcurrent events and optimizing the operation stability of the surgical robot during surgery.

[0076] Figure 4 This is a schematic diagram of the structure of a surgical robot control device provided by an embodiment of the present invention. This embodiment is applicable to scenarios where surgical robots are controlled. The surgical robot control device can be implemented using software and / or hardware and integrated into a computer terminal device with application development capabilities.

[0077] like Figure 4As shown, the surgical robot control device includes: a data acquisition module 410, a data analysis module 420 and a robotic arm control module 430.

[0078] Among them, the data acquisition module 410 is used to obtain the serial data of the joint angle, joint angular acceleration, end force and bone density in the osteotomy area of the surgical robot's robotic arm during operation in a preset time window; the data analysis module 420 is used to input the serial data into a pre-trained control parameter analysis model to obtain the stiffness parameters and the end current integral prediction value; wherein, the constraints of the control parameter analysis model during the training process include the dynamic constraint relationship between the end force and the joint torque, the preset stiffness model and the preset current integral threshold; the robotic arm control module 430 is used to update the control parameters of the robotic arm according to the stiffness parameters and the end current integral prediction value, and control the operation process of the robotic arm based on the updated control parameters.

[0079] The technical solution of this embodiment is to obtain the sequence data of the joint angle, joint angular acceleration, terminal force and bone density of the osteotomy area of the surgical robot's manipulator arm during operation in a preset time window; input the sequence data into a pre-trained control parameter analysis model to obtain the stiffness parameters and the terminal current integral prediction value; wherein, the constraints of the control parameter analysis model during the training process include the dynamic constraint relationship between the terminal force and the joint torque, the preset stiffness model and the preset current integral threshold; update the control parameters of the manipulator arm according to the stiffness parameters and the terminal current integral prediction value, and control the operation process of the manipulator arm based on the updated control parameters. The technical solution of the embodiment of the present invention solves the problem of overcurrent affecting the progress of the operation during the operation of the surgical robot. The power parameters of the manipulator arm can be determined under the constraint of the preset current integral threshold, thereby reducing the probability of overcurrent events and optimizing the operation stability of the surgical robot during the operation.

[0080] In an optional embodiment, the data analysis module 420 is specifically configured to:

[0081] Inputting the sequence data into the input layer of the control parameter analysis model, and performing feature extraction through the shared hidden layer of the control parameter analysis model to obtain a data feature extraction result;

[0082] Inputting the data feature extraction results into the first branch hidden layer and the second branch hidden layer of the control parameter analysis model respectively to obtain the corresponding first feature and second feature respectively;

[0083] Inputting the first feature into the first output layer corresponding to the first branch hidden layer to obtain a stiffness parameter; wherein the first branch hidden layer uses a preset stiffness model as an intermediate constraint;

[0084] The second feature and the stiffness parameter are input into the second output layer corresponding to the second branch hidden layer to obtain the terminal current integral prediction value.

[0085] In an optional embodiment, the shared hidden layer includes a bidirectional long short-term memory network structure and / or a Transformer structure.

[0086] In an optional embodiment, the dynamic constraint relationship is a constraint relationship between the joint torque and the end force of the robotic arm established based on the Newton-Euler dynamic equation.

[0087] In an optional embodiment, the surgical robot control device further includes a model training module, specifically configured to:

[0088] Obtaining model training sample data, wherein the model training sample data includes control parameters, joint angles, joint angular accelerations, end forces, bone density in the osteotomy area, and corresponding stiffness data labels and current integral value labels of the robotic arm during operation;

[0089] Inputting the model training sample data into the control parameter analysis model to be trained to obtain the corresponding first stiffness parameter and the first terminal current integral prediction value; wherein the hidden layer in the control parameter analysis model to be trained uses the preset stiffness model as intermediate constraint information;

[0090] Calculating a first learning loss corresponding to a first stiffness parameter and a corresponding stiffness data label, calculating a second learning loss corresponding to a first terminal current integral prediction value and a corresponding current integral numerical label, and calculating a third learning loss corresponding to a relationship between a manipulator control parameter and a dynamic constraint corresponding to the first stiffness parameter;

[0091] Based on the first learning loss, the second learning loss and the third learning loss, the parameters in the control parameter analysis model that need to be trained are reversely updated to complete the model training process.

[0092] In an optional embodiment, the surgical robot control device further includes a model training module, which can also be used to:

[0093] Update the weight parameters corresponding to the first learning loss, the second learning loss, and the third learning loss.

[0094] The surgical robot control device provided in the embodiment of the present invention can execute the surgical robot control method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0095] Figure 5 A schematic structural diagram of a computer device provided in an embodiment of the present invention. Figure 5 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 12 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capabilities, such as an intelligent controller, a server, a mobile phone, or other terminal devices.

[0096] like Figure 5 As shown, computer device 12 is implemented as a general-purpose computing device. Components of computer device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, and a bus 18 that connects various system components (including system memory 28 and processing unit 16).

[0097] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0098] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0099] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 5 Not shown, often called a "hard drive"). Although Figure 5 Not shown, a magnetic disk drive for reading and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0100] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0101] The computer device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, a pointing device, a display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with the other modules of the computer device 12 via the bus 18. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RFID systems, tape drives, and data backup storage systems.

[0102] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the surgical robot control method provided in the embodiment of the present invention, which includes:

[0103] Obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window;

[0104] The sequence data is input into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values. The constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold.

[0105] The control parameters of the manipulator are updated according to the stiffness parameters and the terminal current integral prediction value, and the operation process of the manipulator is controlled based on the updated control parameters.

[0106] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the surgical robot control method provided in any embodiment of the present invention is implemented. The method includes:

[0107] Obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window;

[0108] The sequence data is input into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values. The constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold.

[0109] The control parameters of the manipulator are updated according to the stiffness parameters and the terminal current integral prediction value, and the operation process of the manipulator is controlled based on the updated control parameters.

[0110] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0111] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0112] Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0113] Computer program code for performing the operations of the present invention can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, Python, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] An embodiment of the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the surgical robot control method provided in any embodiment of the present application.

[0115] The computer program product, during implementation, may be written in one or more programming languages, or a combination thereof, for performing the operations of the present invention. The programming languages include object-oriented programming languages such as Java, Smalltalk, Python, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0116] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computing device. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they can be implemented using program code executable by a computer device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

[0117] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A surgical robot control method, characterized in that: include: Obtaining sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window; Inputting the sequence data into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values; wherein the constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold; The control parameters of the robotic arm are updated according to the stiffness parameter and the terminal current integral prediction value, and the operation process of the robotic arm is controlled based on the updated control parameters.

2. The method according to claim 1, characterized in that The step of inputting the sequence data into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values includes: Inputting the sequence data into the input layer of the control parameter analysis model, and performing feature extraction via the shared hidden layer of the control parameter analysis model to obtain a data feature extraction result; Inputting the data feature extraction results into the first branch hidden layer and the second branch hidden layer of the control parameter analysis model respectively to obtain corresponding first features and second features respectively; Inputting the first feature into a first output layer corresponding to the first branch hidden layer to obtain a stiffness parameter; wherein the first branch hidden layer uses the preset stiffness model as an intermediate constraint; The second feature and the stiffness parameter are input into the second output layer corresponding to the second branch hidden layer to obtain the terminal current integral prediction value.

3. The method according to claim 2, characterized in that The shared hidden layer includes a bidirectional long short-term memory network structure and / or a Transformer structure.

4. The method according to claim 1, wherein The dynamic constraint relationship is a constraint relationship between the joint torque of the robotic arm and the end force established based on the Newton-Euler dynamic equation.

5. The method according to any one of claims 1 to 4, characterized in that: The training process of the control parameter analysis model: Obtaining model training sample data, wherein the model training sample data includes control parameters, joint angles, joint angular accelerations, end forces, bone density in the osteotomy area, and corresponding stiffness data labels and current integral value labels of the robotic arm during operation; Inputting the model training sample data into the control parameter analysis model to be trained to obtain the corresponding first stiffness parameter and the first terminal current integral prediction value; wherein the hidden layer in the control parameter analysis model to be trained uses the preset stiffness model as intermediate constraint information; Calculating a first learning loss corresponding to the first stiffness parameter and the corresponding stiffness data label, calculating a second learning loss corresponding to the first terminal current integral prediction value and the corresponding current integral value label, and calculating a third learning loss of the relationship between the manipulator control parameter corresponding to the first stiffness parameter and the dynamic constraint; The parameters in the control parameter analysis model that needs to be trained are reversely updated based on the first learning loss, the second learning loss, and the third learning loss to complete the model training process.

6. The method according to claim 5, characterized in that The training process of the control parameter analysis model further includes: Update the weight parameters corresponding to the first learning loss, the second learning loss, and the third learning loss.

7. A surgical robot control device, characterized in that: include: A data acquisition module is used to obtain sequence data of joint angles, joint angular accelerations, end forces, and bone density in the osteotomy area of the surgical robot's robotic arm during operation within a preset time window; a data analysis module, configured to input the sequence data into a pre-trained control parameter analysis model to obtain stiffness parameters and terminal current integral prediction values; wherein the constraints of the control parameter analysis model during training include the dynamic constraint relationship between terminal force and joint torque, a preset stiffness model, and a preset current integral threshold; A robotic arm control module is used to update the control parameters of the robotic arm according to the stiffness parameter and the terminal current integral prediction value, and control the operation process of the robotic arm based on the updated control parameters.

8. A computer device, characterized in that: The computer device comprises: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the surgical robot control method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the surgical robot control method as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the surgical robot control method according to any one of claims 1 to 6.

Images