Remote surgical robot control system based on a force feedback handle and its control method

Abstract

The present invention relates to the technical field of force feedback handles, and particularly to a remote surgical robot control system based on a force feedback handle and a control method thereof. The method includes the following steps: collecting multi-dimensional force data of the end effector of the surgical robot, and extracting the magnitude, direction, and timing characteristics of the contact force to obtain multi-dimensional force characteristic data; performing surgical operation safety baseline analysis based on the multi-dimensional force characteristic data to generate surgical safety operation baseline data; calculating a dynamic force feedback threshold according to the surgical safety operation baseline data to obtain force feedback threshold range data; performing adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters. Through real-time control instruction optimization and closed-loop feedback strategies, the system of the present invention can quickly respond to operation intentions and give early warnings of potential risks, ensuring that the surgical process is safer and more reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of force feedback handles, and in particular to a remote surgical robot control system based on a force feedback handle and a control method thereof. Background Art

[0002] A force feedback handle is a device that can provide physical feedback to users. It enhances the user's immersion and interaction experience by simulating the forces and tactile sensations in the real world. The working principle of a force feedback handle is based on sensing and measuring the forces exerted on an object, and then converting these force signals into feedback signals and transmitting them to the user to achieve the user's force perception or control of the object. A force feedback system generally includes the following components: Sensors: Used to sense the forces exerted on an object, such as strain gauges, force sensors, and pressure sensors, etc. They convert the magnitude and direction of the force into electrical signals. Processing unit: Receives the force signals transmitted by the sensors, processes and analyzes them, and performs operations such as filtering, amplification, and calibration according to preset algorithms. Feedback actuator: Provides force feedback to the user according to the instructions of the processing unit, and can be achieved through methods such as motors, pneumatic devices, vibrators, or tactile devices. User interface: An interface for the user to interact with the force feedback system, such as a touch screen, buttons, handles, etc. The user sends instructions or operations through it and perceives the forces generated by the feedback actuator. A remote surgical robot is a high-tech medical device that uses wireless networks and robotic technologies to allow surgeons to perform surgeries on patients at remote locations. The core of the control method of a remote surgical robot lies in force feedback technology, which allows doctors to feel the forces of the interaction between surgical tools and tissues in the patient's body at remote locations.

[0003] However, the traditional control methods of remote surgical robots based on force feedback handles often have the following problems: In traditional remote surgeries, doctors mainly rely on visual feedback for operations and lack the tactile sensation of physical operations, which limits the accuracy and safety of surgeries. Data transmission takes time, which may cause communication delays and affect the real-time nature of operations. At the same time, the accuracy and stability of force feedback devices have a significant impact on the operation effect. Summary of the Invention

[0004] Based on this, it is necessary for the present invention to provide a remote surgical robot control system based on a force feedback handle and a control method thereof to solve at least one of the above technical problems.

[0005] To achieve the above object, a control method of a remote surgical robot based on a force feedback handle includes the following steps:

[0006] Step S1: Collect multi-dimensional force data of the end effector of the surgical robot, extract the magnitude, direction, and timing characteristics of the contact force to obtain multi-dimensional force characteristic data; perform a surgical operation safety baseline analysis based on the multi-dimensional force characteristic data to generate surgical safety operation baseline data;

[0007] Step S2: Calculate the dynamic force feedback threshold according to the surgical safety operation baseline data to obtain force feedback threshold range data; perform adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters;

[0008] Step S3: Real-time collect the control input data of the operator through the force feedback handle to obtain operation intention data; perform a safety assessment on the operation intention data based on the dynamic force feedback control parameters to generate operation safety level data; perform multi-level response processing on the operation safety level data and optimize network transmission to generate real-time control instruction data;

[0009] Step S4: Plan the robot motion trajectory according to the real-time control instruction data to obtain robot execution trajectory data; perform real-time update of the force feedback parameters based on the robot execution trajectory data, thereby obtaining a closed-loop force feedback control strategy.

[0010] The present invention collects multi-dimensional force data from the end effector of a surgical robot, enabling a comprehensive understanding of the interaction force between the robot and the patient's tissue, including the magnitude, direction, and temporal characteristics of the contact force. By performing a surgical operation safety baseline analysis on the multi-dimensional force characteristic data, the system can provide a basis for real-time force monitoring during the surgical process. The generated surgical safety operation baseline data can dynamically adjust the force feedback parameters according to the specific surgical environment to prevent excessive or inappropriate forces from damaging the patient's tissue. The multi-dimensional data collection method ensures that the robot can adapt to changes in different tissue types, surgical conditions, and operation inputs, thereby reducing risks. The safety baseline data is tailored to the surgical environment to achieve dynamic safety monitoring, promptly identify abnormal force conditions, and ensure that surgical operations are always within the safe range. Based on the surgical safety operation baseline data, the force feedback threshold is dynamically calculated, enabling the force feedback system to accurately match the physical characteristics of the tissue and the surgical scenario. The dynamic threshold calculation can effectively cope with changes in different tissue stiffnesses, ensuring that when the operation approaches the critical range, feedback is promptly provided to the operator. The adaptive compression process optimizes different force feedback signals, improving the response efficiency of the force feedback system, enhancing both the precision and safety of surgical operations. The system can adaptively adjust the force feedback range according to different tissue types and surgical scenarios, providing a more precise tactile experience; the real-time fine-tuning of the force feedback signal helps the operator more precisely control the robot's movements, especially playing a key role in high-precision operations. By using a force feedback handle to collect the operator's control input data in real time and convert it into operation intention data, the system can quickly identify and respond to the operator's intentions. Based on the dynamic force feedback control parameters, a safety assessment is performed on the operation intention data to ensure that the operation is always within the safe range. The multi-level response mechanism and network transmission optimization technology prioritize the processing of critical control signals, reduce latency, and enhance the system's response ability in high-risk scenarios. The system performs real-time safety analysis on the actions input by the operator, promptly feedbacks the risks during the operation, and avoids dangerous actions. Through network optimization, critical command data is preferentially transmitted to ensure that the robot can quickly respond to the operator's operation requirements. According to the control instructions generated in real time, the robot's motion trajectory is planned to ensure that its actions always meet the surgical safety requirements. By performing real-time feedback parameter updates on the robot's execution trajectory data to form a closed-loop force feedback control strategy, the system can dynamically adjust the operation parameters according to the actual situation, further improving the safety and precision of the surgery. The real-time updated force feedback parameters ensure that the robot can continuously adjust according to the operation situation, maintain optimal performance, and reduce the error risk; the combination of trajectory planning and feedback update not only ensures the accuracy of the robot's actions but also prevents operation behaviors that exceed the safe force limit, further safeguarding the patient's safety.

[0011] The present invention also provides a remote surgical robot control system based on a force feedback handle for performing the above-mentioned remote surgical robot control method based on a force feedback handle. The remote surgical robot control system based on a force feedback handle includes:

[0012] A force feature acquisition and analysis module, which is used to collect multi-dimensional force data of the end effector of the surgical robot, extract the magnitude, direction and timing features of the contact force to obtain multi-dimensional force feature data; perform surgical operation safety baseline analysis based on the multi-dimensional force feature data to generate surgical safety operation baseline data;

[0013] A dynamic force feedback threshold calculation module, which is used to calculate the dynamic force feedback threshold according to the surgical safety operation baseline data to obtain the force feedback threshold range data; perform adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters;

[0014] An operation intention and safety evaluation module, which is used to collect the control input data of the operator in real time through the force feedback handle to obtain operation intention data; perform safety evaluation on the operation intention data based on the dynamic force feedback control parameters to generate operation safety level data; perform multi-level response processing on the operation safety level data and optimize network transmission to generate real-time control instruction data;

[0015] A closed-loop force feedback control module, which is used to plan the robot motion trajectory according to the real-time control instruction data to obtain the robot execution trajectory data; perform real-time update of the force feedback parameters based on the robot execution trajectory data, so as to obtain the closed-loop force feedback control strategy.

[0016] By collecting and analyzing multi-dimensional force characteristics (such as the magnitude, direction, and timing characteristics of force), the system can comprehensively perceive and analyze different forces generated during the surgical process. This can help identify and monitor key surgical operation stages, ensuring the accuracy and safety of the operation. Based on the collected multi-dimensional force data, a safety baseline analysis can be performed to provide reliable basic data for subsequent surgical operation safety assessment. This means that the system can establish a standard safety operation baseline for different surgical scenarios, thus providing effective support for real-time evaluation and feedback. The surgical safety operation baseline data can be dynamically adjusted according to the needs of the actual surgical scenario to adapt to the characteristics of different patients and surgical sites, thereby providing personalized safety operation references for doctors. Through the calculation of dynamic force feedback thresholds based on the safety operation baseline data, the system can adjust the force feedback threshold in real time according to the actual situation during the surgical process to ensure that the force feedback at each operation stage meets the safety standards. This can better handle force changes and the requirements in different situations during the surgical process. The adaptive compression processing of different force feedback signals helps reduce redundant signals, improve control accuracy, and avoid system burden and delay caused by excessive information. This processing can make the force feedback control parameters more accurate and reflect the needs of surgical operations in real time. The dynamically adjusted force feedback control parameters provide a more accurate and comfortable force feedback experience, enabling the operator to better perceive the operation environment during the surgery, reducing the surgical difficulty, and improving the stability and safety of the surgery. By collecting the operator's control input data through the force feedback handle, the system can real-time sense the operator's intention and operation behavior. This provides real-time data support for operation safety assessment, enabling the system to judge at any time whether there are potential safety risks in the operator's operation. The combination of dynamic force feedback control parameters and operation intention data helps real-time evaluate the operation safety of the operator. If potential risks are detected, the system can generate operation safety level data according to the safety assessment results, provide safety warnings in a timely manner, and avoid misoperations or high-risk operations. The multi-level response processing and network transmission optimization of the operation safety level data enable the system to provide different responses at different risk levels, ensuring the transmission efficiency and timeliness of control instructions. In this way, the system can automatically optimize operation feedback and transmit control instructions in real time according to the actual situation, improving the reaction speed and operation accuracy. Network transmission optimization can reduce data transmission delay and ensure the timely transmission of real-time control instructions. Especially in scenarios such as remote surgery and robot control, the efficiency and low latency of transmission are crucial for ensuring the precise execution of the surgery. The combination of robot motion trajectory planning and real-time force feedback control can achieve closed-loop control. By tracking the execution of real-time control instructions, the robot system can adjust the execution trajectory and force feedback parameters in real time to ensure that the force feedback during the surgical process always remains within the safe range, improving the manipulation accuracy and stability.Real-time updating of force feedback parameters based on the robot's execution trajectory data helps to make precise force feedback adjustments according to changes in the actual operation process, avoiding operation discomfort or errors caused by lagging force feedback parameters. The adaptive characteristics of closed-loop control enhance the system's adaptability in complex operation environments. Through precise trajectory planning and real-time force feedback adjustment, the system can ensure the accuracy and safety of the robot during movement, avoiding operations beyond the preset safety range. At the same time, the application of closed-loop control can reduce operation errors, optimize surgical efficiency, and improve success rates and patient safety. Description of the Drawings

[0017] Other features, objectives, and advantages of the present invention will become more apparent by reading the detailed description of the non-restrictive embodiments with reference to the following drawings:

[0018] Figure 1 Schematic diagram of the step flow of the remote surgical robot control method based on a force feedback handle of the present invention;

[0019] Figure 2 For Figure 1 Detailed step flow diagram of step S1 in

[0020] Figure 3 For Figure 1 Detailed step flow diagram of step S2 in Detailed Embodiments

[0021] The technical method of the present invention will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0022] In addition, the drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0023] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly, the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve the above object, please refer to Figures 1 to 3 , the present invention provides a method for controlling a remote surgical robot based on a force feedback handle, and the method includes the following steps:

[0025] Step S1: Collect multi-dimensional force data of the end effector of the surgical robot, and extract the magnitude, direction, and timing characteristics of the contact force to obtain multi-dimensional force characteristic data; perform a surgical operation safety baseline analysis based on the multi-dimensional force characteristic data to generate surgical safety operation baseline data;

[0026] Step S2: Calculate a dynamic force feedback threshold based on the surgical safety operation baseline data to obtain force feedback threshold range data; perform an adaptive compression process on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters;

[0027] Step S3: Real-time collect the control input data of the operator through the force feedback handle to obtain operation intention data; perform a safety evaluation on the operation intention data based on the dynamic force feedback control parameters to generate operation safety level data; perform a multi-level response process on the operation safety level data and optimize network transmission to generate real-time control instruction data;

[0028] Step S4: Plan the robot motion trajectory according to the real-time control instruction data to obtain robot execution trajectory data; perform real-time update of the force feedback parameters based on the robot execution trajectory data, so as to obtain a closed-loop force feedback control strategy.

[0029] In an embodiment of the present invention, referring to Figure 1 shown, it is a schematic flow chart of the steps of a method for controlling a remote surgical robot based on a force feedback handle according to the present invention. In this example, the method for controlling a remote surgical robot based on a force feedback handle includes the following steps:

[0030] Step S1: Collect multi-dimensional force data of the end effector of the surgical robot, and extract the magnitude, direction, and timing characteristics of the contact force to obtain multi-dimensional force characteristic data; perform a surgical operation safety baseline analysis based on the multi-dimensional force characteristic data to generate surgical safety operation baseline data;

[0031] In the embodiment of the present invention, an end effector integrated with a six - dimensional force sensor array is used to collect real - time force data through sensor nodes, including linear force components in the x, y, and z directions and torque components around the x, y, and z axes, and record the spatial position and attitude angle data of the end effector. The collected force data and attitude data are merged to calculate the magnitude, direction angle, and spatial distribution matrix of the force vector. The time - domain signal is denoised using wavelet transform, and frequency - domain characteristic parameters are extracted through Fourier transform to generate multi - dimensional force characteristic data including force magnitude, direction, and time - series characteristics. Further, by statistically analyzing the correlation between force characteristics and operations in common surgical scenarios (such as puncture, cutting, suturing), a surgical operation safety baseline is constructed based on a support vector machine classification model, and the statistical mean and threshold range of the baseline data are extracted to generate safety baseline data applicable to minimally invasive surgical operations.

[0032] Step S2: Calculate the dynamic force feedback threshold range data according to the surgical safety operation baseline data; perform adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters;

[0033] In the embodiment of the present invention, according to the surgical safety operation baseline data generated in step S1, an adaptive fuzzy inference algorithm is used to establish a dynamic threshold model according to the tissue hardness characteristics of different surgical sites. Taking the real - time force characteristics input as a variable, calculate the upper and lower limit values of each force interval to generate a preliminary force feedback threshold range. Based on the patient's individualized mechanical property data (such as tissue elastic modulus, shear modulus), use the Bayesian update method to adjust the preliminary threshold range, and finally determine the dynamic force feedback threshold range data. For the force feedback signal within the threshold range, a logarithmic compression function is used for non - linear compression processing to suppress the impact of excessive force values on the feedback handle, and generate dynamically adjustable force feedback control parameters to adapt to the requirements of different surgical scenarios.

[0034] Step S3: Real - time collect the operator's control input data through the force feedback handle to obtain operation intention data; perform a safety assessment on the operation intention data based on the dynamic force feedback control parameters to generate operation safety level data; perform multi - level response processing on the operation safety level data and optimize network transmission to generate real - time control instruction data;

[0035] The embodiment of the present invention collects the operator's control input data, including the joystick displacement, pressing force and finger movement rate, through a high-precision force feedback handle, and uses a neural network model to extract the characteristic data of the operation intention. Combined with the dynamic force feedback control parameters generated in step S2, the operation intention is evaluated for safety, and the risk index is calculated and classified into three operation levels: safe, warning and dangerous. A graded response strategy is set for different levels of operation signals. For example, when a dangerous level signal is detected, the vibration or resistance mechanism of the force feedback handle is immediately activated for prompting; when it is a warning level, a visual warning signal is provided and the feedback force of the handle is enhanced. Further, through the multi-channel 5G communication protocol, the control signal is optimized for network transmission to ensure low-latency and high-precision transmission of real-time control command data.

[0036] Step S4: planning the robot motion trajectory according to the real-time control instruction data to obtain the robot execution trajectory data; updating the force feedback parameters in real time based on the robot execution trajectory data to obtain a closed-loop force feedback control strategy.

[0037] According to the real-time control instruction data generated in step S3, the embodiment of the present invention uses the cubic spline interpolation method to plan the robot's motion trajectory, generate smooth execution trajectory data, and ensure that the end effector movement conforms to the predetermined operation path. Combined with the real-time monitored end effector force characteristic data, the robot's movement speed and force output are dynamically adjusted through the force control algorithm to avoid tissue damage caused by execution errors. At the same time, the force feedback control parameters are updated based on the error between the trajectory execution and the actual feedback, and the control strategy is optimized. For example, in the suturing operation, the surgical needle puncture depth and strength are adjusted in real time to ensure that there is no tissue tearing, and a closed-loop force feedback control strategy is implemented, thereby further improving the accuracy and safety of the surgical operation.

[0038] The present invention collects multi-dimensional force data from the end effector of a surgical robot, enabling a comprehensive understanding of the interaction force between the robot and the patient's tissue, including the magnitude, direction, and temporal characteristics of the contact force. By performing a surgical operation safety baseline analysis on the multi-dimensional force characteristic data, the system can provide a basis for real-time force monitoring during the surgical process. The generated surgical safety operation baseline data can dynamically adjust the force feedback parameters according to the specific surgical environment to prevent excessive or inappropriate forces from damaging the patient's tissue. The multi-dimensional data collection method ensures that the robot can adapt to changes in different tissue types, surgical conditions, and operation inputs, thereby reducing risks. The safety baseline data is tailored to the surgical environment to achieve dynamic safety monitoring, promptly identify abnormal force conditions, and ensure that surgical operations are always within the safe range. Based on the surgical safety operation baseline data, the force feedback threshold is dynamically calculated, enabling the force feedback system to accurately match the physical characteristics of the tissue and the surgical scenario. The dynamic threshold calculation can effectively cope with changes in different tissue stiffnesses, ensuring that when the operation approaches the critical range, feedback is promptly provided to the operator. The adaptive compression process optimizes different force feedback signals, improving the response efficiency of the force feedback system, enhancing both the precision of surgical operations and safety. The system can adaptively adjust the force feedback range according to different tissue types and surgical scenarios, providing a more precise tactile experience; the real-time fine-tuning of the force feedback signal helps the operator more precisely control the robot's movements, especially playing a key role in high-precision operations. By using a force feedback handle to collect the operator's control input data in real time and convert it into operation intention data, the system can quickly identify and respond to the operator's intentions. Based on the dynamic force feedback control parameters, a safety assessment is performed on the operation intention data to ensure that the operation is always within the safe range. The multi-level response mechanism and network transmission optimization technology prioritize the processing of key control signals, reduce latency, and enhance the system's response ability in high-risk scenarios. The system performs real-time safety analysis on the actions input by the operator, promptly feedbacks the risks during the operation, and avoids dangerous actions. Through network optimization, key instruction data is preferentially transmitted to ensure that the robot can quickly respond to the operator's operation requirements. According to the control instructions generated in real time, the robot's motion trajectory is planned to ensure that its actions always meet the surgical safety requirements. By performing real-time feedback parameter updates on the robot's execution trajectory data to form a closed-loop force feedback control strategy, the system can dynamically adjust the operation parameters according to the actual situation, further improving the safety and precision of the surgery. The real-time updated force feedback parameters ensure that the robot can continuously adjust according to the operation situation, maintain optimal performance, and reduce the error risk; the combination of trajectory planning and feedback update not only ensures the accuracy of the robot's actions but also prevents operation behaviors that exceed the safe force limit, further safeguarding the patient's safety.

[0039] Preferably, step S1 includes the following steps:

[0040] Step S11: Collect force data of the end effector of the surgical robot through a multi-dimensional force sensor array to obtain six-dimensional force data; collect the spatial position and attitude angle information of the end effector to obtain spatial attitude data;

[0041] Step S12: Combine the six-dimensional force data and the spatial attitude data into multi-dimensional force data;

[0042] Step S13: Calculate the magnitude of the resultant force and its three-dimensional components based on the multi-dimensional force data, and perform spatial torque decoupling processing to obtain contact force feature vector data;

[0043] Step S14: Perform time-domain sampling and frequency-domain analysis on the contact force feature vector data to obtain multi-dimensional force feature data;

[0044] Step S15: Perform surgical operation safety baseline analysis based on the multi-dimensional force feature data to generate surgical safety operation baseline data.

[0045] As an embodiment of the present invention, referring to Figure 2 shown, for Figure 1 the detailed step flow diagram of step S1 in

[0046] Step S11: Collect force data of the end effector of the surgical robot through a multi-dimensional force sensor array to obtain six-dimensional force data; collect the spatial position and attitude angle information of the end effector to obtain spatial attitude data;

[0047] In the embodiment of the present invention, a multi-dimensional force sensor array is used to collect force data of the end effector of the surgical robot. The array includes force sensors in six directions, which are respectively used to collect the linear forces of the end effector in the X, Y, and Z directions, as well as the torque data around the X, Y, and Z axes. By synchronously collecting the output signals of the sensors, six-dimensional force data is obtained. To ensure high-precision collection, a sampling frequency of 200 Hz is adopted. At the same time, the spatial position of the end effector (for example, the coordinate position of the end effector relative to the robot base) and the attitude angles (for example, pitch angle, yaw angle, and roll angle) are collected in real time through an integrated inertial measurement unit (IMU) module to obtain spatial attitude data, and the sampling frequency is set to 50 Hz to ensure accurate tracking of the spatial movement of the end effector.

[0048] Step S12: Combine the six-dimensional force data and the spatial attitude data into multi-dimensional force data;

[0049] In the embodiment of the present invention, the six - dimensional force data obtained in step S11 and the spatial attitude data are merged through a data fusion algorithm. Using the Extended Kalman Filter (EKF) method, the six - dimensional force data and the spatial attitude data are fused according to time synchronization, so as to generate multi - dimensional force data containing force and spatial attitude information. Specifically, through EKF, the six - dimensional force data and the attitude data collected by the IMU are co - estimated to eliminate the noise interference in the collection process and merge them into a unified data vector, which is convenient for subsequent mechanical analysis and processing. The merged data includes the magnitude and direction of the force, as well as the spatial position and attitude information of the end - effector, forming multi - dimensional force data, which provides a basis for subsequent contact force analysis.

[0050] Step S13: Calculate the magnitude of the resultant force and its three - dimensional components according to the multi - dimensional force data, and perform spatial moment decoupling processing, so as to obtain the contact force feature vector data;

[0051] In the embodiment of the present invention, according to the multi - dimensional force data generated in step S12, the magnitude of the resultant force and its three - dimensional components are first calculated. The magnitude of the resultant force is obtained by calculating the norm of the six - dimensional force data, that is, the total magnitude of the resultant force. Then, using this resultant force data, the force in each direction is decomposed three - dimensionally to obtain the linear force components in the X, Y, and Z axis directions respectively. At the same time, a spatial moment decoupling algorithm is adopted to decouple the moment components from the force distribution to obtain the independent moment components in each direction. To ensure the accuracy of the decoupling process, a spatial moment decoupling algorithm based on the least - squares method is used, and accurate contact force feature vector data representing the force and moment values in each direction is obtained through multiple iterative optimizations.

[0052] Step S14: Perform time - domain sampling and frequency - domain analysis on the contact force feature vector data, so as to obtain multi - dimensional force feature data;

[0053] In the embodiment of the present invention, time - domain sampling and frequency - domain analysis are performed on the contact force feature vector data obtained in step S13. In terms of time - domain sampling, the force feature vector is regularly sampled at a sampling rate of 100Hz to ensure the capture of the high - frequency response of the force data. Then, the time - domain data is converted into frequency - domain data through Fourier transform, and the main components in the frequency domain, such as vibration frequency, amplitude, etc., are extracted to further analyze the regularity of the force change. The Fast Fourier Transform (FFT) method is used to convert the collected force data into a frequency - domain representation, analyze the main spectral components of the force signal, extract the relevant frequency and amplitude information, obtain multi - dimensional force feature data, reflect the change law and time - frequency characteristics of the force, and assist in subsequent safety baseline analysis.

[0054] Step S15: Perform surgical operation safety baseline analysis according to the multi - dimensional force feature data, so as to generate surgical safety operation baseline data.

[0055] In the embodiment of the present invention, based on the multi-dimensional force feature data extracted in step S14, a safety baseline analysis of surgical operations is performed. First, based on historical surgical data and a force feedback model, an association relationship between force and operation safety is established, and a clustering analysis method is used to identify typical force feedback patterns in different surgical scenarios. For example, in the puncture operation, by analyzing the relationship between the collected force data and factors such as tissue elasticity and hardness, safety thresholds for different surgical actions are set. Then, regression analysis and multi-dimensional feature fusion technology are used to model the multi-dimensional force feature data, generating operation safety baseline data that meets the requirements of surgical safety. This baseline data not only reflects the force feedback safety range under different surgical actions but also can adjust safety parameters according to real-time operations to ensure that the operation process does not exceed the safety range, generating surgical safety operation baseline data for real-time monitoring and control.

[0056] The present invention can accurately measure the six-dimensional force data (including three-dimensional force and three-dimensional torque) of the end effector of the surgical robot through a multi-dimensional force sensor array, and at the same time collect the spatial position and attitude angle information of the end effector to obtain complete spatial attitude data. The six-dimensional force data captures the complex force information when the actuator contacts the patient's tissue, and the spatial attitude data provides the position and direction basis for the mechanical behavior. The combination of these two lays the foundation for subsequent data fusion and feature analysis. The six-dimensional force data and the spatial attitude data are merged into multi-dimensional force data to form a more complete description of mechanical information, providing a data basis for the correlation analysis of force and spatial position in complex surgical scenarios. The merged data contains the coupling relationship between force and direction, position, supporting more accurate model analysis and dynamic adaptive adjustment. By calculating the resultant force magnitude of the contact force and its three-dimensional components from the multi-dimensional force data and performing spatial torque decoupling processing, the linear force and rotational torque components in the contact force can be accurately separated, generating contact force feature vector data. This decoupling processing improves the ability to understand complex mechanical environments and provides higher-precision basic data for force feedback and operation safety analysis. By performing time-domain sampling and frequency-domain analysis on the contact force feature vector, the time characteristics and frequency characteristics of force changes can be captured. Time-domain analysis can reflect the instantaneous changes of force, while frequency-domain analysis can reveal possible vibrations or abnormal fluctuations in long-term operations. These data provide richer information dimensions for multi-dimensional force feature extraction, supporting in-depth understanding and real-time monitoring of complex operation processes. According to the extracted multi-dimensional force feature data, a safety baseline analysis of surgical operations is performed to generate surgical safety operation baseline data. These baseline data reflect the reasonable and controllable mechanical range during the surgical process, can provide a reference for the real-time operation of the robot, and ensure that the operation force is always within the safety threshold. Through the monitoring of abnormal mechanical behaviors, the baseline analysis can timely remind the operator to avoid harming the patient's tissue.

[0057] Preferably, step S15 includes the following steps:

[0058] Step S151: Using the multi-dimensional force characteristic data, the three dimensions of force magnitude, spatial direction, and temporal variation are taken as the basic dimensions of the feature space, so as to construct the surgical operation force feedback feature space data with the grid density increasing as the force value increases, where the force magnitude is in Newton units, the spatial direction is in angular degrees, and the temporal variation is in the change rate per unit time;

[0059] In the embodiment of the present invention, according to the multi-dimensional force characteristic data, the three basic dimensions of force magnitude, spatial direction, and temporal variation are respectively extracted to construct the surgical operation force feedback feature space. The force magnitude is in Newton (N) units and directly uses the resultant force value in the multi-dimensional force characteristic data; the spatial direction calculates the direction angle relative to the reference coordinate system through the decoupled three-dimensional components and is represented in angular degrees (unit: degree); the temporal variation is obtained by calculating the rate of change of the force per unit time (unit: N / s). During the construction of the feature space, a grid division method is adopted, and the grid density is set to increase with the force magnitude to capture the subtle changes of larger force values with higher resolution. Specifically, when the force is less than 10N, the grid resolution is 1N, and when the force is greater than 10N, the resolution is 0.1N, forming a feature space with dynamic resolution to reflect the subtle changes of the surgical operation force.

[0060] Step S152: Perform a three-level safety threshold setting including a warning threshold, a danger threshold, and an emergency threshold according to the data of each dimension of the surgical operation force feedback feature space data, so as to obtain the initial safety threshold data;

[0061] In the embodiment of the present invention, using the force feedback feature space data constructed in step S151, three-level safety thresholds are set respectively for the three dimensions of force, direction, and temporal variation. The warning threshold corresponds to the upper limit of normal operation, the danger threshold represents the initial critical point of potential dangerous operations, and the emergency threshold represents the limit value that may cause tissue damage. For example, for the force dimension, the warning threshold is set to 15N, the danger threshold is set to 25N, and the emergency threshold is set to 30N; for the spatial direction dimension, a deviation from the reference direction exceeding 15° is regarded as a warning, exceeding 30° is dangerous, and exceeding 45° is an emergency; for the temporal variation rate, exceeding 10N / s is a warning, exceeding 20N / s is dangerous, and exceeding 30N / s is an emergency. By marking the safety levels in different regions of the feature space, the initial safety threshold data is generated.

[0062] Step S153: Dynamically adjust the threshold magnitude of the initial safety threshold data based on the force-bearing characteristics of human tissues, and establish a force mutation warning mechanism based on the temporal variation rate, so as to obtain the dynamic safety threshold data;

[0063] In the embodiments of the present invention, the initial safety threshold data generated in step S152 is dynamically adjusted in combination with the force-bearing characteristics of human tissues. Based on experimental research and literature data, the safety threshold is fine-tuned according to the force tolerance ranges of different tissue types (such as skin, soft tissue, bone). For example, for operations on soft tissue, the dangerous threshold of force can be reduced to 20 N, and the emergency threshold can be reduced to 25 N. At the same time, a force mutation warning mechanism is established by using the multi-dimensional time series analysis method. When the time series change rate exceeds the dangerous threshold and the duration exceeds 0.5 seconds, the system triggers a warning signal and dynamically lowers the dangerous and emergency thresholds to adapt to sudden changes during the operation. The adjusted dynamic safety threshold data can more accurately reflect the safety boundaries in the real-time surgical scenario.

[0064] Step S154: Construct a scene-adaptive baseline for the surgical scenario according to the dynamic safety threshold data, so as to obtain scene-adaptive baseline data, where the baseline construction includes force feature pattern recognition for different types of surgical actions, pattern matching for real-time surgical operations, and dynamic adjustment of safety baseline parameters based on the matching results according to the scene adaptation;

[0065] In the embodiments of the present invention, based on the dynamic safety threshold data obtained in step S153, a scene-adaptive baseline for the surgical scenario is constructed. Through the clustering analysis of historical surgical data, the force feature patterns of different types of surgical actions are identified. For example, the force change curve of a puncture operation usually shows an initial peak and then tends to be stable, while the force change of a suture operation is relatively uniform. The force features of the real-time surgical operation are matched with the identified patterns, the similarity is calculated using the dynamic time warping (DTW) algorithm, and the safety baseline parameters are adjusted according to the matching results. For example, the weights of different force features are dynamically adjusted. The scene-adaptive baseline can respond to surgical operation changes in real time, ensure the effectiveness of the baseline parameters in diverse scenarios, and the generated scene-adaptive baseline data provides a basis for subsequent evaluation.

[0066] Step S155: Calculate the deviation degree between the current operation force feature and the safety baseline according to the scene-adaptive baseline data and the surgical operation force feedback feature space data, and generate a real-time risk score based on the deviation degree, so as to obtain the safety assessment result data;

[0067] In the embodiment of the present invention, the deviation degree between the current operating force feature and the safety baseline is calculated by using the scene adaptive baseline data and the force feedback feature space data generated in step S154. The specific method is to quantify the deviation size by performing difference analysis on the real-time force data and the safety baseline parameters. For example, when the current force value is higher than the warning threshold but lower than the danger threshold, the deviation degree is defined as medium; when it exceeds the danger threshold but is lower than the emergency threshold, it is defined as high; when it exceeds the emergency threshold, it is defined as extremely high. At the same time, a real-time risk score is generated based on the deviation degree, and the score range is set to 0-100, graded as low risk (0-40), medium risk (41-70), and high risk (71-100), and finally the safety assessment result data is obtained.

[0068] Step S156: Perform feature fusion processing according to the surgical operation force feedback feature space data, the dynamic safety threshold data, the scene adaptive baseline data, and the safety assessment result data, so as to generate the surgical safety operation baseline data.

[0069] In the embodiment of the present invention, feature fusion processing is performed according to the safety assessment result data generated in step S155, as well as the force feedback feature space data, the dynamic safety threshold data, and the scene adaptive baseline data. A feature fusion algorithm based on weight weighting is adopted to perform unified normalization processing on the four types of data to ensure the dimensional consistency of different data dimensions. For example, the force feedback feature space data accounts for 40% of the total weight, the dynamic safety threshold data accounts for 30%, and the scene adaptive baseline data and the safety assessment result data each account for 15%. The comprehensive data obtained through fusion analysis is used to generate the surgical safety operation baseline data, which can dynamically guide the real-time feedback and regulation in the surgical operation process and achieve more precise safety operation guarantee.

[0070] The present invention constructs a multi-dimensional surgical operation force feedback feature space by taking the magnitude of force, spatial direction, and temporal variation as the basic dimensions of the feature space. The magnitude of force is intuitively quantified in Newton units, the spatial direction is accurately described in angular measure to represent the direction attribute, and the temporal variation reflects the dynamic characteristics through the rate of change within a unit of time. The design where the grid density increases with the increase in the force value pays more attention to the fine characteristics of large-force operations and helps to accurately capture the mechanical behavior characteristics of high-risk areas. Three-level safety thresholds, namely the warning threshold, danger threshold, and emergency threshold, are set in the surgical operation force feedback feature space, forming a hierarchical response mechanism. The division of the three-level thresholds enables the system to take different control measures according to the operation force and risk level, gradually improving the operation safety from warning to restricting actions and avoiding problems of over-response or under-response that may be caused by a single threshold. Combining with the force-bearing characteristics of human tissues, the initial safety thresholds are dynamically adjusted to make the thresholds more in line with the actual surgical scenario requirements. At the same time, the force mutation warning mechanism based on the temporal change rate can monitor and warn of sudden abnormalities in the mechanical changes during the operation in real time, effectively avoiding tissue damage or mechanical failures. By identifying the force feature patterns of surgical actions and matching the real-time operation with known patterns, the system can dynamically adjust the safety baseline parameters according to the matching results. The construction of the scenario-adaptive baseline enables the system to flexibly adapt to different surgical scenarios, ensuring the dynamic update and accuracy of the safety baseline. According to the scenario-adaptive baseline data and the surgical operation force feedback feature space data, the deviation degree between the current operation force feature and the safety baseline is calculated, and a real-time risk score is generated. Through the dynamic risk score and the safety assessment results, the system can provide real-time safety tips or take protective measures during the operation. By performing feature fusion processing on the surgical operation force feedback feature space data, dynamic safety threshold data, scenario-adaptive baseline data, and safety assessment result data, surgical safety operation baseline data is finally generated. The fusion processing synthesizes multi-dimensional information to ensure that the generated baseline data is more comprehensive and practical, providing an accurate and reliable reference standard for the operation of surgical robots.

[0071] Preferably, step S156 includes the following steps:

[0072] Step S1561: Normalize and correct the standard deviation of each dimension data of the surgical operation force feedback feature space data to obtain standardized feature mapping data, where the normalization process uses min-max linear transformation and the standard deviation correction is based on the kurtosis and skewness analysis of the data;

[0073] In the embodiment of the present invention, for the dimension data such as the magnitude of force, direction angle, and temporal change rate in the surgical operation force feedback feature space data, normalization is first performed. The min-max linear transformation is used to compress the data into the range of [0,1]. The normalization formula is: where x minand x max They are the minimum and maximum values of the specific dimension data respectively. After normalization, the standard deviation correction is performed on each dimension data, and the skewness and kurtosis of the data are calculated to adjust the symmetry and concentration of its distribution. For example, the feature data with skewness greater than 2 is adjusted to a nearly symmetric distribution (skewness close to 0) through logarithmic transformation. Finally, the standardized feature mapping data is generated, providing a unified standard data basis for subsequent analysis.

[0074] Step S1562: Perform cross-correlation analysis on the standardized feature mapping data, dynamic safety threshold data, and scenario adaptive baseline data to obtain feature subspace correlation data, where the cross-correlation analysis is specifically to identify the feature subspaces with significant correlation by calculating the mutual information gain rate between feature dimensions, and dynamically adjust the weight allocation strategy for each feature dimension;

[0075] Based on the standardized feature mapping data, dynamic safety threshold data, and scenario adaptive baseline data in step S1561, the embodiments of the present invention perform cross-correlation analysis. By calculating the mutual information gain rate (MI) between each data dimension, the feature subspaces with significant correlation are identified. The specific method is to calculate the mutual information gain rate formula for each pair of feature dimension data: where MI(A,B) is the mutual information between features A and B, and H(A) is the entropy value of feature A. When the gain rate is greater than the threshold 0.5, it is determined that this dimension has an important impact on the surgical safety assessment. At the same time, the weight allocation strategy is dynamically adjusted according to the analysis results. For example, the weight of the dimension with the highest correlation is increased by 10%, and the weight of the weakly correlated dimension is decreased by 5%, so as to generate feature subspace correlation data and improve the accuracy of multi-dimensional feature fusion.

[0076] Step S1563: Perform wavelet packet decomposition on the safety assessment result data, extract the multi-scale feature coefficients of the data, and perform adaptive recombination on the feature coefficients of each scale based on the empirical mode decomposition method to obtain the surgically safe operation baseline data reconstructed by inverse transformation.

[0077] The embodiments of the present invention perform wavelet packet decomposition on the safety assessment result data to extract the multi-scale feature coefficients at different frequencies. The specific process is to decompose the data through the Daubechies wavelet function for three layers to obtain low-frequency and high-frequency feature coefficients. The low-frequency features are retained to reflect the long-term trend, and the high-frequency features reveal the instantaneous changes. At the same time, combined with the empirical mode decomposition (EMD) method, the feature coefficients of different scales are adaptively recombined. For example, the high-frequency abnormal signals are separated from the basic signals by using the instantaneous frequency extraction algorithm to avoid noise interference. Finally, the inverse transformation is performed on the recombined feature coefficients to reconstruct the surgically safe operation baseline data combined with multi-dimensional features and dynamic adjustment, providing optimized real-time safety guidance for the surgical process and ensuring both operation accuracy and safety.

[0078] The present invention performs normalization processing and standard deviation correction on the force feedback feature space data of surgical operations to ensure that the data in each dimension is analyzed within the same scale range. The normalization uses the minimum-maximum linear transformation to make the data distribution uniform and facilitate subsequent processing. The standard deviation correction eliminates the influence of outliers on the data distribution through kurtosis and skewness analysis, improving the stability and accuracy of data analysis. Cross-correlation analysis is performed on the standardized feature mapping data, dynamic safety threshold data, and scenario adaptive baseline data to identify significant associated feature subspaces among the dimensions. By calculating the mutual information gain rate, the weight allocation strategy of the feature dimensions is dynamically adjusted to ensure that the analysis results are more in line with the actual surgical scenario. This method can highlight key features, reduce the interference of redundant information, and improve the accuracy of feature fusion. Wavelet packet decomposition is performed on the safety assessment result data to extract the multi-scale feature coefficients of the data, and the empirical mode decomposition method is combined to adaptively reconstruct the features at each scale. Finally, the baseline data for surgical safety operations is reconstructed through inverse transformation. Wavelet packet decomposition provides multi-scale analysis capabilities and can extract key features from both the macroscopic trend and microscopic changes; the empirical mode decomposition has strong adaptability and can effectively remove noise and retain important feature information. The inverse transformation reconstruction ensures the integrity and authenticity of the data and provides an accurate baseline for surgical safety operations.

[0079] Preferably, step S2 includes the following steps:

[0080] Step S21: Obtain the tissue hardness parameter of the surgical site, calculate the upper limit value of the safe puncture force based on the tissue hardness parameter using an adaptive algorithm, and perform threshold division to obtain preliminary force feedback threshold data, where the preliminary force feedback threshold data includes the upper threshold of the safe interval, the threshold interval of the warning interval, and the lower threshold of the dangerous interval;

[0081] Step S22: Perform dynamic force feedback threshold calculation on the preliminary force feedback threshold data according to the surgical safety operation baseline data to obtain force feedback threshold range data;

[0082] Step S23: Perform force feedback signal partition processing on the surgical operation force feedback feature space data according to the force feedback threshold range data, dividing it into a safe interval, a warning interval, and a dangerous interval to obtain force feedback partition data;

[0083] Step S24: Perform adaptive compression processing on the force feedback signals in different intervals according to the force feedback partition data to obtain partition compressed force feedback data;

[0084] Step S25: Transmit the partition compression force feedback data with priority through the 5G network to obtain dynamic force feedback control parameters. The priority transmission specifically means that the force feedback signal in the dangerous interval has the highest transmission priority, followed by the warning interval, and the lowest in the safe interval.

[0085] As an embodiment of the present invention, refer to Figure 3 shown in Figure 1 the detailed step flow schematic diagram of step S2 in

[0086] Step S21: Obtain the tissue hardness parameters of the surgical site, calculate the upper limit value of the safe puncture force based on the tissue hardness parameters using an adaptive algorithm, and perform threshold division to obtain preliminary force feedback threshold data. The preliminary force feedback threshold data includes the upper threshold of the safe interval, the threshold interval of the warning interval, and the lower threshold of the dangerous interval.

[0087] In the embodiment of the present invention, a hardness measuring instrument is used to collect the tissue hardness parameters of the surgical site. For example, a force measuring probe is used to measure the hardness range between 10 - 200 kPa, and the local hardness distribution of the specific surgical site is recorded. According to the tissue hardness parameters, the upper limit value of the safe puncture force is calculated based on an adaptive algorithm (such as a multivariable regression model combined with a tissue stress distribution function) to ensure that the force value adapts to the characteristics of different tissues. Taking the hardness data as input, a safe interval (for example, hardness 10 - 50 kPa corresponds to a puncture force of 0.1 - 1.0 N), a warning interval (hardness 50 - 150 kPa corresponds to a puncture force of 1.0 - 2.5 N), and a dangerous interval (hardness greater than 150 kPa corresponds to a puncture force exceeding 2.5 N) are set, and the preliminary force feedback threshold data is divided to ensure that the force feedback signal can clearly reflect the safety state of the surgical operation during the puncture process.

[0088] Step S22: Perform dynamic force feedback threshold calculation on the preliminary force feedback threshold data according to the surgical safety operation baseline data to obtain force feedback threshold range data.

[0089] In the embodiment of the present invention, according to the preliminary force feedback threshold data generated in step S21, combined with the surgical safety operation baseline data, it is optimized through a dynamic force feedback threshold calculation model. For example, the safe threshold range under different tissue characteristics is dynamically adjusted. For tissues with large hardness fluctuations, the upper and lower threshold values of the warning interval and the dangerous interval are reduced by 20%; for tissues with uniform hardness distribution, the safe interval range is expanded by 10%. The results after dynamic adjustment generate new force feedback threshold range data. For example, the safe interval is adjusted to 0.1 - 1.2 N, the warning interval is 1.2 - 2.0 N, and the dangerous interval is greater than 2.0 N, so as to better adapt to the changes in the actual surgical scenario.

[0090] Step S23: Perform force feedback signal partitioning processing on the surgical operation force feedback feature space data according to the force feedback threshold range data, and divide it into a safe interval, a warning interval, and a dangerous interval, so as to obtain force feedback partition data;

[0091] In the embodiment of the present invention, the force feedback threshold range data obtained in step S22 is matched with the surgical operation force feedback feature space data to divide the force feedback signals in different regions. For example, by detecting the current feedback force value point by point, the signal of 0.8N is classified into the safe interval, the signal of 1.5N is classified into the warning interval, and the signal of 2.8N is classified into the dangerous interval, and finally the force feedback partition data is formed. These data not only include the specific interval of each signal point, but also the interval distribution probability of the continuous feedback force signal, providing a basis for subsequent processing.

[0092] Step S24: Perform adaptive compression processing on the force feedback signals in different intervals according to the force feedback partition data, so as to obtain partition-compressed force feedback data;

[0093] In the embodiment of the present invention, according to the force feedback partition data in step S23, adaptive compression processing is performed on the force feedback signals in different intervals. For the signals in the safe interval, a logarithmic compression algorithm is used to reduce the data volume; the signals in the warning interval are non-linearly compressed (such as squaring the signal) to retain their change trend; the signals in the dangerous interval are linearly amplified to enhance sensitivity. Specifically, the compression ratio of the signals in the safe interval is set to 3:1, the warning interval is 2:1, and the dangerous interval remains 1:1. After the compression processing, the partition-compressed force feedback data is generated, effectively reducing the invalid data transmission bandwidth while retaining the important feature signals.

[0094] Step S25: Transmit the partition-compressed force feedback data through the 5G network with priority, so as to obtain dynamic force feedback control parameters, where the priority transmission specifically means that the force feedback signals in the dangerous interval have the highest transmission priority, followed by the warning interval, and the lowest for the safe interval.

[0095] In the embodiment of the present invention, the partition-compressed force feedback data is transmitted through the 5G network with priority. The force feedback signals in the dangerous interval are set with the highest transmission priority, and their delay is ensured to be less than 1 ms through QoS (Quality of Service) parameters; the warning interval signals are set with medium priority, allowing a delay within 10 ms; the safe interval signals have the lowest transmission priority, and the delay can reach 50 ms. Through the network queuing management mechanism, the data traffic is dynamically adjusted during the transmission process to ensure that the signals in the key areas are transmitted in time, and finally the dynamic force feedback control parameters are generated, which are used for the surgical robot to adjust the operation force feedback strategy in real time, improving the surgical safety and accuracy.

[0096] The present invention measures the tissue hardness parameters of the surgical site, calculates the upper limit value of the safe puncture force according to an adaptive algorithm, and divides the threshold intervals of the safe interval, warning interval, and danger interval, so as to generate preliminary force feedback threshold data. This process ensures that the setting of force feedback can accurately adapt to the physiological characteristics of the surgical site, and avoids damage to tissues caused by improper threshold setting. The threshold division enables the operator to have an intuitive understanding of the risk level during the operation and provides a clear reference for force feedback. Dynamically calculating the preliminary force feedback threshold data in combination with the baseline data of safe surgical operation, the generated force feedback threshold range data can adapt to the requirements of real-time changing surgical scenarios. This dynamic adjustment method effectively responds to the local differences in tissue hardness and the influence of patient individual characteristics during the operation, ensuring the reliability and real-time nature of the force feedback threshold. Dividing the force feedback signal into a safe interval, warning interval, and danger interval according to the threshold makes the force feedback feature space data of the surgical operation more orderly, facilitating subsequent feedback optimization processing. This zoning method enables the operator to quickly identify and adjust the operating force by clearly defining the risk levels, avoiding entering high-risk areas. Adaptive compression processing of the force feedback signals in different intervals can reduce the redundancy of data transmission while maintaining the key characteristics of the signals. This compression method ensures the integrity of the signals in the danger interval and appropriately compresses the signals in the safe interval, optimizing the efficiency of signal transmission. Transmitting the partition-compressed force feedback data through a 5G network and dividing the priorities according to the importance of the signals (the danger interval is the highest, the warning interval is the second, and the safe interval is the lowest). This transmission strategy ensures that key signals can reach the operator first, minimizing delays to the greatest extent and enhancing the safety and real-time nature of the operation.

[0097] Preferably, step S21 includes the following steps:

[0098] Obtain the tissue hardness parameters of the surgical site; calculate the upper limit value of the safe puncture force based on the adaptive algorithm according to the tissue hardness parameters, so as to obtain the safe puncture force value; set 80% of the safe puncture force value as the upper threshold of the safe interval, set the range of 80%-120% of the safe puncture force value as the threshold interval of the warning interval, and set 120% of the safe puncture force value as the lower threshold of the danger interval, so as to obtain the preliminary force feedback threshold data.

[0099] In the embodiments of the present invention, a hardness measuring instrument (e.g., a hardness probe based on a piezoelectric sensor) is first used to measure the tissue hardness of the surgical site. Specifically, the operator contacts the target tissue with the probe and applies a certain pressure, and the tissue hardness value of this site is obtained through the instrument. Taking the liver as an example, the measurement data may be between 10 and 150 kPa, where hard tissue may be measured at 150 kPa and soft tissue is 10 - 50 kPa. The hardness data can be recorded by the sensor and fed back to the system in real time to ensure the accuracy and timeliness of the measurement results, providing a basis for subsequent calculations. Based on the obtained tissue hardness data, the upper limit value of the puncture force is calculated through an adaptive algorithm (such as a model based on artificial neural network or support vector regression). This algorithm takes into account the influence of the change in hardness on the puncture force and automatically adjusts the recommended range of the puncture force for different hardness values based on historical data and empirical models. For example, assuming that for a tissue with a hardness of 50 kPa, the upper limit value of the safe puncture force calculated by the adaptive algorithm is 1.5 N, and if the hardness is 100 kPa, the upper limit value of the safe puncture force is calculated as 3.0 N. This process ensures the precise control of the puncture force and avoids tissue damage caused by excessive or insufficient puncture force. According to the safe puncture force value calculated in the second step, different force feedback threshold intervals are set. Specifically, assuming that the calculated safe puncture force value is 2.0 N, 80% of it (i.e., 1.6 N) is set as the upper threshold of the safe interval; the 80% - 120% of the safe puncture force value (1.6 N to 2.4 N) is set as the threshold interval of the warning interval; 120% of the safe puncture force value (i.e., 2.4 N) is set as the lower threshold of the dangerous interval. This process ensures that in actual surgical operations, the puncture force value is compared with the interval threshold in real time, providing real-time feedback for the surgical operation and guiding the operator to adjust the force, preventing tissue damage or puncture failure caused by improper operation.

[0100] By measuring the tissue hardness parameters at the surgical site, the present invention can accurately understand the actual physical properties of the tissue (such as elasticity, stiffness, etc.). This provides a data basis for subsequent force feedback control and helps to quantify the tissue's response to force. The tissue hardness may vary greatly among different patients. Obtaining this parameter can adjust the surgical strategy according to individual differences, achieve personalized surgical operations, and thus reduce the damage to the tissue caused by improper operation. The adaptive algorithm can dynamically calculate the upper limit value of the safe puncture force based on the tissue hardness data, making the setting of the puncture force more in line with the actual needs of the surgical site and ensuring that tissue damage or operation failure will not occur during the operation due to excessive or insufficient force. The use of the adaptive algorithm can dynamically adjust the puncture force value according to the real-time obtained hardness data, thereby improving the flexibility and safety of the operation and reducing the limitations of manual setting. The automated calculation process reduces the subjective judgment of the surgical staff on the puncture force, enhances the accuracy of puncture force control, and thus improves the controllability and precision of the operation. Taking 80% of the puncture force as the upper threshold of the safe range is a reasonable choice based on the actual tissue response, ensuring that the puncture force is within the safe range in most cases. This threshold can effectively avoid excessive puncture force during the operation and reduce the surgical risk. Setting a reasonable upper threshold of the safe range can help the surgical staff avoid tissue damage caused by excessive puncture force and at the same time provide sufficient operating space to ensure the smooth progress of the surgical operation. Setting the warning range to 80%-120% of the safe puncture force provides a certain warning space for the surgical operation. If the operating force exceeds the safe range but has not reached the dangerous level, the system can remind the operator in real time to avoid further loss of control of the surgical force. The setting of the warning range allows the surgical staff to flexibly respond to emergencies, timely adjust the puncture force, avoid unnecessary pressure on the tissue, and at the same time give the operator time and space for adjustment. Setting 120% as the lower threshold of the dangerous range can ensure that once the puncture force exceeds this range, the system will immediately issue a strong warning to prompt the operator to immediately adjust the operating force. At this time, continuing the operation may cause irreversible damage to the tissue, so timely warning is crucial. By setting a clear dangerous range, it can effectively avoid the puncture force reaching the damage threshold and ensure that the patient's tissue will not suffer excessive damage during the operation, guaranteeing the safety of the operation. By dividing the safe puncture force value into three different ranges (safe range, warning range, dangerous range), a multi-level risk assessment mechanism is formed, making the force feedback signal more hierarchical. The surgical staff can judge the safety and risk of the current operation according to the threshold information and timely adjust the operation strategy. The preliminary force feedback threshold data provides a clear operation boundary for the surgical process, helping the doctor to avoid excessive or insufficient force during the operation, thereby improving the accuracy and safety of the operation. Through the division of the threshold, the operator can monitor the force change during the operation in real time, make a quick response, ensure that the operation remains within the safe range, and effectively avoid potential risks.

[0101] Preferably, step S22 includes the following steps:

[0102] Step S221: Conduct probability statistics based on the baseline data of surgical safety operations, and perform confidence interval estimation on the preliminary force feedback threshold data through the multi-dimensional random variable analysis method to establish probability density function data including threshold uncertainty;

[0103] In the embodiment of the present invention, based on the baseline data of surgical safety operations, the statistical distribution analysis of the preliminary force feedback threshold data is performed by using the probability statistics method. The multi-dimensional random variable analysis method is adopted, and the magnitude of force, spatial direction and time sequence change are used as random variables, and the mean value, variance and joint probability distribution of each dimension are calculated respectively. Taking a certain surgical scenario as an example, the mean value of the magnitude of force dimension is 2.0N, and the variance is 0.2N 2 , and based on this, the confidence interval is calculated and the probability density function data including threshold uncertainty is established by using the Gaussian distribution hypothesis. The probability density function describes the threshold uncertainty distribution in different intervals and provides a mathematical basis for subsequent correction.

[0104] Step S222: Perform Bayesian inference on the probability density function data, and dynamically adjust the prior probability distribution of the force feedback threshold based on the surgical type, patient physiological characteristics and surgical site, so as to obtain the threshold range data after conditional probability correction;

[0105] In the embodiment of the present invention, the probability density function data obtained in step S221 is input into the Bayesian inference model, and the prior probability distribution of the force feedback threshold is dynamically adjusted based on the surgical type, patient physiological characteristics and surgical site. For example, for craniotomy (lower force threshold) and laparoscopic surgery (higher force threshold), the initial prior probability distributions are set respectively. Combining the patient's weight, age and the hardness parameter of the surgical site, the prior probability is corrected to obtain the threshold range data after conditional probability correction. For example, the upper threshold range after correction for craniotomy is 1.2 - 1.6N, while for laparoscopic surgery it is 2.5 - 3.0N.

[0106] Step S223: Perform multi-dimensional feature coupling analysis on the threshold range data to generate threshold dynamic correction data, where the multi-dimensional features include the magnitude of force dimension, spatial direction dimension, time sequence change dimension, scene adaptability dimension and risk assessment dimension;

[0107] In an embodiment of the present invention, multi-dimensional feature coupling analysis is performed based on the threshold range data in step S222. Using the mutual information gain analysis method, dimensional features such as force magnitude, spatial direction, temporal variation, scene adaptability, and risk assessment are combined to calculate the contribution degree of each dimension to the threshold correction. For example, the contribution degree of the scene adaptability dimension to the dynamic adjustment of the threshold is 30%, and that of the risk assessment dimension is 25%. Through feature-weighted linear combination, dynamic threshold correction data is generated. The correction data is adjusted in real time under different surgical scenarios. For example, when the risk assessment shows that the operation deviates from the safe range, the sensitivity of the warning interval is dynamically increased.

[0108] Step S224: Perform multi-scale feature extraction and adaptive reconstruction on the dynamic threshold correction data to eliminate noise interference in the threshold calculation process, thereby obtaining accurate threshold range data;

[0109] In an embodiment of the present invention, multi-scale feature extraction and adaptive reconstruction are performed on the dynamic threshold correction data generated in step S223. The specific operations include performing wavelet transform on the correction data to decompose it into feature signals of different scales; using the empirical mode decomposition (EMD) method to extract the main feature components and filtering the high-frequency noise components at the same time. Taking the data of a certain operation as an example, the main frequency feature is extracted at the third layer scale after wavelet decomposition, and the noise-free accurate threshold range data is reconstructed. This method effectively reduces the threshold deviation introduced by measurement equipment errors or environmental interference.

[0110] Step S225: Perform incremental correction on the accurate threshold range data and perform closed-loop self-correction based on the real-time feedback during the surgical process, thereby generating force feedback threshold range data with dynamic adaptive capabilities.

[0111] In an embodiment of the present invention, incremental correction is performed on the accurate threshold range data obtained in step S224, and closed-loop self-correction is performed in combination with the real-time feedback during the surgical process. Specifically, the system monitors the difference between the operation force feedback and the patient tissue response in real time, and dynamically adjusts the threshold range through an incremental algorithm. Taking the real-time feedback as an example, when it is detected that the actual operation force exceeds the upper limit of the warning interval, the system will automatically narrow the safe interval and increase the sensitivity of the dangerous interval. Finally, force feedback threshold range data with dynamic adaptive capabilities is generated. For example, in a tissue environment with changing hardness, the dynamic threshold range can be adjusted to 1.0 - 1.5 N (safe), 1.5 - 2.0 N (warning), and above 2.0 N (dangerous).

[0112] The present invention estimates the confidence interval of force feedback threshold data through a multi-dimensional random variable analysis method, which can reveal the fluctuation range and uncertainty of force feedback during surgical operations. This helps to quantify the potential errors of the force feedback threshold and provide a scientific basis for the operating range during surgery. The establishment of the probability density function data provides theoretical support for subsequent threshold correction and dynamic adjustment, making the threshold setting more in line with the uncertainty in actual operations. By comprehensively considering multiple variables (such as force magnitude, direction, time, etc.) for analysis, it can effectively capture potential changes in complex operations and improve the adaptability and accuracy of the force feedback threshold. Bayesian inference can dynamically adjust the threshold range by combining factors such as surgical type, patient physiological characteristics, and surgical site. This method can be adaptively adjusted according to real-time patient data, making the force feedback threshold more personalized and accurate. The dynamic adjustment of the prior probability distribution ensures that the threshold correction is not only based on current data but also comprehensively considers historical experience and physiological differences, which helps to improve the precision and safety during surgery. This step enables the system to adjust the threshold according to different operating environments, the specific conditions of the patient, and the characteristics of the surgical site, thereby realizing intelligent and personalized force feedback control. Through multi-dimensional feature coupling analysis, multiple dimensions of force (such as force magnitude, spatial direction, temporal variation, etc.) are considered, as well as the adaptability and risk assessment of the surgical scenario. This can integrate information from each dimension and correct the threshold more accurately. Different surgical scenarios, patient physiological states, and operating conditions will all affect the force feedback threshold. Through the coupling analysis of multi-dimensional features, dynamic correction and optimization can be achieved in a complex and changing operating environment. By fusing multiple influencing factors for analysis, the errors caused by single-dimensional analysis can be eliminated, enhancing the precision of threshold dynamic correction and ensuring the safety during the surgical process. Multi-scale feature extraction and adaptive reconstruction help to eliminate noise generated by data fluctuations or external interference during the threshold calculation process, thus ensuring that the finally calculated threshold data is more accurate. Adaptive reconstruction can automatically optimize the threshold calculation process, remove irrelevant or redundant information, further improve the data quality, and ensure the reliability of the threshold. Through multi-scale processing, different levels of data information can be processed to ensure that the threshold accuracy of force feedback during the surgical process reaches the best, thereby improving the safety and effectiveness of surgical operations. The combination of incremental correction and closed-loop self-correction enables the threshold range to be dynamically adjusted according to real-time feedback during the surgical process. This process ensures that each step during the surgical process is within the optimal force feedback range, improving the precision of the surgery. As the surgery progresses, the adaptive ability of the force feedback threshold can be adjusted according to the actual situation, avoiding risks caused by changes in conditions during the surgical process. Through real-time feedback, the threshold can be continuously optimized to cope with different surgical processes. Incremental correction makes the adjustment of the threshold more refined and gradually optimized, avoiding the problem of operational inadaptability that may be caused by a large one-time adjustment.Through continuous correction, the system can handle complex situations during the operation, maintaining high flexibility and safety.

[0113] Preferably, step S25 includes the following steps:

[0114] Step S251: Establish a multi-level transmission priority policy including a danger interval, a warning interval, and a safety interval according to the partition compression force feedback data, where the force feedback signal in the danger interval is set with the highest transmission priority, the force feedback signal in the warning interval is set with the medium transmission priority, and the force feedback signal in the safety interval is set with the lowest transmission priority;

[0115] In the embodiment of the present invention, different intervals of the partition compression force feedback data are first identified, and the signal in the danger interval is marked as the highest priority, the signal in the warning interval is marked as the medium priority, and the signal in the safety interval is marked as the lowest priority. For example, during a surgical operation, the detected danger interval feedback signal may include a sudden high-intensity touch force, which is immediately set as the highest priority to ensure timely transmission; the warning interval feedback signal, such as the operating force gradually approaching the warning range, is marked as the medium priority; the safety interval signal, such as the normal contact force remaining within the normal range, is assigned the lowest priority. After establishing the multi-level transmission priority policy, identification tags are attached to the data packets of signals with different priorities for subsequent network transmission optimization.

[0116] Step S252: Dynamically allocate the network channel bandwidth resources according to the multi-level transmission priority data, so as to obtain the channel resource scheduling data;

[0117] In the embodiment of the present invention, according to the multi-level transmission priority data in step S251, the bandwidth resources of the network channel are allocated through a dynamic bandwidth allocation algorithm. The danger interval signal is allocated priority bandwidth resources to ensure the lowest latency and the highest data integrity; the warning interval signal is allocated the remaining bandwidth resources; the safety interval signal is allocated the bandwidth resources with the lowest priority. For example, in a 5G network environment, 20 MHz of bandwidth is allocated to the danger interval signal, 10 MHz to the warning interval signal, and 5 MHz to the safety interval signal. The dynamic allocation of bandwidth resources is completed by the network scheduling module to generate the channel resource scheduling data.

[0118] Step S253: Perform anti-interference coding processing on the partition compression force feedback data according to the channel resource scheduling data, so as to generate the channel-coded force feedback data;

[0119] In an embodiment of the present invention, data is scheduled according to channel resources, and anti-interference coding processing is performed on partitioned compression force feedback data. For example, low-density parity-check (LDPC) coding technology is used to perform error correction coding on signals in the danger zone to ensure reliable transmission in a high-noise environment; convolutional coding is used for signals in the warning zone to balance transmission reliability and bandwidth occupancy; for signals in the safe zone, simple Hamming coding is used to reduce system resource consumption. Taking the signals in the danger zone as an example, the original data is converted into channel-coded force feedback data with high anti-interference ability through an LDPC encoder.

[0120] Step S254: Perform real-time evaluation of the transmission quality of the channel-coded force feedback data based on the transmission success rate, so as to obtain network transmission quality evaluation data;

[0121] In an embodiment of the present invention, real-time evaluation of the transmission quality of the channel-coded force feedback data based on the transmission success rate is performed, and a sliding window mechanism is used to monitor the packet loss rate, delay, and bit error rate of each data packet. For example, during the transmission of signals in the danger zone, if the system detects that the average packet loss rate of the data packets within the sliding window is less than 0.01%, the delay is less than 10 milliseconds, and the bit error rate is less than 0.0001, the transmission quality is evaluated as "excellent"; if the delay of the signals in the warning zone is close to the threshold (50 milliseconds), the transmission quality is evaluated as "medium". By evaluating the network transmission quality in real time, network transmission quality evaluation data is generated and the result is fed back to the network scheduling module for adjustment.

[0122] Step S255: Perform transmission processing on the partitioned compression force feedback data based on multiplexing and intelligent routing according to the network transmission quality evaluation data, so as to generate dynamic force feedback control parameters.

[0123] In an embodiment of the present invention, transmission processing on the partitioned compression force feedback data based on multiplexing and intelligent routing is performed according to the network transmission quality evaluation data in step S254. Specifically, during the transmission of signals in the danger zone, the system preferentially selects a low-delay path and enables time-division multiplexing (TDM) technology to ensure real-time performance; for the transmission of signals in the warning zone, frequency-division multiplexing (FDM) technology is used to balance transmission efficiency and resource utilization rate; for signals in the safe zone, redundant routing strategies are used for transmission to improve reliability. For example, in a complex network environment, signals in the danger zone are transmitted in parallel through the main route and the backup route, further reducing the risk of transmission interruption. The finally generated dynamic force feedback control parameters are applied in real-time control to optimize the response speed and accuracy of the force feedback signal.

[0124] By setting different transmission priorities according to different intervals of force feedback (dangerous, warning, safe), the present invention can ensure that the most critical data (such as force feedback signals in the dangerous interval) is transmitted in a timely manner when the network bandwidth is limited. This strategy ensures that the system can give priority to processing high-risk signals in the face of network congestion or bandwidth limitations, thus guaranteeing surgical safety. The high-priority setting for signals in the dangerous interval can reduce the latency in high-risk operations, thereby more quickly reflecting operation changes, reducing possible errors, increasing the response speed, and ensuring that corresponding safety measures are taken in a timely manner. The multi-level priority transmission strategy effectively allocates network bandwidth resources, enabling important signals to be guaranteed and non-critical data to be transmitted later, thus improving the utilization rate of network resources and enhancing the overall system efficiency. The dynamic allocation of channel bandwidth according to multi-level transmission priorities helps to effectively manage and optimize network bandwidth resources in a real-time system. In the case of limited bandwidth, it can automatically adjust the resource allocation according to the transmission priority to ensure the smooth transmission of high-priority signals. The dynamic allocation of resources makes the network bandwidth no longer fixed but adjusted in real time according to actual needs, avoiding waste of network resources. The system can automatically adjust to adapt to the changing needs during the surgical process, improving the usage efficiency of the bandwidth. By dynamically allocating channel resources, it can ensure that the key signals during the surgery are not affected by insufficient bandwidth, thus guaranteeing the real-time nature and accuracy of surgical operations. In an environment of multipath propagation and noise interference, anti-interference coding processing can effectively improve the reliability of signals. By performing anti-interference coding on the partition-compressed force feedback data, it can ensure that the signals are not interfered by external factors during transmission and maintain the integrity of the data. Anti-interference coding can reduce the bit error rate and signal loss, ensuring the accuracy and high-quality transmission of force feedback data. Especially during the surgical process with high real-time requirements, it is crucial to ensure the correct transmission of feedback signals. Through anti-interference coding, the system can maintain a high transmission quality in an unstable network environment and reduce data loss or errors caused by network problems. The real-time evaluation based on the transmission success rate can dynamically monitor the network transmission quality, promptly detect problems in transmission, and make adjustments. For the real-time evaluation of transmission quality, remedial measures can be quickly taken when signal packet loss or errors occur to ensure the stable operation of the system. Through real-time transmission quality evaluation, the system can make decisions based on the current network conditions, such as adjusting the data transmission rate or re-scheduling resources, to ensure that data can be transmitted efficiently and accurately. Evaluating the transmission quality can continuously optimize the performance of the system under different network conditions, ensuring that critical data can be effectively transmitted in any network state and guaranteeing the reliability and stability of the system. The combination of multiplexing and intelligent routing helps to efficiently utilize network bandwidth resources. By distributing data streams to different paths, it can reduce network load, avoid bottlenecks in a single channel, and enhance the overall system transmission efficiency.The intelligent router can select the best transmission path according to the real-time network status, further improving the stability and efficiency of data transmission. Adjusting the transmission mode of force feedback data based on network quality assessment data can ensure that when the network environment changes, the system can automatically optimize the transmission scheme, guarantee the accurate transmission of force feedback data, and dynamically generate control parameters to maintain the real-time performance and reliability of the system. Through intelligent routing and multiplexing, data transmission congestion and delay can be avoided. Especially when high-priority signals (such as danger zone signals) need to be transmitted quickly, it ensures that they reach the target in the shortest time, improves the response speed, and enhances the real-time control ability of the system.

[0125] Preferably, step S3 includes the following steps:

[0126] Step S31: Extract operation intention feature data by extracting the operation intention features from the control input data collected by the force feedback handle.

[0127] In the embodiment of the present invention, feature extraction is first performed on the control input data collected by the force feedback handle in the surgical robot system. The collected data includes the moving speed, acceleration, force sensing data, etc. of the handle, and features are extracted through time-domain and frequency-domain analysis methods. For example, the data of each operation input of the handle will be collected by sensors and sent to the data preprocessing module to remove noise and outliers, and then the PCA (Principal Component Analysis) method is used to reduce the dimension of the data and extract the main features related to the operation intention, such as the force change and direction change of the handle, so as to obtain the operation intention feature data. These data serve as the basis for subsequent analysis and can reflect the specific operation intention of the operator.

[0128] Step S32: Perform safety evaluation according to the operation intention feature data and the dynamic force feedback control parameters, so as to generate operation safety level data.

[0129] In the embodiment of the present invention, safety evaluation is performed according to the operation intention feature data and the dynamic force feedback control parameters. First, the operation intention feature data is combined with the current dynamic force feedback control parameters, and machine learning algorithms (such as Support Vector Machine SVM or Random Forest) are used to classify and evaluate the data. Through training the model, the operation intention is associated with the safety level data to evaluate the safety of the current operation. For example, during the operation, when the operation intention feature shows that the operator is about to perform a high-intensity puncture, combined with the current force feedback control parameters, the system generates operation safety level data, such as "safe", "warning", "dangerous" levels, through the evaluation model, reflecting the potential risks of the operation.

[0130] Step S33: Perform risk stratification response processing on the operation safety level data, so as to obtain risk classification response data.

[0131] In the embodiment of the present invention, risk stratification response processing is performed on the operation safety level data. According to the operation safety level data generated in step S32, a risk stratification response algorithm is used to process data of different safety levels. For example, for operations at the "dangerous" level, the system will give priority to starting an emergency response mechanism for automatic adjustment or interruption; for operations at the "warning" level, the system can issue a reminder and require the operator to adjust the intensity; operations at the "safe" level will maintain normal operation. The fuzzy control theory is used to perform fuzzy matching and response adjustment on data of different levels to generate specific risk classification response data to ensure the real-time safety of surgical operations.

[0132] Step S34: According to the risk classification response data, perform real-time correction and constraint on the operation intention, so as to generate intention regulation data;

[0133] In the embodiment of the present invention, the operation intention is corrected and constrained in real time according to the risk classification response data. For operations detected at the "dangerous" or "warning" level, the system will perform real-time correction and constraint on the operation intention through a dynamic control algorithm. The specific method includes modifying the force feedback signal fed back by the handle to forcibly limit the operation force of the operator or change the operation direction. For example, when the system recognizes that the operator is about to perform a dangerous operation (such as excessive force puncture), the system adjusts the force perception provided by the force feedback handle to limit the hand force of the operator, causing it to reduce the force or stop the operation, thereby generating intention regulation data to ensure the safety of the operation.

[0134] Step S35: According to the intention regulation data, perform network transmission optimization processing, so as to generate real-time control instruction data;

[0135] In the embodiment of the present invention, network transmission optimization processing is performed according to the intention regulation data. In a surgical robot system, the intention regulation data needs to be transmitted to the control system in real time for processing. In order to ensure the timely transmission and efficient processing of operation instructions, an adaptive transmission strategy based on the network load condition is adopted. Through a network transmission optimization algorithm (such as congestion control technology based on the TCP / IP protocol or the priority scheduling mechanism of the 5G network), the control instruction data is optimized for transmission according to factors such as the current network bandwidth and delay. For critical high-risk operations, the system will give priority to ensuring high-priority transmission, reducing delay, and ensuring that the operation intention can be quickly responded to. The finally generated real-time control instruction data ensures the precise operation of the surgical robot and avoids risks caused by network delay or errors.

[0136] By extracting the feature of the control input data collected by the force feedback handle, the present invention can accurately capture the operation intention of the user. This step converts the physical input of the handle into operation intention feature data, enabling the system to understand and analyze the user's behavior and goals, and provide personalized and precise feedback. The extraction of operation intention features enables the system to perceive the user's operations in real time and make appropriate responses based on these features, enhancing the interactive perception between the user and the system and improving the intuitiveness and comfort of operation. The operation intention feature data provides an important input data basis for subsequent steps such as safety assessment, risk stratification response processing, and real-time correction, ensuring that the system can make accurate safety assessments and feedback controls based on the analysis of the user's intention. By combining the operation intention feature data and dynamic force feedback control parameters, the system can evaluate the safety of the operation in real time and avoid potential risks caused by improper operations or behaviors beyond the safe range. This helps to detect unsafe operations at an early stage, thereby reducing the probability of accidents. The dynamic force feedback control parameters can be adjusted in real time according to the current operation environment and force feedback situation, making the safety assessment more accurate and timely. In this way, the user's behavior can be continuously evaluated during the operation, and potential safety hazards can be detected and corrected in a timely manner. Conducting a safety assessment of the operation helps to ensure that the user always stays within the safe operation range when performing the operation, thereby enhancing the reliability and safety of the overall operation process. Especially in high-risk environments such as surgery and robot control, it can effectively prevent dangerous operations. By performing risk stratification response processing on the operation safety level data, the system can take different response measures according to different safety levels. For example, for high-risk operations, the system can issue warnings, restrict certain operations, or take preventive measures, while for low-risk operations, it can maintain the normal operation process. This hierarchical response can improve the accuracy and response ability of the system. The risk stratification response helps to identify high-risk behaviors in a timely manner and process them through preset response strategies to prevent accidents. At the same time, for low-risk behaviors, the response can be simplified to avoid excessive intervention and improve the efficiency of the system. This hierarchical response mechanism can be adjusted according to different operation safety levels to ensure that the system can adjust the response strategy according to real-time feedback in any situation, enhancing the flexibility and dynamic adaptability of the system. Real-time correction and constraint of the operation intention according to the risk classification response data can effectively prevent the user from deviating from the safe range when performing high-risk operations, automatically correct the user's operation behavior, and avoid dangerous or unsafe behaviors. For example, when the operation enters the dangerous area, the system can actively intervene or automatically adjust the operation intention to ensure that it always stays within the safe range. By actively regulating the user's operation intention, it can effectively prevent accidents caused by user misoperations or external interferences, enhancing the system's risk control ability, which is particularly important in high-risk environments (such as surgical robot operations).Real-time correction and constraint of operation intentions enable the system to actively intervene in user behavior and improve the system's control ability. This can better protect users and the operation environment, ensuring that tasks can be completed smoothly and safely. Optimizing network transmission according to the intention-regulated data can ensure the efficient transmission of control instructions in the network, reducing latency and bandwidth occupancy. This is particularly important in applications with high real-time requirements (such as remote control, robot operation, etc.), which can ensure the rapid response and execution of instructions. The network transmission optimization process can adjust the data transmission strategy according to the network status, ensuring the stable transmission of real-time control instructions in various network environments and reducing control failures or data losses caused by network fluctuations. Through the optimization of network transmission, real-time control instructions can reach the target device or system more quickly, thus ensuring the timeliness of control feedback. This is crucial for operations that require quick responses (such as precise operations during surgery, real-time control tasks), avoiding operation delays or interruptions.

[0137] The present invention also provides a remote surgical robot control system based on a force feedback handle for performing the above-mentioned remote surgical robot control method based on a force feedback handle. The remote surgical robot control system based on a force feedback handle includes:

[0138] A force feature acquisition and analysis module for collecting multi-dimensional force data of the end effector of the surgical robot, extracting the magnitude, direction, and timing characteristics of the contact force, and obtaining multi-dimensional force feature data; performing a surgical operation safety baseline analysis based on the multi-dimensional force feature data to generate surgical safety operation baseline data;

[0139] A dynamic force feedback threshold calculation module for calculating a dynamic force feedback threshold based on the surgical safety operation baseline data to obtain force feedback threshold range data; performing adaptive compression processing on different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters;

[0140] An operation intention and safety evaluation module for collecting the control input data of the operator in real time through the force feedback handle to obtain operation intention data; performing a safety evaluation on the operation intention data based on the dynamic force feedback control parameters to generate operation safety level data; performing multi-level response processing on the operation safety level data and optimizing network transmission to generate real-time control instruction data;

[0141] A closed-loop force feedback control module for planning the robot motion trajectory according to the real-time control instruction data to obtain robot execution trajectory data; updating the force feedback parameters in real time based on the robot execution trajectory data to obtain a closed-loop force feedback control strategy.

[0142] Therefore, in all aspects, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Thus, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0143] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A remote surgical robot control system based on a force feedback handle, characterized in that: The following steps are involved: The force feature acquisition and analysis module is used to collect multi-dimensional force data of the end effector of the surgical robot, and extract the size, direction and timing characteristics of the contact force to obtain multi-dimensional force feature data; Perform surgical operation safety baseline analysis based on multi-dimensional force characteristic data to generate surgical operation safety baseline data; A dynamic force feedback threshold calculation module is used to calculate the dynamic force feedback threshold according to the surgical safety operation baseline data to obtain the force feedback threshold range data; Adaptively compress different force feedback signals according to the force feedback threshold range data to obtain dynamic force feedback control parameters; The operation intention and safety assessment module is used to collect the operator's control input data in real time through the force feedback handle to obtain the operation intention data; Conduct safety assessment on operation intention data based on dynamic force feedback control parameters to generate operation safety level data; Perform multi-level response processing on operational safety level data and optimize network transmission to generate real-time control instruction data; A closed-loop force feedback control module is used to plan the robot motion trajectory according to the real-time control instruction data and obtain the robot execution trajectory data; The force feedback parameters are updated in real time based on the robot execution trajectory data, thereby obtaining a closed-loop force feedback control strategy.

2. The remote surgical robot control system based on the force feedback handle according to claim 1, characterized in that: The force characteristic acquisition and analysis module is used to perform the following steps: Step S11: collecting force data of the end effector of the surgical robot through a multi-dimensional force sensor array to obtain six-dimensional force data; collecting spatial position and attitude angle information of the end effector to obtain spatial attitude data; Step S12: merging the six-dimensional force data and the spatial posture data into multi-dimensional force data; Step S13: Calculate the magnitude of the resultant force and its three-dimensional components according to the multi-dimensional force data, and perform spatial moment decoupling processing to obtain contact force characteristic vector data; Step S14: performing time domain sampling and frequency domain analysis on the contact force characteristic vector data, thereby obtaining multi-dimensional force characteristic data; Step S15: Perform surgical operation safety baseline analysis based on the multi-dimensional force feature data to generate surgical operation safety baseline data.

3. The remote surgical robot control system based on the force feedback handle according to claim 2, characterized in that: Step S15 includes the following steps: Step S151: according to the multi-dimensional force feature data, the three dimensions of force magnitude, spatial direction and time series change are used as the basic dimensions of the feature space, so as to construct the surgical operation force feedback feature space data whose grid density increases with the increase of the force value, wherein the force magnitude adopts the Newton unit, the spatial direction adopts the angle system, and the time series change adopts the rate of change per unit time; Step S152: setting three levels of safety thresholds including a warning threshold, a danger threshold and an emergency threshold according to the data of each dimension of the surgical operation force feedback feature space data, thereby obtaining initial safety threshold data; Step S153: dynamically adjusting the threshold value of the initial safety threshold data based on the force characteristics of human tissue, and establishing a force mutation warning mechanism based on the time series change rate, thereby obtaining dynamic safety threshold data; Step S154: constructing a surgical scene adaptive baseline according to the dynamic safety threshold data, thereby obtaining scene adaptive baseline data, wherein the baseline construction includes force feature pattern recognition for different types of surgical actions, pattern matching for real-time surgical operations, and dynamic adjustment of scene adaptive safety baseline parameters according to the matching results; Step S155: calculating the degree of deviation between the current operation force characteristic and the safety baseline according to the scene adaptive baseline data and the surgical operation force feedback characteristic space data, and generating a real-time risk score based on the degree of deviation, thereby obtaining safety assessment result data; Step S156: Perform feature fusion processing based on the surgical operation force feedback feature space data, dynamic safety threshold data, scene adaptive baseline data and safety assessment result data to generate surgical safety operation baseline data.

4. The remote surgical robot control system based on the force feedback handle according to claim 3, characterized in that: Step S156 includes the following steps: Step S1561: normalizing and standard-deviation correcting the data of each dimension of the surgical operation force feedback feature space data, thereby obtaining standardized feature mapping data, wherein the normalization process uses a minimum-maximum linear transformation, and the standard-deviation correction is based on a kurtosis and skewness analysis of the data; Step S1562: performing cross-correlation analysis on the standardized feature mapping data, the dynamic safety threshold data and the scene adaptive baseline data, thereby obtaining feature subspace correlation data, wherein the cross-correlation analysis specifically calculates the mutual information gain rate between feature dimensions, identifies feature subspaces with significant correlation, and dynamically adjusts the weight allocation strategy of each feature dimension; Step S1563: Perform wavelet packet decomposition on the safety assessment result data to extract the multi-scale characteristic coefficients of the data, and adaptively reorganize the characteristic coefficients of each scale based on the empirical mode decomposition method to obtain the surgical safety operation baseline data reconstructed by inverse transformation.

5. The remote surgical robot control system based on the force feedback handle according to claim 1, characterized in that: The dynamic force feedback threshold calculation module is used to perform the following steps: Step S21: obtaining tissue hardness parameters of the surgical site, calculating the upper limit value of the safe puncture force based on the adaptive algorithm according to the tissue hardness parameters, and performing threshold division to obtain preliminary force feedback threshold data, wherein the preliminary force feedback threshold data includes an upper limit threshold of a safe interval, a threshold interval of a warning interval, and a lower limit threshold of a dangerous interval; Step S22: performing dynamic force feedback threshold calculation on the preliminary force feedback threshold data according to the surgical safety operation baseline data to obtain force feedback threshold range data; Step S23: performing force feedback signal partitioning processing on the surgical operation force feedback feature space data according to the force feedback threshold range data, dividing the data into a safe interval, a warning interval, and a dangerous interval, thereby obtaining force feedback partitioning data; Step S24: performing adaptive compression processing on force feedback signals in different intervals according to the force feedback partition data, thereby obtaining partition compressed force feedback data; Step S25: The partitioned compressed force feedback data is transmitted with priority through the 5G network to obtain dynamic force feedback control parameters, wherein the priority transmission specifically means that the force feedback signal in the danger zone has the highest transmission priority, followed by the warning zone, and the safety zone has the lowest priority.

6. The remote surgical robot control system based on the force feedback handle according to claim 5, characterized in that: Step S21 includes the following steps: Obtain tissue hardness parameters of the surgical site; calculate the upper limit of the safe puncture force based on an adaptive algorithm according to the tissue hardness parameters, thereby obtaining a safe puncture force value; set 80% of the safe puncture force value as the upper limit threshold of the safety interval, set 80%-120% of the safe puncture force value as the threshold interval of the warning interval, and set 120% of the safe puncture force value as the lower limit threshold of the danger interval, thereby obtaining preliminary force feedback threshold data.

7. The remote surgical robot control system based on the force feedback handle according to claim 5, characterized in that: Step S22 includes the following steps: Step S221: Probabilistic statistics are performed based on the surgical safety operation baseline data, and confidence interval estimation is performed on the preliminary force feedback threshold data by a multi-dimensional random variable analysis method to establish probability density function data including threshold uncertainty; Step S222: performing Bayesian inference on the probability density function data, and dynamically adjusting the prior probability distribution of the force feedback threshold value based on the type of surgery, the physiological characteristics of the patient, and the surgical site, so as to obtain the threshold range data after conditional probability correction; Step S223: performing coupling analysis on the multi-dimensional features of the threshold range data to generate threshold dynamic correction data, wherein the multi-dimensional features include the dimension of force size, the dimension of spatial direction, the dimension of temporal change, the dimension of scene adaptability, and the dimension of risk assessment; Step S224: performing multi-scale feature extraction and adaptive reconstruction on the threshold dynamic correction data to eliminate noise interference in the threshold calculation process, thereby obtaining accurate threshold range data; Step S225: incrementally correct the precise threshold range data, and perform closed-loop self-correction according to real-time feedback of the surgical process, thereby generating force feedback threshold range data with dynamic self-adaptation capability.

8. The remote surgical robot control system based on the force feedback handle according to claim 5, characterized in that: Step S25 includes the following steps: Step S251: establishing a multi-level transmission priority strategy including a danger zone, a warning zone and a safety zone according to the partition compression force feedback data, wherein the danger zone force feedback signal is set with the highest transmission priority, the warning zone force feedback signal is set with a medium transmission priority, and the safety zone force feedback signal is set with the lowest transmission priority; Step S252: dynamically allocating network channel bandwidth resources according to the multi-level transmission priority data, thereby obtaining channel resource scheduling data; Step S253: performing anti-interference coding processing on the partitioned compressed force feedback data according to the channel resource scheduling data, thereby generating channel coded force feedback data; Step S254: performing a real-time transmission quality evaluation based on the transmission success rate on the channel coding force feedback data, thereby obtaining network transmission quality evaluation data; Step S255: performing transmission processing based on multiplexing and intelligent routing on the partitioned compressed force feedback data according to the network transmission quality evaluation data, thereby generating dynamic force feedback control parameters.

9. The remote surgical robot control system based on the force feedback handle according to claim 1, characterized in that: The operation intention and safety assessment module is used to perform the following steps: Step S31: extracting operation intention features according to the control input data collected by the force feedback handle, thereby obtaining operation intention feature data; Step S32: performing safety assessment according to the operation intention characteristic data and the dynamic force feedback control parameters, thereby generating operation safety level data; Step S33: performing risk stratification response processing on the operation safety level data, thereby obtaining risk grading response data; Step S34: modifying and constraining the operation intention in real time according to the risk classification response data, thereby generating intention control data; Step S35: Perform network transmission optimization processing according to the intended control data, thereby generating real-time control instruction data.

Images