Remote operation control system and control method based on Internet of Things technology

Abstract

The invention relates to the technical field of Internet of Things, in particular to a remote operation control system and method based on the Internet of Things technology. The method comprises the following steps: acquiring multi-source sensing data of a surgical system; performing communication quality evaluation based on network delay, data integrity and signal strength on the multi-source sensing data to obtain initial system state data; setting an operation parameter threshold according to the initial system state data, and performing system parameter calibration on the multi-source sensing data to obtain system calibration data; performing force feedback characteristic analysis on the system calibration data to generate six-dimensional force control characteristic data; and performing force feedback space-time distribution calculation according to the six-dimensional force control characteristic data, and generating force feedback statistical characteristic data. Through six-dimensional force control characteristic analysis and force feedback space-time distribution calculation, the system can recognize and compensate force sense deviation in operation in real time, and more accurate operation control experience is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a remote surgery control system and a control method based on the Internet of Things technology. Background Art

[0002] Remote surgical control relies on high-speed and stable network technology to ensure real-time and reliable data transmission during surgical operations. This usually involves 4G / 5G networks as the main communication method, as well as fiber optic connections. The network layer includes network access modules, data transmission modules, and network security modules. The user layer is responsible for functions such as surgical operations, remote collaboration, and data storage and analysis. This includes surgical control modules, remote collaboration modules, data storage and analysis modules, etc. The surgical control module receives instructions sent by the surgical control terminal, controls the movement of the surgical robot, and realizes remote surgical operations.

[0003] Traditional remote surgery control methods based on IoT technology often have the following problems: Remote surgery relies on high-speed, stable network communications. In the past, network delays and packet loss often led to slower response times during surgery, affecting the accuracy of the operation and the surgeon's feel. The delicate operation of surgery requires surgeons to accurately sense the applied force. In traditional remote operations, force feedback is usually not timely or accurate enough, resulting in increased surgical risks. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a remote surgery control system and control method based on Internet of Things technology to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a remote surgery control method based on Internet of Things technology includes the following steps:

[0006] Step S1: Acquire multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system state data; set operating parameter thresholds according to the initial system state data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data;

[0007] Step S2: performing force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; performing force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; performing difference analysis on the six-dimensional force control feature data to generate force feedback deviation data;

[0008] Step S3: performing motion delay identification according to the force feedback deviation data, and performing real-time delay evaluation to generate delay feature evaluation data; performing data noise reduction processing on the system calibration data, and performing delay compensation processing through the delay feature evaluation data to generate dynamic compensation data;

[0009] Step S4: performing operation mode recognition on the dynamic compensation data through force feedback statistical feature data to generate operation type feature data; performing operation risk classification on the operation type feature data to obtain risk level data; performing safety strategy matching on the risk level data to generate surgical safety control data;

[0010] Step S5: construct a remote surgery control model based on the dynamic compensation data and the surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction and analysis to generate robotic arm control data.

[0011] The present invention obtains multi-source sensor data of the surgical system and evaluates network delay, data integrity and signal strength, so that the system can understand its initial communication state in real time. Such evaluation not only improves the stability of network connection, but also ensures the reliability of data transmission. By setting the operating parameter threshold and calibrating the system parameters of the multi-source sensor data, the surgical system is optimized and calibrated before operation, which improves the overall stability and response speed of the system during operation. This can effectively reduce network and communication problems that occur during operation, laying a foundation for the smooth progress of remote surgery. Force feedback feature analysis is performed on the calibration data and six-dimensional force control feature data is generated to ensure that the surgical system can capture real and subtle tactile feedback, thereby providing a more realistic surgical experience for the operating doctor. Through the calculation of the spatiotemporal distribution of force feedback and the analysis of deviation data, the surgical system can accurately locate the abnormality in the feedback, thereby reducing the risk of misoperation. These processes improve the accuracy of the operation, enable doctors to better perceive and manipulate surgical instruments in a remote environment, and improve the accuracy and safety of surgical operations. By performing motion delay recognition and real-time delay evaluation based on force feedback deviation data, the surgical system can identify and analyze the operation lag caused by network delay. The response time of the surgical system is optimized by performing data noise reduction and delay compensation on the system calibration data. This step effectively reduces the delay during the operation, improves the real-time and smoothness of the operation, ensures the precise operation of the operation, and reduces the errors and uncertainties caused by network delays. Operation mode recognition and risk classification provide an additional layer of security for the remote surgery system. Through the analysis of force feedback statistical feature data and dynamic compensation data, the system can quickly identify the current operation type and classify the risk according to the operation type, so as to effectively respond when potential risks are detected. The generated surgical safety control data enables the system to adjust and respond to emergencies in a timely manner by matching the preset safety strategy, reducing possible accidents and risks during the operation and enhancing the safety of remote surgery. The remote surgery control model is constructed using dynamic compensation data and surgical safety control data, so that the system has the ability to predict and optimize the surgical operation trajectory. Real-time operation trajectory prediction analysis not only improves the control stability of surgical instruments, but also enables the movement of the robotic arm to be accurately adjusted according to the surgical situation, ensuring the continuity and high precision of the action. The generation and application of robotic arm control data significantly improves the execution efficiency and success rate of the operation, so that remote surgery can be carried out stably and efficiently in different scenarios.

[0012] The present invention further provides a remote surgery control system based on the Internet of Things technology, which is used to execute the above-mentioned remote surgery control method based on the Internet of Things technology. The remote surgery control system based on the Internet of Things technology includes:

[0013] A multi-source data acquisition and calibration module is used to obtain multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system status data; set operating parameter thresholds according to the initial system status data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data;

[0014] The force feedback feature extraction module is used to perform force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; perform force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data;

[0015] The delay compensation processing module is used to identify motion delay according to the force feedback deviation data, and to perform real-time delay evaluation to generate delay feature evaluation data; perform data noise reduction processing on the system calibration data, and perform delay compensation processing through the delay feature evaluation data to generate dynamic compensation data;

[0016] The operation safety assessment module is used to identify the operation mode of the dynamic compensation data through the force feedback statistical feature data to generate the operation type feature data; to classify the operation risk of the operation type feature data to obtain the risk level data; to match the safety strategy of the risk level data to generate the operation safety control data;

[0017] The trajectory prediction control module is used to build a remote surgery control model based on dynamic compensation data and surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction analysis and generate robotic arm control data.

[0018] The present invention ensures that the collected data has high credibility and reduces signal interference and data loss through communication quality evaluation based on network delay, data integrity and signal strength. Preliminary evaluation of the operating status of the system helps to timely discover system problems such as delay, data loss or signal instability, and provide accurate starting data for subsequent operations. By calibrating multi-source sensor data, the data of each sensor can work together, reduce the error between different sensors, and ensure the accuracy in the subsequent data processing process. By generating six-dimensional force control feature data (including force size, direction, position, etc.), it can accurately perceive the force changes in surgical operations and provide data support for subsequent compensation and control. Force feedback spatiotemporal distribution calculation can analyze the distribution pattern of force, timely reflect the change trend of force in operation, and enable the system to dynamically adjust the operation. Through difference analysis, force feedback deviation data is identified and generated, which helps to eliminate the deviation and inaccuracy that may occur during the operation and ensure the high precision of the robot arm operation. Through delay identification and evaluation, the source and impact of the delay can be clearly understood, compensation can be made in time, the interference of delay on surgical operations can be reduced, and the real-time operation can be improved. Combined with dynamic compensation data, the motion trajectory of the robot arm can be dynamically adjusted so that it can respond to operation requirements more accurately, thereby improving operation accuracy and safety. Noise reduction processing can effectively remove noise from the signal, ensure the accuracy of compensation data, and enable the system to respond quickly and stably to different operation environments and requirements. By identifying the operation mode, the risk level of the current operation can be evaluated in real time, potential problems can be discovered in advance, and timely adjustments can be made. By risk grading the operation type, high-risk operations can be identified and more stringent safety strategies can be configured for them to ensure the smooth progress of the operation. The matching of safety strategies can select appropriate control data according to different risk levels, further enhance the safety of the operation process, and reduce the risks caused by improper operation. The establishment of a remote surgery control model enables the system to accurately predict the motion trajectory of the robot arm and avoid unnecessary surgical risks caused by prediction errors. Through real-time prediction and control, precise operation of the robot arm can be achieved, avoiding unstable factors caused by manual intervention, which is particularly important in remote surgery. Trajectory prediction and adjustment combined with safety control data ensure that each step of the operation is within a safe range, which helps prevent operational errors from causing harm to patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0020] Figure 1 This is a schematic diagram of the steps of the remote surgery control method based on the Internet of Things technology of the present invention;

[0021] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0022] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0023] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying 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 implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0025] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. 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.

[0026] To achieve this, please refer to Figures 1 to 3 The present invention provides a remote surgery control method based on Internet of Things technology, the method comprising the following steps:

[0027] Step S1: Acquire multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system state data; set operating parameter thresholds according to the initial system state data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data;

[0028] Step S2: performing force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; performing force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; performing difference analysis on the six-dimensional force control feature data to generate force feedback deviation data;

[0029] Step S3: performing motion delay identification according to the force feedback deviation data, and performing real-time delay evaluation to generate delay feature evaluation data; performing data noise reduction processing on the system calibration data, and performing delay compensation processing through the delay feature evaluation data to generate dynamic compensation data;

[0030] Step S4: performing operation mode recognition on the dynamic compensation data through force feedback statistical feature data to generate operation type feature data; performing operation risk classification on the operation type feature data to obtain risk level data; performing safety strategy matching on the risk level data to generate surgical safety control data;

[0031] Step S5: construct a remote surgery control model based on the dynamic compensation data and the surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction and analysis to generate robotic arm control data.

[0032] In the embodiment of the present invention, reference Figure 1 As shown, it is a schematic diagram of the steps of a remote surgery control method based on the Internet of Things technology of the present invention. In this example, the remote surgery control method based on the Internet of Things technology includes the following steps:

[0033] Step S1: Acquire multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system state data; set operating parameter thresholds according to the initial system state data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data;

[0034] In the telesurgery system of the embodiment of the present invention, multi-source sensors (including six-dimensional force sensors, position encoders, end effector posture sensors and network communication modules) are used to obtain sensor data, including signal strength, network delay, packet loss rate, six-dimensional force data, torque data and position information of each joint of the robotic arm. The sensor data is integrated into a multi-source sensor data matrix through a data fusion algorithm, and the network delay is evaluated based on the average delay time and delay jitter. The data packet loss rate, retransmission rate and data synchronization performance are calculated through data integrity check to generate an integrity evaluation matrix, and the signal quality vector is calculated based on the signal-to-noise ratio, bandwidth utilization and signal stability. Based on the above evaluation results, a weighted scoring algorithm is used to calculate the comprehensive score of the system state to obtain the initial system state data. According to the initial state data, the network delay tolerance threshold is set to 150ms, the data packet loss rate tolerance threshold is set to 0.5%, and the signal strength lower limit is set to -70dBm, and the multi-source sensor data is calibrated based on the iterative weighted least squares method to obtain the calibrated multi-source data.

[0035] Step S2: performing force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; performing force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; performing difference analysis on the six-dimensional force control feature data to generate force feedback deviation data;

[0036] The embodiment of the present invention analyzes the multi-source data after calibration in step S1. First, the time domain features of the initial force data matrix are extracted, and the peak force is calculated to be 20N, the average force is 10N, the torque peak is 5Nm, and the torque mean is 2Nm, thereby generating an initial force feature vector; then, the force data is subjected to frequency domain feature analysis based on Fourier transform, and the main frequency component is extracted to be 1Hz, and the energy distribution is mainly concentrated in the frequency band of 0.5Hz-2Hz, thereby generating a frequency domain feature matrix. The time domain and frequency domain features are fused, and spatial mapping is performed in combination with the end-of-arm posture data to generate six-dimensional force control feature data; further, by constructing a spatiotemporal distribution model, the spatial distribution density of the calculated force is 50N / m 2 , the time series change rate is 0.1N / ms, and the distribution characteristics of the torque are extracted to generate force feedback statistical feature data; finally, a historical comparative analysis is performed on the six-dimensional force control feature data, and the mean deviation of the calculated force is 5%, and the direction angle deviation is 2° to generate force feedback deviation data.

[0037] Step S3: performing motion delay identification according to the force feedback deviation data, and performing real-time delay evaluation to generate delay feature evaluation data; performing data noise reduction processing on the system calibration data, and performing delay compensation processing through the delay feature evaluation data to generate dynamic compensation data;

[0038] The embodiment of the present invention performs a delay feature analysis on the force feedback deviation data generated in step S2, extracts the average value of the operation delay as 120ms and the maximum delay as 200ms, and generates an initial delay feature vector. Subsequently, the dynamic delay evaluation model is used to analyze the fluctuation range (±10ms) and trend (linear increase) of the delay, and a delay dynamic evaluation matrix is ​​generated; the impact of the delay on the system operation accuracy and safety is identified according to the evaluation matrix, and the delay feature evaluation data is calculated. The system calibration data is denoised by wavelet transform, and the delay compensation is performed in combination with the delay feature evaluation data after removing the high-frequency noise. The optimization model is used to calculate the optimal compensation parameters in different scenarios, including a time compensation coefficient of 0.8 and a spatial compensation displacement of 0.05mm, and finally the dynamic compensation data is generated.

[0039] Step S4: performing operation mode recognition on the dynamic compensation data through force feedback statistical feature data to generate operation type feature data; performing operation risk classification on the operation type feature data to obtain risk level data; performing safety strategy matching on the risk level data to generate surgical safety control data;

[0040] The embodiment of the present invention uses an operation mode classification algorithm to analyze the dynamic compensation data and force feedback statistical feature data generated in step S3, identifies the operation type as "fine operation", and extracts the operation intensity parameter as 8N. The operation feature data is combined with the risk assessment model, and the risk level of the current operation is determined to be medium according to the risk level classification rules. Subsequently, according to the preset surgical safety strategy library, the safety strategy corresponding to the medium risk level is matched, and the safety control parameters are set, including the maximum allowable force deviation of 2N and the maximum allowable posture error of 0.1mm, thereby generating surgical safety control data.

[0041] Step S5: construct a remote surgery control model based on the dynamic compensation data and the surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction and analysis to generate robotic arm control data.

[0042] The embodiment of the present invention constructs a remote surgery control model based on the dynamic compensation data of step S3 and the surgical safety control data of step S4, which includes a motion trajectory prediction sub-model and a safety control sub-model. First, the motion trajectory prediction sub-model is used to predict the end position of the robotic arm at the next moment, and the prediction result is the target position (x, y, z) = (100.05, 50.02, 75.10) mm; then, the trajectory safety constraint verification is performed in combination with the surgical safety control data to ensure that the predicted trajectory meets the accuracy and safety requirements. The predicted trajectory vector is optimized and the joint control parameters are calculated, where the joint angle is 45°, the angular velocity is 1.2 rad / s, and the angular acceleration is 0.8 rad / s 2, generate the joint control matrix. Finally, the stability and reliability of the joint control matrix are verified through the dynamic model, and the verified data is converted into a robot arm control instruction sequence, and the robot arm operation parameters are updated in real time to complete remote surgery control.

[0043] Preferably, step S1 comprises the following steps:

[0044] Step S11: Acquire signal quality data of network transmission, including signal strength, network delay, data packet loss rate and bandwidth occupancy rate;

[0045] Step S12: collecting the initial force data matrix of the six-dimensional force sensor at the end of the robotic arm, including the force data and torque data of the six-dimensional force sensor;

[0046] Step S13: Collecting information of position encoders of each joint of the robot arm and position information of the end effector, and performing a coordinate system conversion to obtain position matrix data; merging signal quality data, initial force data matrix and position matrix data into multi-source sensor data;

[0047] Step S14: performing network delay evaluation on multi-source sensor data based on average delay time calculation and delay jitter analysis, thereby obtaining delay feature vector data;

[0048] Step S15: performing data integrity assessment on the multi-source sensor data to obtain an integrity assessment matrix, wherein the data integrity assessment includes a data packet sequence integrity check, data packet loss rate and retransmission rate calculation, and data synchronization performance assessment;

[0049] Step S16: performing signal strength evaluation of the signal-to-noise ratio, bandwidth utilization, and signal stability index on the multi-source sensor data, thereby obtaining signal quality vector data; performing system state comprehensive score calculation based on the delay feature vector data, the integrity evaluation matrix, and the signal quality vector data, thereby obtaining initial system state data;

[0050] Step S17: Calculate the system state comprehensive score according to the delay feature vector data, the integrity assessment matrix and the signal quality vector data, so as to obtain the initial system state data;

[0051] Step S18: setting operating parameter thresholds according to the initial system state data, and performing system parameter calibration on the multi-source sensor data to obtain system calibration data, wherein the operating parameter thresholds include a network delay tolerance threshold, data integrity requirements, and a signal quality lower limit.

[0052] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0053] Step S11: Acquire signal quality data of network transmission, including signal strength, network delay, data packet loss rate and bandwidth occupancy rate;

[0054] The embodiment of the present invention obtains the signal quality data of network transmission in real time through the network communication module, and uses a dedicated communication monitoring device to record the signal strength (e.g. -65dBm), network delay (e.g. average 120ms, maximum 150ms), packet loss rate (e.g. 0.3%) and bandwidth occupancy rate (e.g. 75%). The signal strength is measured by RSSI (received signal strength indication), the network delay is measured by the Ping tool, the packet loss rate is calculated by packet comparison, and the bandwidth occupancy rate is obtained by network traffic statistics analysis, and finally a signal quality data vector is generated.

[0055] Step S12: collecting the initial force data matrix of the six-dimensional force sensor at the end of the robotic arm, including the force data and torque data of the six-dimensional force sensor;

[0056] The embodiment of the present invention uses a six-dimensional force sensor installed at the end of the mechanical arm to collect the initial force data matrix, including the three-axis force data (e.g., Fx=10N, Fy=8N, Fz=12N) and three-axis torque data (e.g., Mx=1.2Nm, My=0.8Nm, Mz=1.0Nm) measured by the force sensor. The sampling frequency of the collected data is set to 100Hz, and the signal preprocessing module is used for filtering to eliminate high-frequency noise and ensure the accuracy and stability of the force data and torque data.

[0057] Step S13: Collecting information of position encoders of each joint of the robot arm and position information of the end effector, and performing a coordinate system conversion to obtain position matrix data; merging signal quality data, initial force data matrix and position matrix data into multi-source sensor data;

[0058] In the embodiment of the present invention, the encoder installed on each joint of the robot arm obtains real-time position data, for example, the joint angle is [30°, 45°, 60°], and the position and direction information is collected through the end effector posture sensor, for example, the posture matrix is ​​[(x, y, z) = (100mm, 50mm, 75mm), (roll, pitch, yaw) = (5°, 3°, 10°)]. The joint and end data are converted using the same coordinate system conversion algorithm to ensure data consistency and generate a unified position information matrix. The signal quality data, the initial force data matrix, and the position information matrix are synthesized into multi-source sensor data through a matrix splicing method.

[0059] Step S14: performing network delay evaluation on multi-source sensor data based on average delay time calculation and delay jitter analysis, thereby obtaining delay feature vector data;

[0060] The embodiment of the present invention analyzes the network delay data in the multi-source sensor data and uses the average delay calculation formula

[0061] The average delay is 120ms, and the delay jitter calculation formula is used The calculated jitter value is 10ms, and the delay feature vector data T is generated. vector =[120ms,10ms], used for subsequent delay evaluation.

[0062] Step S15: performing data integrity assessment on the multi-source sensor data to obtain an integrity assessment matrix, wherein the data integrity assessment includes a data packet sequence integrity check, data packet loss rate and retransmission rate calculation, and data synchronization performance assessment;

[0063] The embodiment of the present invention performs data integrity analysis on multi-source sensor data. First, the transmission order is verified by a data packet sequence integrity check algorithm. Then, the formula The data packet loss rate is calculated to be 0.3%, and the retransmission rate is 0.1% through retransmission packet statistics. Finally, the data synchronization performance analysis (error analysis based on timestamp comparison) confirms that the data synchronization error is less than 5ms, and the integrity assessment matrix M is generated comprehensively. integrity =[0.3%,0.1%,5ms].

[0064] Step S16: performing signal strength evaluation of the signal-to-noise ratio, bandwidth utilization, and signal stability index on the multi-source sensor data, thereby obtaining signal quality vector data; performing system state comprehensive score calculation based on the delay feature vector data, the integrity evaluation matrix, and the signal quality vector data, thereby obtaining initial system state data;

[0065] The embodiment of the present invention calculates the signal-to-noise ratio (SNR) of the signal quality in the multi-source sensor data using the formula The signal-to-noise ratio is 25dB, and the bandwidth utilization is 75% based on traffic statistics. The signal stability is evaluated by the mean and fluctuation range of signal strength (the fluctuation range is ±5dBm). Finally, the signal quality vector data Q is generated by combining the above indicators. signal =[25dB,75%,±5dBm].

[0066] Step S17: Calculate the system state comprehensive score according to the delay feature vector data, the integrity assessment matrix and the signal quality vector data, so as to obtain the initial system state data;

[0067] According to the delay feature vector data T vector =[120ms,10ms], integrity assessment matrix M integrity =[0.3%,0.1%,5ms] and signal quality vector data Q signal =[25dB,75%,±5dBm], through the weighted comprehensive scoring algorithm S system =ω1T avg +ω2P loss +ω3SNR (weights are 0.4, 0.3, and 0.3 respectively) calculates the system status score to be 85 points, and obtains the initial system status data.

[0068] Step S18: setting operating parameter thresholds according to the initial system state data, and performing system parameter calibration on the multi-source sensor data to obtain system calibration data, wherein the operating parameter thresholds include a network delay tolerance threshold, data integrity requirements, and a signal quality lower limit.

[0069] According to the initial system state data of step S17 (system score is 85 points), the embodiment of the present invention sets the operating parameter threshold, wherein the network delay tolerance threshold is set to 150ms, the data integrity requirement is set to a packet loss rate of <0.5%, and the signal quality lower limit is -70dBm. The least squares fitting algorithm is used to systematically adjust the various parameters in the multi-source sensor data, for example, fitting the abnormally fluctuating signal strength value into a continuous trend, thereby obtaining the optimized system calibration data, providing reliable input for subsequent operations.

[0070] The present invention obtains the signal quality data of network transmission such as signal strength, network delay, packet loss rate and bandwidth occupancy rate, so that the surgical system can monitor and understand the transmission status of the network in real time. This helps to identify and solve potential network problems, thereby ensuring the stability and reliability of data transmission during surgery. The transparency of the network status helps to reduce the operation errors caused by network fluctuations and improve the response speed and overall performance of the system. The initial force data matrix of the six-dimensional force sensor at the end of the mechanical arm is collected, including force and torque data, so that the system can capture the force feedback information during the surgical operation. These data provide a basis for the precise control of surgical operations, help doctors better perceive the resistance or subtle tactile changes encountered during surgery, and increase the accuracy of operations and the safety of surgery. The information of the position encoders of each joint of the mechanical arm and the position information of the end effector are collected, and the coordinate system is converted to ensure that the system has unified and standardized position information. These data are integrated into multi-source sensor data with signal quality data and the initial force data matrix, providing a comprehensive data basis for subsequent analysis. This data integration improves the collaborative performance of the system, promotes the effective cooperation between different data sources, and makes the operation control more accurate and stable. The network delay is evaluated by calculating the average delay time and analyzing the delay jitter to obtain the delay feature vector data, which helps the system identify potential delay problems in network transmission. Real-time delay analysis can help the system to compensate and adjust the delay in advance, reduce the inaccuracy caused by delay in operation, and thus improve the real-time and smoothness of surgery. Data integrity evaluation ensures the integrity and reliability of the transmitted data. The evaluation content includes the integrity check of the data packet sequence, the calculation of the packet loss rate and retransmission rate, and the data synchronization performance. These evaluations help the system identify possible omissions or anomalies in the transmission, prevent surgical interruptions or misoperations caused by data loss, and ensure the continuity and accuracy of data transmission. The signal-to-noise ratio, bandwidth utilization and signal stability of multi-source sensor data are evaluated to obtain signal quality vector data, which can help the system judge the overall quality of the network and signal transmission. By evaluating the stability of the signal, the system can determine whether there is interference or fluctuation during the transmission process, thereby reducing surgical uncertainty and errors caused by signal problems. The system status is comprehensively scored based on the delay feature vector data, integrity evaluation matrix and signal quality vector data, so that the system can comprehensively evaluate its initial status under multi-factor considerations. The comprehensive score provides a pre-operation understanding of the overall health of the system, allowing doctors and operators to adjust system parameters before the operation begins to prevent potential problems. Setting operating parameter thresholds and calibrating system parameters based on initial system status data helps ensure that the system operates within the optimal parameter range. Setting operating parameter thresholds such as network delay tolerance thresholds, data integrity requirements, and signal quality minimums allows the system to operate normally even when small fluctuations occur without affecting the accuracy of surgical operations.Calibration of system parameters can effectively improve the response speed of operations and the stability of the system, ultimately increasing the success rate of surgery and patient safety.

[0071] Preferably, step S2 comprises the following steps:

[0072] Step S21: performing time domain feature extraction on the initial force data matrix in the system calibration data to obtain an initial force feature vector, wherein the time domain feature extraction includes calculation of peak force, average force, torque peak value and torque mean;

[0073] Step S22: performing frequency domain feature analysis based on force signal spectrum analysis, main frequency component extraction, and frequency band energy distribution calculation on the initial force data matrix in the system calibration data, thereby obtaining frequency domain feature matrix data;

[0074] Step S23: performing feature fusion on the initial force feature vector and the frequency domain feature matrix data, and performing spatial mapping conversion according to the robot arm posture information, thereby generating six-dimensional force control feature data;

[0075] Step S24: establishing a force feedback spatiotemporal distribution model based on the six-dimensional force control characteristic data, calculating the force feedback distribution characteristics at different operation stages, thereby obtaining force feedback statistical characteristic data, wherein the force feedback distribution characteristics include the spatial distribution density of the force, the time series variation law, and the torque distribution characteristics;

[0076] Step S25: Perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data.

[0077] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0078] Step S21: performing time domain feature extraction on the initial force data matrix in the system calibration data to obtain an initial force feature vector, wherein the time domain feature extraction includes calculation of peak force, average force, torque peak value and torque mean;

[0079] The embodiment of the present invention extracts time domain features from the initial force data matrix in the system calibration data. The specific operation includes calculating the maximum value (peak force, for example, Fx=12N), average value (average force, for example, Fy=8.5N) of each dimensional force signal (such as Fx, Fy, Fz), the maximum value (torque peak value, for example, Mz=1.5Nm) and average value (torque mean, for example, My=0.9Nm) of the torque signal (Mx, My, Mz). The sliding window technology is used to extract these time domain features (the window size is 100ms and the step size is 10ms) to ensure that the characteristic value can reflect the transient changes and overall trends of the signal, and finally generate the initial force characteristic vector.

[0080] Step S22: performing frequency domain feature analysis based on force signal spectrum analysis, main frequency component extraction, and frequency band energy distribution calculation on the initial force data matrix in the system calibration data, thereby obtaining frequency domain feature matrix data;

[0081] The embodiment of the present invention uses fast Fourier transform (FFT) to perform spectrum analysis on the initial force data matrix and extract the main frequency component (for example, the main frequency of the Fx signal is 3Hz). At the same time, the energy distribution of different frequency bands is calculated, for example, the frequency is divided into a low frequency band (0-5Hz), a medium frequency band (5-15Hz) and a high frequency band (15-30Hz), and the energy proportion of each frequency band is counted separately (for example, the low frequency proportion is 70%). The above operation generates frequency domain feature matrix data, and stores the main frequency and energy distribution characteristics of each dimensional force signal in matrix form to ensure that frequency domain feature support is provided for subsequent analysis.

[0082] Step S23: performing feature fusion on the initial force feature vector and the frequency domain feature matrix data, and performing spatial mapping conversion according to the robot arm posture information, thereby generating six-dimensional force control feature data;

[0083] The embodiment of the present invention fuses the initial force feature vector with the frequency domain feature matrix data. The principal component analysis (PCA) is used to reduce the dimension of the force signal features and remove redundant features. At the same time, the space mapping conversion is performed in combination with the end posture information of the manipulator, for example, the force signal is mapped to the Cartesian coordinate system of the operating space. The spatial conversion is completed through the homogeneous coordinate transformation matrix to ensure the consistency of the force control features in the spatial dimension. Finally, six-dimensional force control feature data is generated, which includes the comprehensive features of the fused force value and torque value.

[0084] Step S24: establishing a force feedback spatiotemporal distribution model based on the six-dimensional force control characteristic data, calculating the force feedback distribution characteristics at different operation stages, thereby obtaining force feedback statistical characteristic data, wherein the force feedback distribution characteristics include the spatial distribution density of the force, the time series variation law, and the torque distribution characteristics;

[0085] The embodiment of the present invention establishes a spatiotemporal distribution model of force feedback based on six-dimensional force control characteristic data, and uses a spatiotemporal distribution analysis algorithm to calculate the force feedback distribution characteristics of different operation stages. Specifically, it includes: calculating the force distribution density of each unit in the operation space through three-dimensional grid division (for example, the maximum density is 0.85N / cm 3 ); extract the time series variation law of force value (for example, the average force variation trend in the operation stage is gradually increasing from 6N to 10N); analyze the torque distribution characteristics (for example, the torque distribution density in the Z-axis direction is the largest). Finally, generate force feedback statistical feature data to provide input for operation mode analysis.

[0086] Step S25: Perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data.

[0087] The embodiment of the present invention performs a difference analysis on the six-dimensional force control feature data. By comparing the theoretical values ​​in the standard force control model with the actual force control feature data, the deviation matrix is ​​calculated. For example, the mean square error of the force deviation is 0.15N, and the torque deviation is 0.02Nm. The error analysis algorithm is used to identify the key deviation points in the feature and generate a visual distribution map, and finally the force feedback deviation data is obtained for real-time correction of the parameter configuration of the force control system.

[0088] The present invention extracts time domain features from the initial force data matrix in the system calibration data to obtain the initial force feature vector. This process helps the system capture the core features of the force signal in the time dimension. Time domain feature extraction calculates peak force, average force, torque peak and torque mean, which helps to evaluate the basic mechanical performance of the manipulator under different operating conditions. This improves the system's perception accuracy of the force and torque required in the operation, helps the surgical system to better adapt to real-time changes and external interference, and ensures the stability and accuracy of the operation. Perform frequency domain feature analysis, including spectrum analysis, main frequency component extraction and frequency band energy distribution calculation, to obtain frequency domain feature matrix data, so that the system can identify the frequency characteristics of the force signal. This analysis provides energy distribution information of the signal in different frequency bands, revealing the potential periodic behavior and dynamic characteristics of force feedback during the operation. Frequency domain analysis can help detect abnormal vibration and instability, thereby improving the system's responsiveness and early warning function. The initial force feature vector and the frequency domain feature matrix data are feature fused, and spatial mapping conversion is performed based on the manipulator's posture information to generate six-dimensional force control feature data. Feature fusion provides comprehensive force feedback features, and the spatial mapping of posture information ensures the physical meaning and multi-dimensional coordination of the data. This step enhances the system's adaptability to complex operation scenarios, enabling the system to better handle force feedback in different directions in space, and improve precision control and stability during surgery. Based on the six-dimensional force control feature data, a force feedback spatiotemporal distribution model is established and the force feedback distribution characteristics at different operation stages are calculated, which helps to analyze and understand the distribution and change law of force during the operation. The obtained force feedback statistical feature data, including the spatial distribution density of force, the time series change law and the torque distribution characteristics, provide in-depth analysis of operation data. This can help the surgical system optimize the response mechanism of force feedback, improve the accuracy of operation and reduce possible over-force or under-force problems, thereby improving the safety and effectiveness of surgery. The six-dimensional force control feature data is analyzed for differences and force feedback deviation data is generated, so that the system can identify deviations from expected feedback. This analysis helps detect abnormal force changes or emergencies that may occur during the operation, thereby triggering the system's early warning mechanism or automatic adjustment strategy. This can reduce the risk of misoperation, ensure that doctors obtain more stable and controllable force feedback during remote surgery, and improve the reliability and accuracy of the overall surgery.

[0089] Preferably, step S25 comprises the following steps:

[0090] Step S251: performing historical data comparison and analysis on the six-dimensional force control characteristic data to obtain force feedback reference data, wherein the historical data includes historical records of force magnitude, force direction and torque characteristics;

[0091] The embodiment of the present invention performs historical data comparative analysis on the six-dimensional force control feature data, specifically by querying the historical force control feature records stored in the database, including historical force magnitude (such as the range of Fx is 10-12N), force direction (such as the direction vector [0.7, 0.7, 0.1]) and torque characteristics (such as the range of Mx is 0.5-0.8Nm). The comparative analysis uses a sliding time window technology (the window length is 5 seconds) to perform similarity matching between the current six-dimensional force control feature and the historical data of the corresponding time period, calculate the force feedback reference value of each dimension (such as the Fx reference value is 11N), and generate force feedback reference data.

[0092] Step S252: performing real-time comparison between the six-dimensional force control characteristic data and the force feedback reference data, and calculating the force amplitude deviation, the force mean deviation and the force fluctuation deviation, so as to obtain the force magnitude deviation vector;

[0093] The embodiment of the present invention compares the six-dimensional force control characteristic data with the force feedback reference data in real time, and specifically uses the differential calculation method to calculate the amplitude deviation of each dimensional force signal (such as the amplitude deviation of Fx is 0.3N), and simultaneously calculates the mean deviation of the signal (such as the mean deviation of Fy is 0.2N) and the fluctuation deviation (such as the fluctuation deviation of Fz is ±0.15N). The deviation vector (such as [0.3N, 0.2N, ±0.15N]) is stored in matrix form to represent the real-time force magnitude deviation, and finally a force magnitude deviation vector is generated.

[0094] Step S253: performing force direction deviation analysis according to the force magnitude deviation vector, calculating the force direction angle deviation, direction stability deviation and direction consistency deviation, thereby obtaining a force direction deviation matrix;

[0095] The embodiment of the present invention performs force direction deviation analysis based on the force magnitude deviation vector. Specifically, the angle deviation between the actual force direction vector and the reference force direction vector is calculated (e.g., the direction angle deviation is 2.5°), and the standard deviation of the direction signal (e.g., the direction stability deviation is 0.1) and the correlation coefficient of the direction vector (e.g., the direction consistency deviation is 0.98) are calculated at the same time. The above calculation results are stored as a force direction deviation matrix for further analysis of the change characteristics of the force direction.

[0096] Step S254: calculating the torque deviation including the torque amplitude deviation, the torque direction deviation and the torque balance deviation according to the force magnitude deviation vector, thereby obtaining torque deviation characteristic data;

[0097] The embodiment of the present invention uses the force magnitude deviation vector to calculate the deviation of the torque characteristic. Specifically, the difference between the actual torque amplitude and the reference value (such as the Mx amplitude deviation of 0.2Nm), the deviation of the direction angle (such as the torque direction deviation of 1.8°) and the balance deviation of the torque on the three coordinate axes (such as the Z axis deviation of 0.05Nm) are calculated. The above eigenvalues ​​are represented by a matrix, and finally the torque deviation characteristic data is generated for evaluating the force control performance.

[0098] Step S255: performing statistical analysis on the force direction deviation matrix and the torque deviation characteristic data, including the average deviation rate, the maximum deviation value and the deviation fluctuation range, so as to obtain deviation statistical data;

[0099] The embodiment of the present invention performs statistical analysis on the force direction deviation matrix and the torque deviation characteristic data, and calculates the average deviation rate (e.g., the average direction deviation is 1.5°), the maximum deviation value (e.g., the maximum torque deviation is 0.25Nm), and the deviation fluctuation range (e.g., the force direction fluctuation range is ±0.3°). The statistical analysis results are used to generate deviation statistical data, and the changing trends of various deviation characteristics are displayed in the form of charts.

[0100] Step S256: setting a deviation threshold system including an average deviation threshold, a maximum deviation threshold and a fluctuation range threshold according to the deviation statistical data, thereby obtaining a deviation evaluation matrix;

[0101] The embodiment of the present invention sets a deviation threshold system according to deviation statistical data, including an average deviation threshold (such as 1.2°), a maximum deviation threshold (such as 0.3Nm), and a fluctuation range threshold (such as ±0.25°). These thresholds are used to construct a deviation evaluation matrix, which characterizes the critical situation of each dimension deviation in the current operation stage through the matrix, ensuring that the deviation information can be quantitatively evaluated.

[0102] Step S257: performing abnormality discrimination processing according to the deviation assessment matrix, calculating abnormal characteristic parameters based on the degree of deviation exceeding the limit, the duration of deviation and the trend of deviation change, thereby obtaining an abnormal characteristic vector;

[0103] The embodiment of the present invention determines the abnormality of the system based on the deviation assessment matrix, specifically calculating the degree of deviation exceeding the limit (such as the directional deviation exceeding the threshold by 1.3 times), the duration of the deviation (such as the overlimit lasting for 3 seconds) and the trend of the deviation change (such as the deviation gradually intensifies, with an increase of 10% / second). The above data is aggregated and analyzed to generate an abnormal feature vector, which is used to judge the abnormality level in the operation in real time and provide a basis for decision-making.

[0104] Step S258: Perform feature fusion on the deviation evaluation matrix and the abnormal feature vector to obtain force feedback deviation data.

[0105] The embodiment of the present invention performs feature fusion of the deviation assessment matrix and the abnormal feature vector, which is specifically realized by a weighted feature aggregation method. First, according to the deviation values ​​of each dimension in the deviation assessment matrix (such as the directional deviation is 1.3°, the torque deviation is 0.2Nm) and the abnormal degree factor in the abnormal feature vector (such as the over-limit degree weight is 0.6, and the duration weight is 0.4), a weight coefficient is allocated to highlight the contribution of the abnormal feature to the overall deviation. Then, the two are matrix-weighted summed to obtain a comprehensive deviation feature matrix (such as the directional comprehensive deviation is 1.5°, and the torque comprehensive deviation is 0.25Nm). Finally, the comprehensive deviation feature matrix is ​​normalized to ensure that the numerical ranges between different feature dimensions are consistent, and standardized force feedback deviation data are generated. Through the force feedback deviation data, accurate force control performance evaluation and real-time adjustment basis can be provided for subsequent steps.

[0106] The present invention performs historical data comparison and analysis on six-dimensional force control feature data to generate force feedback reference data, which can help the system establish an operation benchmark based on history. By recording historical data of force magnitude, direction and torque characteristics, the system can better understand the scope and trend of normal operation status. This enables the system to quickly detect abnormal force changes in real-time operation, improve the system's abnormal detection accuracy and response speed, and thus reduce operational risks. By comparing the six-dimensional force control feature data with the reference data in real time and calculating various force deviation vectors, a detailed analysis of the system's real-time feedback can be provided, including the amplitude, mean and fluctuation deviation of the force. This detailed analysis helps the system identify subtle but potentially harmful force changes, provides basic data support for the system's force feedback adjustment, and improves the precise control and feedback adjustment capabilities during surgery. Performing force direction deviation analysis and calculating direction angle deviation, stability deviation and consistency deviation enables the system to not only detect force magnitude changes, but also identify abnormalities in force direction. In this way, the system can more comprehensively evaluate the stability and consistency of the operation, which helps to improve the directional control accuracy of the robotic arm under complex operations. The calculation of torque deviation, including the analysis of torque amplitude, direction and balance deviation, can provide in-depth understanding of the abnormal performance of torque during operation. This helps to timely identify torque anomalies that affect the balance and stability of the robot arm, avoid operational deviations caused by torque anomalies, and improve the safety and stability of operation. Statistical analysis of deviation data, calculation of average deviation rate, maximum deviation value and fluctuation range, provides an overall evaluation view of the deviation. This helps the system identify long-term or large force deviations, so that corresponding measures can be taken to ensure that the system maintains stable and accurate feedback control under different operating conditions. The deviation threshold system is set according to the deviation statistical data, so that the system can set a reasonable deviation acceptance range, including average deviation, maximum deviation and fluctuation range thresholds. This threshold system provides a standard for subsequent abnormality detection and system adjustment, which helps to improve the system's adaptive ability and reduce unnecessary false alarms and undetected abnormalities. The degree of abnormality is discriminated and processed according to the deviation evaluation matrix and the abnormal characteristic parameters are calculated, which helps to identify potential risks that may exist in the operation, such as deviation exceeding the limit, long duration and change trend. This step provides a multi-dimensional analysis of abnormalities, enhances the system's early warning and rapid response capabilities, and thus improves the safety and reliability of remote operation. The deviation evaluation matrix and the abnormal feature vector are feature-fused to obtain the final force feedback deviation data, providing a comprehensive deviation analysis result. This step integrates all deviation information, allowing the system to make comprehensive decisions and control adjustments during operation, helping to reduce operational errors caused by deviations and improve the system's adaptive and feedback control capabilities in complex surgical operations.

[0107] Preferably, step S3 comprises the following steps:

[0108] Step S31: performing time series analysis on the force feedback deviation data to extract the delay characteristics of the force feedback signal, thereby obtaining an initial delay characteristic vector, wherein the delay characteristics include signal transmission delay, system response delay and operation delay;

[0109] The embodiment of the present invention performs a time series analysis on the force feedback deviation data, uses the sliding time window technology to segment the continuously sampled force feedback signal (for example, each time window is 100ms long), and extracts the delay characteristics of each segment of the force feedback signal. The signal transmission delay is calculated by analyzing the difference between the peak time point of the signal and the trigger time point of the operation instruction; the system response delay is calculated by detecting the time difference between the system output response and the input instruction; the operation delay is calculated by combining the synchronization of the operation task execution time and the force feedback signal change. Finally, the initial delay feature vector is generated by integrating various delay features, for example, the signal transmission delay is 30ms, the system response delay is 50ms, and the operation delay is 20ms. The initial delay feature vector provides basic data for subsequent delay evaluation.

[0110] Step S32: performing a real-time dynamic evaluation on the initial delay feature vector based on the delay fluctuation range, delay trend and delay stability index calculation, thereby obtaining a delay dynamic evaluation matrix;

[0111] The embodiment of the present invention performs real-time dynamic evaluation on the initial delay feature vector. First, the delay fluctuation range calculation method is used to analyze the difference between the maximum and minimum values ​​of the delay feature in multiple time windows (for example, the signal transmission delay fluctuation range is 10ms); secondly, the delay trend analysis algorithm is used to calculate the change trend of the delay over time (such as the delay gradually decreases with a slope of -2ms / s); finally, the stability of the delay signal is quantified according to the delay stability index (such as the mean square error and the coefficient of variation) (for example, the mean square error of the system response delay is 5ms). The above results are integrated to generate a delay dynamic evaluation matrix, including dimensions such as delay fluctuation range, trend change and stability score, to provide a reference for the comprehensive evaluation of delay characteristics.

[0112] Step S33: performing delay impact assessment according to the delay dynamic assessment matrix, thereby obtaining delay feature assessment data, wherein the delay impact assessment includes an assessment of the impact of the delay on operation accuracy, system stability and security;

[0113] The embodiment of the present invention evaluates the impact of delay based on the delay dynamic evaluation matrix, specifically evaluating the impact of delay on operational accuracy, system stability and safety. Through simulation tests and experimental data verification, the impact of delay on the position deviation of the end effector of the robotic arm is analyzed (for example, a delay increase of 10ms causes a decrease in operational accuracy of 0.5mm); the frequency domain analysis method is used to evaluate the impact of delay on the vibration amplitude and stability index of the system (such as a 10% reduction in the system stability score); combined with real-time safety threshold analysis during surgery, the impact of delay on safety is quantified (for example, a delay exceeding 50ms may cause the risk of misoperation). The above evaluation results are integrated to generate delay feature evaluation data to provide a basis for dynamic compensation.

[0114] Step S34: performing data noise reduction processing on the system calibration data, and performing delay compensation processing on the delay characteristic evaluation data to generate dynamic compensation data.

[0115] The embodiment of the present invention performs data denoising on the system calibration data. First, a denoising method based on wavelet transform is used to separate the high-frequency noise component in the force feedback signal (such as reducing the noise amplitude to 20% of the original value); then, combined with the delay feature evaluation data, the force feedback signal is dynamically adjusted through a delay compensation algorithm to compensate for the time difference caused by network delay and system response delay. For example, in the case where the signal transmission delay is 30ms, real-time synchronous compensation is achieved by adjusting the timing distribution of the force feedback signal to ensure operation accuracy and response speed. Finally, the output dynamic compensation data is used to optimize the real-time control performance of the surgical system and improve the stability and safety of the system.

[0116] The present invention performs time series analysis on force feedback deviation data and extracts delay features, which can help the system identify and quantify various types of delays in signal transmission and system response. This step ensures the generation of delay feature vectors, so that the system can grasp the signal transmission situation, the system response speed and the delay caused by the operation in real time. This is crucial to maintaining the smoothness and responsiveness of remote surgery in a dynamic environment, and helps prevent operational deviations and risks caused by delays. Real-time dynamic evaluation of the initial delay feature vector helps the system identify the fluctuation range, trend and stability of the delay. Through this dynamic evaluation matrix, the system can understand the short-term and long-term behavioral characteristics of the delay, thereby providing detailed data support for further delay compensation and adjustment. This evaluation improves the system's adaptive ability in dealing with network jitter and response time fluctuations, ensuring the accuracy and continuity of operations in remote surgery. The delay impact assessment is performed to quantify the potential impact of delay on operational accuracy, system stability and safety. This step provides the system with an intuitive evaluation of the delay on the overall operational effect, helping to identify potential operational risks and prevent operational errors that may be caused by delays. Through this detailed delay feature evaluation data, the system can more accurately predict risks and make advance adjustments to ensure the efficiency and safety of operations. Performing data noise reduction on system calibration data and delay compensation processing in combination with delay feature evaluation data is an important step to reduce the impact of delay on the system. Through noise reduction processing, the system can eliminate interfering noise and enhance the clarity and accuracy of data. Subsequent delay compensation processing can adjust the system response in real time, thereby reducing the impact of delay on operational accuracy and safety, and improving the sensitivity and stability of the system in actual operation. This comprehensive processing improves the system's stable operation capability and real-time response performance in complex remote environments.

[0117] Preferably, step S34 includes the following steps:

[0118] Step S341: performing wavelet transform analysis on the system calibration data to identify and filter high-frequency noise components, thereby obtaining noise reduction system calibration data;

[0119] The embodiment of the present invention performs wavelet transform analysis on the system calibration data, selects Daubechies wavelet (such as db4) to perform multi-layer decomposition on the force feedback signal, and decomposes the signal into high-frequency and low-frequency components. For the high-frequency noise component, a soft threshold filtering method is used, and the noise threshold is set to 10% of the maximum amplitude of the original signal. The high-frequency components exceeding the threshold are attenuated, while the effective signals in the low-frequency components are retained. The signal is reconstructed by inverse transformation to generate noise reduction system calibration data, and the signal-to-noise ratio of the force feedback signal is increased to more than 20 times of the original data, meeting the accuracy requirements of subsequent compensation processing.

[0120] Step S342: establishing a delay compensation prediction model based on the delay characteristic evaluation data, and calculating optimal compensation parameters under different operation scenarios according to the noise reduction system calibration data, thereby obtaining optimal compensation parameter data;

[0121] The embodiment of the present invention uses a long short-term memory network (LSTM) to establish a delay compensation prediction model based on delay feature evaluation data, takes the delay dynamic evaluation matrix as the model input, and minimizes the operation delay as the target output. Combined with the noise reduction system calibration data, the compensation parameters are optimized through multi-scenario simulation calculations (such as surgical cutting scenarios and suturing scenarios). For example, the compensation parameter in the network transmission delay scenario is 15ms, and the compensation parameter in the system response delay scenario is 25ms. Finally, the output optimal compensation parameter data covers the best compensation values ​​under different operation scenarios, which is suitable for real-time dynamic adjustment.

[0122] Step S343: adjusting the dynamic compensation parameters based on adaptive compensation according to the optimal compensation parameter data, thereby obtaining a compensation optimization matrix;

[0123] The embodiment of the present invention uses an adaptive control algorithm to dynamically adjust the compensation parameters based on the optimal compensation parameter data. The specific method is to dynamically adjust the compensation parameter weights using the change trend of the real-time force feedback signal to ensure a stable compensation effect. For example, when the network delay increases, the compensation rate is increased by the weighted compensation parameter gain. If the transmission delay increases by 10ms, the compensation value after adaptive adjustment increases to 20ms. The generated compensation optimization matrix contains the dynamically adjusted real-time compensation parameters, which are used to guide the compensation control of system delays.

[0124] Step S344: The compensation optimization matrix is ​​integrated with the noise reduction system calibration data, and real-time compensation control of the system delay is performed to generate dynamic compensation data.

[0125] The embodiment of the present invention fuses the compensation optimization matrix with the calibration data of the noise reduction system, and realizes the precise alignment of the force feedback signal through signal reconstruction. Subsequently, based on the real-time feedback operation delay data, a method combining feedforward and feedback control is used to perform real-time compensation control. For example, for the operation delay deviation detected in real time, the signal timing is adjusted in advance through feedforward compensation, and the compensation amplitude is refined by feedback control, and finally the delay error is controlled within 2ms. The generated dynamic compensation data is used to optimize the operation accuracy and response speed of the robotic arm in real time, and improve the performance and safety of the remote surgery system.

[0126] The present invention performs wavelet transform analysis on the system calibration data, which can effectively identify and filter high-frequency noise components. This processing method conducts an in-depth analysis of the time-frequency characteristics of the data, thereby effectively removing noise and retaining key signal details, making the system calibration data after noise reduction clearer and more stable. This processing ensures the accuracy and reliability of subsequent data processing and analysis, improves the accuracy of the system during remote surgery, and reduces misoperation and misjudgment caused by noise interference. A delay compensation prediction model is established based on the delay feature evaluation data, and the optimal compensation parameters under different operation scenarios can be calculated according to the noise-reduced system calibration data. This process ensures that the system can dynamically adapt to the changes in delay under various operating conditions, and provides appropriate compensation adjustment parameters, which improves the system's ability to cope with complex network environments and changes in operating loads. Through the accurately calculated optimal compensation parameters, the system can more efficiently adjust the delay in actual operation and maintain the smoothness and accuracy of the operation. Dynamic adjustment of adaptive compensation based on the optimal compensation parameter data and generation of the compensation optimization matrix are important steps to ensure that the system can adjust and optimize the compensation effect in real time. This dynamic adjustment process enables the system to quickly adjust the compensation strategy according to the real-time situation, so as to always maintain the best system response during operation. Through this adaptive mechanism, the system can actively respond to delay changes, reduce the adverse effects of unstable factors on operations, and improve the overall response speed and accuracy of remote surgery. Fusion of the compensation optimization matrix with the noise reduction system calibration data and real-time compensation control of system delay are important links in achieving real-time compensation effects. This step ensures the effective combination of noise reduction data and dynamic compensation parameters, thereby achieving high-precision delay compensation control. Through this process, the system can continuously perform compensation optimization in actual operations, dynamically adapt to network delays and system responses in different scenarios, and reduce operational lags and potential risks. Ultimately, the generated dynamic compensation data can significantly improve the consistency and stability of operations, and ensure safety and efficiency during remote surgery.

[0127] Preferably, step S342 includes the following steps:

[0128] Performing time series analysis on the delay feature evaluation data to extract delay pattern features, thereby obtaining a delay pattern matrix, wherein the delay pattern features include periodic delay features, burst delay features, and cumulative delay features;

[0129] A surgical operation scene library is established based on the calibration data of the noise reduction system. The surgical operations in the surgical operation scene library are classified into three basic types: fine operation, fast operation and routine operation. The characteristic parameters of each type of operation are extracted to obtain the scene feature vector.

[0130] Training a delay compensation prediction model based on a delay pattern matrix and a scene feature vector, wherein the delay compensation prediction model includes a delay prediction sub-model and a compensation amount prediction sub-model;

[0131] Cross-validate the delay compensation prediction model under different operation scenarios, evaluate the model performance, and improve the prediction accuracy by optimizing the model parameters to obtain the optimized prediction model;

[0132] Compensation parameters for various operation scenarios are calculated based on the optimization prediction model to obtain a compensation parameter set, where the compensation parameters include a time compensation coefficient, a spatial position compensation coefficient, and a force feedback compensation coefficient;

[0133] Conduct stability analysis on the compensation parameter set, evaluate the reliability of the compensation effect, and establish a compensation parameter credibility evaluation index to obtain a parameter credibility matrix;

[0134] The compensation parameter set is screened and optimized according to the parameter credibility matrix, and unreliable compensation parameters are eliminated to obtain the optimal compensation parameter data.

[0135] The embodiment of the present invention extracts the periodicity, burstiness and cumulative characteristics of the delay data by performing an autoregressive moving average model (ARIMA) analysis on the delay feature evaluation data. Specifically, the periodic delay identifies the main periodic components through FFT spectrum analysis. For example, it is found that the delay signal has a periodic fluctuation of 10Hz; the burst delay uses a sliding window standard deviation detection algorithm to extract the instantaneous peak change; the cumulative delay calculates the long-term cumulative effect of the delay through a trend fitting model (such as linear or polynomial fitting), and finally generates a delay pattern matrix containing multidimensional delay characteristics. The operation trajectory, force feedback signal and response time characteristics are extracted from the system calibration data after noise reduction processing, and are classified into fine operation (such as vascular suturing), fast operation (such as tissue cutting) and conventional operation (such as instrument adjustment) according to the complexity and speed requirements of the operation. The key characteristic parameters of the three types of operations are extracted respectively, including the average length of the operation path, the peak force variation range and the time response coefficient, and a multidimensional scene feature vector is constructed. For example, the feature vector of the fine operation is a path length of 10mm, a force variation amplitude of 0.5N, and a response time of 0.2s. The Bidirectional Long Short-Term Memory (BiLSTM) model in deep learning is used. The delay pattern matrix is ​​used as input and the scene feature vector is used as auxiliary feature. The delay prediction sub-model and the compensation amount prediction sub-model are trained in stages. The delay prediction sub-model is used to estimate the delay characteristics of different operation scenarios. For example, the average delay of the fine operation scenario is predicted to be 15ms. The compensation amount prediction sub-model calculates the corresponding compensation value. For example, the recommended time compensation coefficient is 12ms. The mean square error (MSE) is used as the loss function during model training to perform hyperparameter tuning. K-fold cross validation (such as K=5) is used to evaluate the model performance. The indicators include prediction accuracy (R2 ), mean square error (MSE), and prediction stability (standard deviation). In the fast operation scenario, model R 2 Reached 0.95, MSE is 3ms 2 . By performing grid search optimization on model parameters (such as learning rate and number of hidden layer nodes), the prediction accuracy is improved, and the prediction error of all scenarios is reduced by more than 10%, and the optimized prediction model is finally output. The scene feature vector of the optimized prediction model is input, and the compensation parameter set for each type of operation scenario is calculated, including the time compensation coefficient (such as 14ms for fine operation scenario), the spatial position compensation coefficient (position deviation correction 0.8mm) and the force feedback compensation coefficient (force deviation adjustment 0.3N). The compensation parameter set is graded for the specific needs of different operation scenarios to ensure that the compensation results accurately match the scene characteristics. The random noise perturbation method is used to evaluate the robustness of the compensation parameters under different delay conditions. Taking the time compensation coefficient as an example, the compensation effect is evaluated by adding 5%-20% delay perturbation, and the average compensation error and fluctuation range are calculated. Combined with the reliability index of the compensation effect, a credibility evaluation model is established, and the parameter credibility matrix is ​​output. For example, the credibility score of the time compensation coefficient is 0.92. Set a credibility threshold (such as 0.85), eliminate the compensation parameters below the threshold, and perform weighted average optimization on the remaining parameters to generate the optimal compensation parameter data. For example, the spatial position compensation coefficients with a credibility lower than 0.8 in the fast operation scenario are eliminated, and finally the optimized compensation parameter set is obtained, including the time compensation coefficient of 14ms, the spatial position compensation coefficient of 0.75mm and the force feedback compensation coefficient of 0.28N.

[0136] The present invention performs time series analysis on the delay feature evaluation data, extracts periodic, bursty and cumulative delay features, and can identify the behavior patterns of delays at different time points. This helps to gain a deep understanding of the changing trend of delay features and their potential impact on the system during the operation. This step can predict and identify potential delay risks in advance during the operation, thereby providing a reference for subsequent compensation strategies and ensuring the consistency and accuracy of remote surgery. A surgical operation scenario library is established based on the calibration data of the noise reduction system, and the operations are divided into three categories: fine operations, fast operations and routine operations, which helps to accurately identify and understand the different types of surgical operations and their requirements. Extracting the characteristic parameters of these operations to generate scene feature vectors helps to refine the pertinence of the delay compensation model, so that the system can dynamically adapt and compensate according to the different needs of the operation scenario. The delay compensation prediction model is trained by the delay pattern matrix and the scene feature vector, and a delay prediction submodel and a compensation amount prediction submodel are established, which can predict the delay in advance and calculate the appropriate compensation amount. This process improves the system's predictability of network delays and surgical scene changes, makes compensation measures more accurate and efficient, and improves the system's response speed and overall performance. Cross-validation and performance evaluation of the delay compensation prediction model under different operation scenarios help to discover potential defects and optimization points of the model. The prediction accuracy of the model is improved through parameter optimization, making it more stable and reliable in practical applications. This optimization improves the applicability of the model in various surgical scenarios and ensures the effectiveness and consistency of delay compensation. Compensation parameter calculations are performed based on the optimized prediction model to generate compensation parameter sets including time compensation coefficients, spatial position compensation coefficients, and force feedback compensation coefficients, so that the system can automatically select the best compensation measures according to different operation scenarios. The use of these parameters helps to maintain the precise position and force feedback of surgical instruments in remote surgery, ensuring the accuracy of the operation and the quality of the operation. The stability analysis of the compensation parameter set is performed, and a credibility evaluation index is established to ensure the reliability of the compensation parameters. The use of the credibility matrix can identify and filter out less stable or unreliable compensation parameters, thereby avoiding operational deviations caused by unstable compensation strategies. This process enhances the stability of the system and reduces potential risks. The compensation parameter set is screened and optimized according to the parameter credibility matrix, unreliable compensation parameters are eliminated, and the optimal compensation parameter data is finally obtained. This step ensures that the compensation strategy used by the system in actual operation is highly reliable and consistent, improves the compensation effect, reduces the operational risks caused by errors, and ultimately improves the overall safety and success rate of remote surgery.

[0137] Preferably, step S4 comprises the following steps:

[0138] Step S41: performing operation mode recognition according to the force feedback statistical feature data and the dynamic compensation data, thereby generating an operation feature vector, wherein the operation mode recognition includes operation type classification and operation intensity evaluation;

[0139] The embodiment of the present invention utilizes the spatial distribution density of force, time series variation law and other information in the force feedback statistical feature data, combined with the dynamic adjustment trend of compensation parameters in the dynamic compensation data, and adopts the support vector machine (SVM) classification algorithm to identify the surgical operation mode. The specific method includes classifying the operation into three categories: fine type, conventional type and strong type, setting the threshold value by the mean and standard deviation of the force feedback (for example, the peak value of the fine operation force is less than 1N, and the mean value of the strong operation force is greater than 5N) for classification, and calculating the operation intensity, for example, the intensity of the fine operation is evaluated as 0.3 (normalized value). The final output operation feature vector contains two key features: operation type and intensity.

[0140] Step S42: identifying the current surgical operation type and operation parameters of the operation feature vector, thereby generating operation type feature data;

[0141] The embodiment of the present invention identifies the specific type of the current surgical operation and related operation parameters by matching the operation feature vector with the constructed operation mode database. For example, the operation feature vector matches the "vascular suturing" mode, and the extracted parameters include the suturing path length of 5mm and the operation time interval of 2s. The operation parameter identification adopts the similarity matching method based on the K nearest neighbor (KNN) algorithm, and the closest operation type is selected by calculating the Euclidean distance to generate the operation type feature data containing the operation type name and parameter details.

[0142] Step S43: performing operation risk classification on the operation type feature data to obtain risk level data;

[0143] The embodiment of the present invention performs real-time risk classification of the operation based on the operation parameters in the operation type characteristic data and the key part information of the operation, combined with the preset risk assessment rules. The specific implementation method includes establishing a risk assessment model based on fuzzy logic. For example, in the fine operation mode, if the force fluctuation range exceeds the set threshold (such as 0.5N), the risk level is assessed as "high risk"; if the fluctuation range is within 0.2N, it is assessed as "low risk". The output risk level data includes risk level (such as "medium risk") and risk cause analysis (such as "large force feedback fluctuation").

[0144] Step S44: Perform real-time safety policy matching according to the preset surgical safety policy library and risk level data, and select a combination of safety control parameters to generate surgical safety control data.

[0145] The embodiment of the present invention queries the surgical safety strategy library based on risk level data and selects a combination of safety control strategies that meet the current operation scenario. For example, for medium-risk operations, the matching strategies include increasing the response speed of dynamic compensation parameters, reducing the peak limit of the operating force to 3N, and activating the real-time force feedback alarm function. The selected safety control parameters (such as the response speed is adjusted to 100ms and the force limit is adjusted to 2.8N) are combined using a multi-objective optimization algorithm based on priority sorting to generate surgical safety control data for guiding real-time safety adjustments for subsequent operations.

[0146] The present invention helps the system understand the operation mode of the current operation by classifying the operation type (such as fine operation, fast operation, etc.) and evaluating the operation intensity (such as the magnitude of force and duration, etc.). Through the accurate analysis of the operation type and intensity, the system can identify the current operation state, thereby providing a basis for subsequent safety and precision adjustments. This not only helps to optimize the operation process in real time, but also improves the automation and accuracy of the operation. This step helps to further refine the operation information, identify the specific type of operation (such as fine operation or fast operation) and its parameter settings (such as operation intensity, speed, etc.). Accurately identifying the operation type and parameters enables the system to make flexible and effective adjustments in different surgical environments to adapt to different types of surgical needs, thereby improving the adaptability and accuracy of the operation. Through in-depth analysis of the operation mode and type, the system can perform a graded assessment of the potential risks in the operation process (such as low risk, medium risk, high risk). This risk classification helps to provide targeted safety management strategies for different types of surgical processes, timely identify potential dangers and deal with them, thereby reducing the risks in the surgical process and ensuring the safety of the operation. According to different risk levels, the system will match corresponding safety control measures from the preset safety policy library, such as adjusting the control parameters of surgical instruments, limiting the operation force or speed, etc., to ensure that the operation is performed within a safe range. This process realizes dynamic safety control of the surgical process, and can adjust the operation strategy according to real-time risks to improve the safety and controllability of the operation.

[0147] Preferably, step S5 comprises the following steps:

[0148] Step S51: establishing a robot arm motion state model according to the dynamic compensation data, and setting motion constraint conditions according to the surgical safety control data, thereby obtaining an initial motion control model;

[0149] The embodiment of the present invention utilizes the real-time compensation data and historical motion trajectory in the dynamic compensation data, and applies the Kalman filter to establish the motion state model of the robot arm. The model includes the prediction and correction of the position, velocity and acceleration of the robot arm. The motion constraints in the surgical safety control data, such as the limitation of the operating space (for example, the operating radius does not exceed 200mm) and the maximum allowable motion speed (for example, the maximum linear speed does not exceed 0.5m / s) are included in the constraints to ensure that the motion model meets the safety and accuracy requirements of the surgical operation, thereby obtaining the initial motion control model.

[0150] Step S52: performing real-time state parameter identification on the initial motion control model, extracting motion features including position state parameters, velocity state parameters and acceleration state parameters, thereby obtaining a state feature matrix;

[0151] According to the initial model of motion control, the embodiment of the present invention collects the position, velocity and acceleration data of the robot arm through sensors at each moment. The state parameters are extracted by using a data fusion algorithm (such as weighted average filtering), and the position state (for example, X=0.12m, Y=0.09m), velocity state (for example, V=0.02m / s) and acceleration state (for example, A=0.005m / s) are calculated. 2 ). These state parameters are combined with real-time surgical environment information (such as the precise location of the surgical site) to form a state feature matrix, in which the value of each parameter represents the actual operating state of the current robotic arm.

[0152] Step S53: establishing a remote surgery control model according to the state feature matrix, wherein the remote surgery control model includes a motion trajectory prediction sub-model and a safety control sub-model;

[0153] The embodiment of the present invention uses the state feature matrix to construct a remote surgery control model through a deep learning algorithm (such as an LSTM neural network). The model includes a motion trajectory prediction sub-model (predicting the future position and speed of the robot arm) and a safety control sub-model (performing real-time safety constraint judgment based on surgical safety control data). The model input is a state feature matrix, including the position information, speed, acceleration, etc. of the robot arm, and the output is the trajectory prediction data and safety assessment results of future operations (such as whether the current trajectory violates the surgical safety space restrictions).

[0154] Step S54: using the remote surgery control model to predict the position of the robot arm at the next moment, and performing safety constraint verification according to the surgery safety control data, so as to obtain a predicted trajectory vector;

[0155] The embodiment of the present invention is based on the remote surgery control model, inputs the state feature matrix at the current moment, and predicts the position of the robot arm at the next moment. Assuming that the current position of the robot arm is X=0.12m, Y=0.09m, and the current speed is V=0.02m / s, the system calculates the predicted position X'=0.125m, Y'=0.095m according to the model. Then, the prediction result is verified by safety constraints according to the surgical safety control data to ensure that the predicted position is within the predetermined safety range. If the predicted position exceeds the safety range (for example, the maximum radius of the surgical area is 200mm), the predicted trajectory is adjusted to meet the safety conditions.

[0156] Step S55: performing trajectory optimization processing on the predicted trajectory vector based on the operation accuracy requirement, system stability requirement and safety requirement, so as to obtain optimized trajectory data;

[0157] The embodiment of the present invention uses an optimization algorithm (such as a particle swarm optimization algorithm) to optimize the predicted trajectory vector. First, the trajectory is fine-tuned based on the operation accuracy requirements (for example, the target point error is less than 1mm); second, the system stability requirements are considered (for example, the maximum acceleration change does not exceed 0.02m / s 2 ) to smooth the trajectory; finally, integrating safety requirements (for example, avoiding collisions and excessive operating forces) to generate optimized trajectory data that ensures that the robot arm’s movement is both precise and safe.

[0158] Step S56: Calculate the motion control parameters of each joint of the robot arm according to the optimized trajectory data, thereby obtaining a joint control matrix, wherein the motion control parameters include joint angle, angular velocity and angular acceleration;

[0159] The embodiment of the present invention uses the inverse kinematics method to calculate the motion control parameters of each joint of the robot arm according to the optimized trajectory data. By knowing the expected position of the end effector (for example, X'=0.125m, Y'=0.095m) and the expected posture, the iterative method is used to solve the angle, angular velocity and angular acceleration of each joint to ensure that the robot arm movement can achieve the optimized trajectory and meet the real-time control requirements. At this time, the motion control parameters of each joint (for example, joint 1 angle θ1=30°, joint 2 angle θ2=45°) are calculated.

[0160] Step S57: Performing dynamic constraint verification on the joint control matrix to obtain verification control data;

[0161] The embodiment of the present invention performs dynamic constraint verification on the calculated joint control matrix. The mechanical arm dynamics model is used to check whether the angle, velocity and acceleration of the joint meet the physical constraint conditions, such as whether the maximum load and acceleration of the joint exceed the safety limit. By calculating the torque requirement of each joint, it is ensured that the joint movement is within its load capacity, and the vibration that may be generated by the movement is evaluated to obtain verification control data.

[0162] Step S58: convert the verification control data into a robot arm control instruction sequence, and perform real-time control parameter updates to generate robot arm control data.

[0163] The embodiment of the present invention generates a robotic arm control instruction sequence based on the verification control data. The sequence includes control instructions for each joint of the robotic arm, such as joint 1 angle adjustment instructions, joint 2 speed control instructions, etc. The PID controller is used to make real-time adjustments to ensure that the joints perform corresponding actions according to the instructions, and is updated in real time according to the feedback signal to generate robotic arm control data to perform the predetermined task.

[0164] The present invention creates a preliminary motion model of the robot arm and performs safety constraints on its motion by combining dynamic compensation data and surgical safety control information. This model can effectively reflect the motion characteristics of the robot arm and provide a basis for subsequent precise control, ensuring that the motion of the robot arm during surgery meets safety and operation requirements. This process can monitor the motion state of the robot arm in real time, capture parameters such as the real-time position, speed and acceleration of the joints, and help accurately understand the specific state of the robot arm during operation. Through this state feature matrix, the system can provide necessary dynamic information for subsequent motion control and trajectory prediction, and improve the accuracy of motion control. Combining the state feature matrix with the core requirements of remote surgical control, a new model is established, which enables the motion trajectory of the robot arm to be predicted during surgery, and the control strategy is adjusted in real time according to the safety control sub-model. This multi-model integration method can not only improve the accuracy of surgery, but also maintain real-time monitoring and control of safety throughout the process. The system can predict the position changes of the robot arm in the future and verify whether the trajectory meets the safety standards according to the preset safety control requirements. This provides important data support for subsequent trajectory optimization and precise control, and ensures that the operation is always carried out within a safe range. By optimizing the trajectory, it is ensured that the robot arm can meet the requirements of operation accuracy, system stability and safety when performing tasks. With the optimized trajectory, surgical operations can be performed more accurately, while also reducing potential risks caused by inaccurate or too fast / slow trajectories, and improving the reliability of surgery. By calculating the specific motion parameters of each joint, such as angle, angular velocity and angular acceleration, it is ensured that the robot arm can accurately execute the required motion trajectory. The generation of the joint control matrix enables the robot arm to operate according to precise paths and motion requirements when performing complex surgeries, thereby ensuring the smooth progress of the surgical process. Through strict dynamic verification of the control matrix, it is ensured that all calculated joint control parameters meet the physical limitations and operation requirements of the robot arm. This step helps to avoid equipment damage or operation failure caused by control parameters exceeding the physical capability range, and ensures the safety and stability of system operation. By converting the verified control data into specific control instructions, the robot arm is driven to perform precise motion operations. At the same time, the real-time update of the control parameters can be flexibly adjusted in a dynamic surgical environment, so that the robot arm can always work in the best control state.

[0165] The present invention further provides a remote surgery control system based on the Internet of Things technology, which is used to execute the above-mentioned remote surgery control method based on the Internet of Things technology. The remote surgery control system based on the Internet of Things technology includes:

[0166] A multi-source data acquisition and calibration module is used to obtain multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system status data; set operating parameter thresholds according to the initial system status data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data;

[0167] The force feedback feature extraction module is used to perform force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; perform force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data;

[0168] The delay compensation processing module is used to identify motion delay according to the force feedback deviation data, and to perform real-time delay evaluation to generate delay feature evaluation data; perform data noise reduction processing on the system calibration data, and perform delay compensation processing through the delay feature evaluation data to generate dynamic compensation data;

[0169] The operation safety assessment module is used to identify the operation mode of the dynamic compensation data through the force feedback statistical feature data to generate the operation type feature data; to classify the operation risk of the operation type feature data to obtain the risk level data; to match the safety strategy of the risk level data to generate the operation safety control data;

[0170] The trajectory prediction control module is used to build a remote surgery control model based on dynamic compensation data and surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction analysis and generate robotic arm control data.

[0171] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0172] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A remote surgery control method based on Internet of Things technology, characterized in that: The following steps are involved: Step S1: Acquire multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system state data; set operating parameter thresholds according to the initial system state data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data; Step S2: performing force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; performing force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; performing difference analysis on the six-dimensional force control feature data to generate force feedback deviation data; Step S3: performing motion delay identification according to the force feedback deviation data, and performing real-time delay evaluation to generate delay feature evaluation data; Perform data noise reduction processing on the system calibration data, and perform delay compensation processing on the delay characteristic evaluation data to generate dynamic compensation data; Step S4: performing operation mode recognition on the dynamic compensation data through force feedback statistical feature data to generate operation type feature data; performing operation risk classification on the operation type feature data to obtain risk level data; Match risk level data with safety strategies to generate surgical safety control data; Step S5: construct a remote surgery control model based on the dynamic compensation data and the surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction and analysis to generate robotic arm control data.

2. The remote surgery control method based on Internet of Things technology according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire signal quality data of network transmission, including signal strength, network delay, data packet loss rate and bandwidth occupancy rate; Step S12: collecting the initial force data matrix of the six-dimensional force sensor at the end of the robotic arm, including the force data and torque data of the six-dimensional force sensor; Step S13: Collecting information of position encoders of each joint of the robot arm and position information of the end effector, and performing a coordinate system conversion to obtain position matrix data; merging signal quality data, initial force data matrix and position matrix data into multi-source sensor data; Step S14: performing network delay evaluation on multi-source sensor data based on average delay time calculation and delay jitter analysis, thereby obtaining delay feature vector data; Step S15: performing data integrity assessment on the multi-source sensor data to obtain an integrity assessment matrix, wherein the data integrity assessment includes a data packet sequence integrity check, data packet loss rate and retransmission rate calculation, and data synchronization performance assessment; Step S16: performing signal strength evaluation of the signal-to-noise ratio, bandwidth utilization, and signal stability index on the multi-source sensor data, thereby obtaining signal quality vector data; performing system state comprehensive score calculation based on the delay feature vector data, the integrity evaluation matrix, and the signal quality vector data, thereby obtaining initial system state data; Step S17: Calculate the system status comprehensive score according to the delay feature vector data, the integrity assessment matrix and the signal quality vector data, so as to obtain the initial system status data; Step S18: setting operating parameter thresholds according to the initial system state data, and performing system parameter calibration on the multi-source sensor data to obtain system calibration data, wherein the operating parameter thresholds include a network delay tolerance threshold, data integrity requirements, and a signal quality lower limit.

3. The remote surgery control method based on Internet of Things technology according to claim 2 is characterized in that: Step S2 includes the following steps: Step S21: performing time domain feature extraction on the initial force data matrix in the system calibration data to obtain an initial force feature vector, wherein the time domain feature extraction includes calculation of peak force, average force, torque peak value and torque mean; Step S22: performing frequency domain feature analysis based on force signal spectrum analysis, main frequency component extraction, and frequency band energy distribution calculation on the initial force data matrix in the system calibration data, thereby obtaining frequency domain feature matrix data; Step S23: performing feature fusion on the initial force feature vector and the frequency domain feature matrix data, and performing spatial mapping conversion according to the robot arm posture information, thereby generating six-dimensional force control feature data; Step S24: establishing a force feedback spatiotemporal distribution model based on the six-dimensional force control characteristic data, calculating the force feedback distribution characteristics at different operation stages, thereby obtaining force feedback statistical characteristic data, wherein the force feedback distribution characteristics include the spatial distribution density of the force, the time series variation law, and the torque distribution characteristics; Step S25: Perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data.

4. The remote surgery control method based on Internet of Things technology according to claim 3 is characterized in that: Step S25 includes the following steps: Step S251: performing historical data comparison and analysis on the six-dimensional force control characteristic data to obtain force feedback reference data, wherein the historical data includes historical records of force magnitude, force direction and torque characteristics; Step S252: performing real-time comparison between the six-dimensional force control characteristic data and the force feedback reference data, and calculating the force amplitude deviation, the force mean deviation and the force fluctuation deviation, so as to obtain the force magnitude deviation vector; Step S253: performing force direction deviation analysis according to the force magnitude deviation vector, calculating the force direction angle deviation, direction stability deviation and direction consistency deviation, thereby obtaining a force direction deviation matrix; Step S254: calculating the torque deviation including the torque amplitude deviation, the torque direction deviation and the torque balance deviation according to the force magnitude deviation vector, thereby obtaining torque deviation characteristic data; Step S255: performing statistical analysis on the force direction deviation matrix and the torque deviation characteristic data, including the average deviation rate, the maximum deviation value and the deviation fluctuation range, so as to obtain deviation statistical data; Step S256: setting a deviation threshold system including an average deviation threshold, a maximum deviation threshold and a fluctuation range threshold according to the deviation statistical data, thereby obtaining a deviation evaluation matrix; Step S257: performing abnormality discrimination processing according to the deviation assessment matrix, calculating abnormal characteristic parameters based on the degree of deviation exceeding the limit, the duration of deviation and the trend of deviation change, thereby obtaining an abnormal characteristic vector; Step S258: Perform feature fusion on the deviation evaluation matrix and the abnormal feature vector to obtain force feedback deviation data.

5. The remote surgery control method based on Internet of Things technology according to claim 4 is characterized in that: Step S3 includes the following steps: Step S31: performing time series analysis on the force feedback deviation data to extract the delay characteristics of the force feedback signal, thereby obtaining an initial delay characteristic vector, wherein the delay characteristics include signal transmission delay, system response delay and operation delay; Step S32: performing a real-time dynamic evaluation on the initial delay feature vector based on the delay fluctuation range, delay trend and delay stability index calculation, thereby obtaining a delay dynamic evaluation matrix; Step S33: performing delay impact assessment according to the delay dynamic assessment matrix, thereby obtaining delay feature assessment data, wherein the delay impact assessment includes an assessment of the impact of the delay on operation accuracy, system stability and security; Step S34: performing data noise reduction processing on the system calibration data, and performing delay compensation processing on the delay characteristic evaluation data to generate dynamic compensation data.

6. The remote surgery control method based on Internet of Things technology according to claim 5 is characterized in that: Step S34 includes the following steps: Step S341: performing wavelet transform analysis on the system calibration data to identify and filter high-frequency noise components, thereby obtaining noise reduction system calibration data; Step S342: establishing a delay compensation prediction model based on the delay characteristic evaluation data, and calculating optimal compensation parameters under different operation scenarios according to the noise reduction system calibration data, thereby obtaining optimal compensation parameter data; Step S343: adjusting the dynamic compensation parameters based on adaptive compensation according to the optimal compensation parameter data, thereby obtaining a compensation optimization matrix; Step S344: The compensation optimization matrix is ​​integrated with the noise reduction system calibration data, and real-time compensation control of the system delay is performed to generate dynamic compensation data.

7. The remote surgery control method based on Internet of Things technology according to claim 6 is characterized in that: Step S342 includes the following steps: Performing time series analysis on the delay feature evaluation data to extract delay pattern features, thereby obtaining a delay pattern matrix, wherein the delay pattern features include periodic delay features, burst delay features, and cumulative delay features; A surgical operation scene library is established based on the calibration data of the noise reduction system. The surgical operations in the surgical operation scene library are classified into three basic types: fine operation, fast operation and routine operation. The characteristic parameters of each type of operation are extracted to obtain the scene feature vector. Training a delay compensation prediction model based on a delay pattern matrix and a scene feature vector, wherein the delay compensation prediction model includes a delay prediction sub-model and a compensation amount prediction sub-model; Cross-validate the delay compensation prediction model under different operation scenarios, evaluate the model performance, and improve the prediction accuracy by optimizing the model parameters to obtain the optimized prediction model; Compensation parameters for various operation scenarios are calculated based on the optimization prediction model to obtain a compensation parameter set, where the compensation parameters include a time compensation coefficient, a spatial position compensation coefficient, and a force feedback compensation coefficient; Conduct stability analysis on the compensation parameter set, evaluate the reliability of the compensation effect, and establish a compensation parameter credibility evaluation index to obtain a parameter credibility matrix; The compensation parameter set is screened and optimized according to the parameter credibility matrix, and unreliable compensation parameters are eliminated to obtain the optimal compensation parameter data.

8. The remote surgery control method based on Internet of Things technology according to claim 7 is characterized in that: Step S4 includes the following steps: Step S41: performing operation mode recognition according to the force feedback statistical feature data and the dynamic compensation data, thereby generating an operation feature vector, wherein the operation mode recognition includes operation type classification and operation intensity evaluation; Step S42: identifying the current surgical operation type and operation parameters of the operation feature vector, thereby generating operation type feature data; Step S43: performing operation risk classification on the operation type feature data to obtain risk level data; Step S44: Perform real-time safety policy matching according to the preset surgical safety policy library and risk level data, and select a combination of safety control parameters to generate surgical safety control data.

9. The remote surgery control method based on Internet of Things technology according to claim 8, characterized in that: Step S5 includes the following steps: Step S51: establishing a robot arm motion state model according to the dynamic compensation data, and setting motion constraint conditions according to the surgical safety control data, thereby obtaining an initial motion control model; Step S52: performing real-time state parameter identification on the initial motion control model, extracting motion features including position state parameters, velocity state parameters and acceleration state parameters, thereby obtaining a state feature matrix; Step S53: establishing a remote surgery control model according to the state feature matrix, wherein the remote surgery control model includes a motion trajectory prediction sub-model and a safety control sub-model; Step S54: using the remote surgery control model to predict the position of the robot arm at the next moment, and performing safety constraint verification according to the surgery safety control data, so as to obtain a predicted trajectory vector; Step S55: performing trajectory optimization processing on the predicted trajectory vector based on the operation accuracy requirement, system stability requirement and safety requirement, so as to obtain optimized trajectory data; Step S56: Calculate the motion control parameters of each joint of the robot arm according to the optimized trajectory data, thereby obtaining a joint control matrix, wherein the motion control parameters include joint angle, angular velocity and angular acceleration; Step S57: Performing dynamic constraint verification on the joint control matrix to obtain verification control data; Step S58: convert the verification control data into a robot arm control instruction sequence, and perform real-time control parameter updates to generate robot arm control data.

10. A remote surgery control system based on Internet of Things technology, characterized in that: Used to execute the remote surgery control method based on the Internet of Things technology as claimed in claim 1, the remote surgery control system based on the Internet of Things technology comprises: A multi-source data acquisition and calibration module is used to obtain multi-source sensor data of the surgical system; perform communication quality evaluation on the multi-source sensor data based on network delay, data integrity and signal strength to obtain initial system status data; set operating parameter thresholds according to the initial system status data, and perform system parameter calibration on the multi-source sensor data to obtain system calibration data; The force feedback feature extraction module is used to perform force feedback feature analysis on the system calibration data to generate six-dimensional force control feature data; perform force feedback spatiotemporal distribution calculation based on the six-dimensional force control feature data to generate force feedback statistical feature data; perform difference analysis on the six-dimensional force control feature data to generate force feedback deviation data; The delay compensation processing module is used to identify motion delay according to the force feedback deviation data, and to perform real-time delay evaluation to generate delay feature evaluation data; perform data noise reduction processing on the system calibration data, and perform delay compensation processing through the delay feature evaluation data to generate dynamic compensation data; The operation safety assessment module is used to identify the operation mode of the dynamic compensation data through the force feedback statistical feature data to generate the operation type feature data; to classify the operation risk of the operation type feature data to obtain the risk level data; to match the safety strategy of the risk level data to generate the operation safety control data; The trajectory prediction control module is used to build a remote surgery control model based on dynamic compensation data and surgical safety control data; use the remote surgery control model to perform real-time operation trajectory prediction analysis and generate robotic arm control data.

Images