Intelligent control method and system for robot with body based on man-machine interaction data

Abstract

The invention provides an intelligent control method and system for a body-equipped robot based on man-machine interaction data, and relates to the technical field of medical robots, and the method comprises the steps: building a 5G network two-way transmission channel between a control end and a surgical robot end; through a 5G network two-way transmission channel, synchronously transmitting two types of data streams between the control end and the surgical robot end, namely a control instruction data stream generated by the control end and a surgical image data stream collected by the surgical robot end; in the transmission process of the two types of data streams, real-time network time delay is continuously monitored, whether the current network time delay exceeds a preset time delay threshold value or not is judged, and the preset time delay threshold value can be dynamically corrected according to the network state; and when it is judged that the current network delay exceeds the preset delay threshold, dynamically deploying a plurality of network probes on the 5G network two-way transmission channel. According to the invention, the real-time performance, the accuracy and the safety of the remote operation can be ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical robots, in particular to a somatic robot intelligent control method and system based on human-computer interaction data. BACKGROUND

[0002] With the continuous improvement of the demand for medical resource equalization, remote surgery robots have become a key technical means to solve the problem of lack of high-quality medical resources in remote areas due to their advantage of breaking through space limitations.

[0003] To meet the real-time requirements of remote surgery, the existing technology generally uses 5G network as a data transmission carrier, and establishes a bidirectional transmission channel between the control end and the surgery robot end to synchronously transmit expert control instructions and surgery image data. For example, a fixed network delay threshold is usually preset, and whether the real-time delay exceeds the threshold is monitored during transmission. If it exceeds the threshold, the data compression rate is adjusted or the transmission path is switched for simple processing.

[0004] The existing technology has the following technical defects: the 5G network delay is affected by many factors such as base station access user quantity fluctuation, core network gateway load change, edge computing node resource occupation difference, etc., showing dynamic and irregular fluctuation characteristics, and the fixed delay threshold cannot match the dynamic changes of network state in real time. For example, when the network delay temporarily rises but does not affect the operation of the surgery, the fixed threshold is easy to trigger unnecessary transmission parameter adjustment, causing interruption of control instruction transmission or lag of surgery image; when the network delay suddenly increases, the fixed threshold is also difficult to respond quickly, so that the control instruction and the surgery image are out of synchronization, the expert's control intention cannot be transmitted to the surgery robot in time, and the mechanical arm action lags behind the change of the surgery scene, which may cause the surgical instrument to touch the normal tissue by mistake, not only reducing the safety and operation precision of the remote surgery, but also limiting the popularization and application of the remote surgery robot in complex surgical operations. SUMMARY

[0005] The technical problem to be solved by the present application is to provide a somatic robot intelligent control system based on human-computer interaction data, which is suitable for remote surgery scenes that rely on low-latency and high-reliability data transmission, and can realize precise and real-time control of surgery robots by experts in different places.

[0006] To solve the above technical problems, the technical solutions of the present application are as follows: In a first aspect, a somatic robot intelligent control method based on human-computer interaction data, the method comprising: establishing a 5G network bidirectional transmission channel between the control end and the surgery robot end; Through the two-way transmission channel of the 5G network, two types of data streams are transmitted synchronously between the control end and the surgical robot end: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. During the transmission of the two types of data streams, the real-time network latency is continuously monitored, and it is determined whether the current network latency exceeds the preset latency threshold, which can be dynamically adjusted according to the network status. When the current network latency is determined to exceed the preset latency threshold, multiple network probes are dynamically deployed on the bidirectional transmission channel of the 5G network. The network probes are deployed at the base station access point, the core network gateway, and the edge computing node. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed. Calculate the area change rate of the polygonal time delay distribution map; form a network state adjustment factor based on the area change rate; dynamically correct the preset time delay threshold using the adjustment factor to obtain the optimized time delay threshold; Based on the numerical characteristics of the optimized delay threshold, the surgical image data stream and the control command data stream are processed in a hierarchical manner to obtain the hierarchical control command data stream. Based on the hierarchical transmission and processing of the control command data stream, a control command sequence matching the motion characteristics of the surgical robot arm is formed; the control command sequence drives the surgical robot arm to perform corresponding surgical operations.

[0007] Furthermore, a two-way 5G network transmission channel is established between the control terminal and the surgical robot terminal, including: The surgical robot initiates a connection establishment request to the network management platform; the connection establishment request includes the device identifier of the surgical robot and the required network service quality level. Based on the connection establishment request, the execution result of the connection establishment request is obtained; based on the execution result of the connection establishment request, the allocation process of dedicated network slice resources is triggered, and a logical connection path with the control end is established at the same time. Through the logical connection path, the surgical robot sends a channel test data packet to the control terminal, and after the transmission is completed, it initiates a confirmation channel data packet reception waiting mechanism. Based on the surgical robot, the round-trip time and packet loss rate of the channel test data packets are analyzed to obtain the analysis results. Based on the analysis results, it is used to verify whether the transmission performance of the logical connection path meets the requirements of surgical operation. Once the transmission performance is verified to meet the surgical requirements, a stable two-way transmission channel for the 5G network is established through collaborative confirmation between the surgical robot and the control unit.

[0008] Furthermore, through the 5G network's bidirectional transmission channel, two types of data streams are simultaneously transmitted between the control end and the surgical robot end: the control command data stream generated by the control end and the surgical image data stream acquired by the surgical robot end, including: The initial configuration result is obtained by performing the initial configuration of data stream transmission, that is, setting the working parameters of the surgical image acquisition device and determining the frame rate and resolution standard of the data to be transmitted. Based on the initial configuration results, the surgical image data acquired by the surgical robot is encapsulated in real time, and the control command data from the control terminal is parsed and its validity verified. Two types of standard data streams conforming to transmission specifications are formed; Based on the real-time requirements of surgical operations, priority is assigned to the two types of standard data streams: the control command data stream is marked as the highest transmission priority, and the surgical image data stream is marked as the second highest transmission priority. Through the two-way transmission channel of the 5G network, the two types of data streams are transmitted synchronously according to the set priority strategy, and the timing consistency between the data is maintained during the transmission process. The 5G network transmission quality is continuously monitored by the surgical robot to obtain monitoring feedback results. Based on the monitoring feedback results, the compression ratio parameters of the data stream are dynamically adjusted. The compression ratio parameters are a set of control parameters used to dynamically balance data volume and transmission quality to ensure the integrity and real-time transmission of critical surgical data.

[0009] Furthermore, during the transmission of both types of data streams, real-time network latency is continuously monitored, and it is determined whether the current network latency exceeds a preset latency threshold. This preset latency threshold can be dynamically adjusted based on network conditions, including: Start the network latency monitoring mechanism, configure the monitoring period parameters and latency sampling frequency, and establish a latency data buffer to store historical latency data; After the monitoring mechanism is activated, based on the configured monitoring cycle parameters, the round-trip latency of the control command data stream and the one-way transmission latency of the surgical image data stream are collected synchronously during the data transmission process. The collected multi-channel latency data is preprocessed in real time, and the average latency value is calculated after removing abnormal latency sampling points; the average latency value is used as the evaluation index of the current network latency. The current network latency evaluation metrics are compared and analyzed in real time with preset latency thresholds to obtain latency status judgment results. The corresponding processing flow is triggered based on the delay status judgment result. When the delay status judgment result exceeds the threshold, the dynamic adjustment mechanism is started; when the delay status judgment result does not exceed the threshold, the existing transmission strategy is maintained.

[0010] Furthermore, when the current network latency is determined to exceed a preset latency threshold, multiple network probes are dynamically deployed on the 5G network's bidirectional transmission channel. These network probes are deployed at base station access points, core network gateways, and edge computing nodes. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed, including: The network probe dynamic deployment process is triggered by the surgical robot to identify key deployment nodes in the 5G network transmission path. Key deployment nodes include base station access points, core network gateways and edge computing nodes. Based on the location information of the identified key deployment nodes, a probe deployment request is sent to the network management platform to activate the network probe function on the specified nodes. The network probe function collects real-time latency data of each key node and records the corresponding data collection timestamps to form a time-stamped latency dataset. Based on the latency dataset and the geometric positional relationship of each node, the base station access point, core network gateway, and edge computing node are used as vertices, and the vertices are connected in the order of the transmission path to construct a dynamic polygon latency distribution map. Topological verification is performed on the polygon time delay distribution map to ensure the closure of the polygon structure and the correctness of the vertex connections, thereby obtaining an effective time delay distribution map that can be used to calculate the rate of change of area.

[0011] Furthermore, the area change rate of the polygonal time delay distribution map is calculated; a network state adjustment factor is formed based on the area change rate; and the preset time delay threshold is dynamically corrected using the adjustment factor to obtain an optimized time delay threshold, including: Obtain the vertex coordinate sequence of the effective time delay distribution map; divide the polygonal time delay distribution map into multiple triangular sub-regions based on the vertex coordinate sequence; The area of ​​each triangular sub-region is calculated based on the triangular face accumulation method. That is, a common vertex of the polygon is selected; the common vertex is connected to the remaining non-adjacent vertices to form a triangular division, and the area value of each division triangle is calculated to obtain the area value of each triangular sub-region. The area values ​​of each triangular sub-region are summed to obtain the current total area of ​​the polygon time delay distribution map; at the same time, the total area of ​​the polygon time delay distribution map at the previous moment is obtained from the time delay data buffer, that is, the historical total area. Calculate the ratio of the difference between the current total area and the historical total area; use this ratio as the area change rate of the polygon time-delay distribution map; A network state adjustment factor is generated based on the magnitude of the area change rate; the preset delay threshold is proportionally corrected by the adjustment factor to form an optimized delay threshold suitable for the current network state.

[0012] Furthermore, based on the numerical characteristics of the optimized latency threshold, the surgical image data stream and the control command data stream are subjected to hierarchical transmission processing to obtain a hierarchically processed control command data stream, including: Analyze the numerical characteristics of the optimized latency threshold, and determine the hierarchical transmission strategy based on the different ranges of the numerical characteristics; Based on a hierarchical transmission strategy, the control command data stream is divided into critical command sub-streams and non-critical command sub-streams; at the same time, the surgical image data stream is subjected to multi-level compression processing to obtain a compressed image data stream. The transmission bandwidth is dynamically allocated based on network latency, that is, guaranteed bandwidth resources are allocated to critical instruction sub-streams, and shared bandwidth resources are allocated to non-critical instruction sub-streams and compressed image data streams. The data is transmitted in a hierarchical manner through a preset data scheduling mechanism, prioritizing the transmission of critical instruction sub-streams, and then transmitting non-critical instruction sub-streams and compressed image data streams in sequence. The transmission quality of each sub-stream is monitored in real time during data transmission to obtain monitoring results. Based on the monitoring results, the bandwidth allocation ratio is dynamically adjusted to form a control command data stream after hierarchical transmission processing.

[0013] Furthermore, based on the hierarchical transmission and processing of the control command data stream, a control command sequence matching the motion characteristics of the surgical robot arm is formed; the control command sequence drives the surgical robot arm to perform corresponding surgical operations, including: Key tissues and organs within the surgical area are abstracted into polygonal boundaries. An obstacle constraint model is then established based on these polygonal boundaries to determine the coordinates of the starting and target points of the robotic arm's movement. The motion trajectory is planned based on the robot obstacle avoidance path calculation method, and the optimal path is calculated according to the obstacle constraint model. The optimal path consists of a series of continuous line segments, that is, the optimal path sequence is formed to ensure that the path trajectory avoids all polygonal obstacle areas. The optimal path sequence is converted into joint motion parameters of the robotic arm, and inverse kinematics is solved based on the kinematics principle of the robotic arm to generate corresponding joint angle control commands. Smoothness optimization is performed by using joint angle control commands to eliminate abrupt changes in the motion trajectory and form a control command sequence that matches the motion characteristics of the surgical robot arm. The surgical robot arm is driven to perform surgical operations by a sequence of control commands. The positional accuracy of the end effector of the robotic arm is monitored in real time during the movement to obtain the monitoring results. The control command parameters are then fine-tuned based on the monitoring results.

[0014] Secondly, the embodied robot intelligent control system based on human-computer interaction data includes: A module is established to create a two-way 5G network transmission channel between the control unit and the surgical robot. The transmission module is used to synchronously transmit two types of data streams between the control end and the surgical robot end through the two-way transmission channel of the 5G network: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. The monitoring module is used to continuously monitor real-time network latency during the transmission of two types of data streams and determine whether the current network latency exceeds a preset latency threshold, wherein the preset latency threshold can be dynamically adjusted according to the network status. The deployment module is used to dynamically deploy multiple network probes on the bidirectional transmission channel of the 5G network when the current network latency is determined to exceed a preset latency threshold. The network probes are deployed at the base station access point, the core network gateway, and the edge computing node. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed. The calculation module is used to calculate the area change rate of the polygon delay distribution map; to form a network state adjustment factor based on the area change rate; and to dynamically correct the preset delay threshold through the adjustment factor to obtain the optimized delay threshold. The processing module is used to perform hierarchical transmission processing on the surgical image data stream and the control command data stream according to the numerical characteristics of the optimized delay threshold, so as to obtain the hierarchical transmission processing control command data stream. The execution module is used to form a sequence of control commands that matches the motion characteristics of the surgical robot arm based on the hierarchical transmission and processing of the control command data stream; and to drive the surgical robot arm to perform corresponding surgical operations through the sequence of control commands.

[0015] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0016] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0017] The above-described solution of the present invention has at least the following beneficial effects: By employing methods such as dynamically deploying network probes at key nodes in the 5G network transmission path, constructing dynamic polygonal latency distribution maps based on probe data, generating network state adjustment factors through calculating area change rates to dynamically correct latency thresholds, and combining data stream hierarchical transmission strategies with command sequence optimization techniques adapted to robotic arm motion characteristics, this approach effectively overcomes the core deficiency of existing fixed latency thresholds that cannot match the dynamic fluctuations of 5G networks in real time. It avoids unnecessary adjustments to transmission parameters or delayed responses that could lead to interruptions in control command transmission, surgical image lag, and data asynchrony, thereby ensuring the real-time performance and stability of remote surgical data transmission, improving the operational accuracy of the surgical robot's robotic arm, reducing the risk of surgical instruments accidentally touching normal tissue, and effectively enhancing the safety and reliability of remote surgery. This provides strong technical support for the widespread application of complex remote surgical procedures. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the intelligent control method for embodied robots based on human-computer interaction data provided in an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of an embodied robot intelligent control system based on human-computer interaction data provided in an embodiment of the present invention. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0021] like Figure 1 As shown, embodiments of the present invention propose an intelligent control method for embodied robots based on human-computer interaction data, the method comprising the following steps: Step 1, Step 1: Establish a two-way 5G network transmission channel between the control terminal and the surgical robot terminal; Step 2: Through the two-way transmission channel of the 5G network, two types of data streams are transmitted synchronously between the control end and the surgical robot end: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. Step 3: During the transmission of the two types of data streams, continuously monitor the real-time network latency and determine whether the current network latency exceeds the preset latency threshold, wherein the preset latency threshold can be dynamically adjusted according to the network status. Step 4: When the current network latency is determined to exceed the preset latency threshold, multiple network probes are dynamically deployed on the bidirectional transmission channel of the 5G network. The network probes are deployed at the base station access point, the core network gateway, and the edge computing node, respectively. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed. Step 5: Calculate the area change rate of the polygon delay distribution map; form a network state adjustment factor based on the area change rate; dynamically correct the preset delay threshold using the adjustment factor to obtain the optimized delay threshold; Step 6: Based on the numerical characteristics of the optimized delay threshold, the surgical image data stream and the control command data stream are subjected to hierarchical transmission processing to obtain the hierarchically processed control command data stream. Step 7: Based on the hierarchical transmission and processing of the control command data stream, a control command sequence matching the motion characteristics of the surgical robot arm is formed; the surgical robot arm is driven to perform corresponding surgical operations through the control command sequence.

[0022] In this embodiment of the invention, by establishing a stable two-way transmission channel in the 5G network, synchronous transmission of control commands and surgical image data streams is achieved, ensuring the basic stability of data transmission. By leveraging network probes dynamically deployed at base station access points, core network gateways, and edge computing nodes, combined with a dynamically constructed polygonal latency distribution map, the dynamic characteristics of network latency changes can be comprehensively and accurately captured. A network state adjustment factor is generated by calculating the area change rate of the polygonal latency distribution map, dynamically correcting the preset latency threshold. This flexibly adapts to the fluctuating state of the 5G network, avoiding invalid transmission adjustments or response lag issues caused by fixed thresholds. Data stream hierarchical transmission processing based on optimized latency thresholds ensures priority transmission and transmission quality of critical control commands, balancing the transmission needs of different types of data. Finally, by generating a control command sequence that matches the motion characteristics of the surgical robot's robotic arm, the accuracy and real-time performance of the robotic arm operation are effectively improved, surgical operation risks are reduced, and the safety and reliability of remote surgery are enhanced, providing strong support for the widespread application of remote surgical robots in complex surgical scenarios.

[0023] In a preferred embodiment of the present invention, step 1 above may include: Step 1.1: The surgical robot initiates a connection establishment request to the network management platform. This request includes the surgical robot's device identifier and the required network service quality level. Specifically, the surgical robot automatically integrates its unique hardware identifier, the medical institution's registration code, and the type code of the surgery to be performed, forming a unique and traceable device identifier system. This system ensures the network management platform accurately identifies the compliance of the surgical entity. Based on the clinical operating procedures corresponding to the current surgical type, and considering the real-time response requirements for control command transmission and the clarity requirements for surgical image transmission, the system clearly defines the upper limit threshold for network transmission latency, the minimum guaranteed bandwidth, and the maximum allowable packet loss rate, constructing network service quality level parameters that meet clinical surgical safety requirements. The integrated device identifier and the set network service quality level parameters are then structurally encapsulated and sent to the network management platform configured for remote medical services in this region, according to the standard format of the 5G network communication protocol.

[0024] Step 1.2: Based on the connection establishment request, obtain the execution result of the connection establishment request; trigger the allocation process of dedicated network slice resources based on the execution result of the connection establishment request, and simultaneously establish a logical connection path with the control end. Specifically, this includes: after receiving the structured encapsulated connection establishment request, firstly performing layered verification on the legality of the hardware identifier in the device identifier, the validity of the medical institution registration code, and the compliance of the surgical type code to confirm that the surgical robot has legitimate network access permissions; based on the network service quality level parameters specified in the request, calling the network resource scheduling algorithm to retrieve key resource statuses such as idle bandwidth, latency redundancy, and transmission link stability in the network resource pool; when available resources that meet the latency upper limit threshold, minimum guaranteed bandwidth, and maximum allowable packet loss rate requirements are retrieved, a connection establishment success execution result is generated, and the dynamic allocation process of the dedicated network slice is triggered; an independent transmission resource slice is allocated for this remote surgical scenario to achieve resource isolation from public network services and other non-emergency medical services, avoiding transmission performance fluctuations caused by cross-service resource competition; simultaneously, under the centralized scheduling of the 5G core network, an encrypted point-to-point logical transmission path is constructed between the surgical robot and the control end based on the IPsec protocol.

[0025] Step 1.3: Through the logical connection path, the surgical robot sends channel test data packets to the control terminal, and after the transmission is completed, an acknowledgment waiting mechanism for receiving the channel data packets is initiated. Specifically, this includes: after the logical transmission path is constructed, generating fixed-length test data blocks according to the standard unit size of remote surgical data transmission, adding a millisecond-precision transmission timestamp to each data block, and calculating the cyclic redundancy check code of the data blocks to form standardized channel test data packets; based on the load balancing principle of network transmission, setting equal data packet transmission intervals to avoid link congestion caused by sending a large number of data packets in a short period of time, and continuously sending multiple sets of standardized test data packets to the control terminal; after each set of test data packets is sent, immediately initiating the acknowledgment waiting mechanism, setting a reasonable timeout threshold based on the initial transmission delay requirements allowed by the surgical operation, which ensures sufficient acknowledgment waiting time while avoiding affecting the subsequent process due to excessive waiting time; if no acknowledgment feedback is received from the control terminal within the timeout threshold range, the retransmission process of that set of test data packets is automatically initiated.

[0026] Step 1.4: Based on the surgical robot, analyze the round-trip time and packet loss rate of the channel test data packets to obtain the analysis results. Based on these results, verify whether the transmission performance of the logical connection path meets the requirements of the surgical operation. Specifically, this includes: after receiving the reception confirmation information returned by the control terminal, extracting the sending timestamp and reception confirmation timestamp of each group of test data packets, calculating the round-trip time of each group of data packets using the timestamp difference, accumulating the round-trip time values ​​of multiple groups of data packets, and calculating the average round-trip time using the arithmetic mean method; counting the total number of all sent test data packets and the number of data packets for which no reception confirmation feedback was received, calculating the overall packet loss rate of the channel using the ratio of the number of unacknowledged data packets to the total number of sent packets; quantitatively comparing the calculated average round-trip time with the preset delay threshold for the surgical operation, and simultaneously performing threshold verification between the calculated overall packet loss rate and the preset maximum allowable packet loss rate; combining the two comparison results, if the average round-trip time is lower than the preset delay threshold and the packet loss rate is lower than the maximum allowable packet loss rate, then the real-time transmission and data integrity of the logical connection path are determined to meet the requirements of remote surgery; otherwise, the transmission performance is determined to be substandard.

[0027] Step 1.5: When the transmission performance is verified to meet the surgical requirements, a stable two-way transmission channel of the 5G network is established through collaborative confirmation between the surgical robot and the control terminal. Specifically, this includes: when the verification results show that the average round-trip latency of the logical connection path is lower than the latency threshold required by the surgery, and the packet loss rate is lower than the maximum allowable packet loss rate, i.e., the transmission performance meets the surgical requirements, the transmission performance verification result is encrypted and sent to the control terminal through the logical connection path; after receiving the result, the control terminal displays key transmission performance parameters on the operation interface. After the remote expert confirms that the transmission performance meets the surgical operation expectations, a collaborative confirmation command is sent to the surgical robot; after receiving the collaborative confirmation command, the surgical robot sends a channel activation request to the network management platform again. This request contains the complete device identification information of the surgical robot, including a unique hardware identifier and the registration code of the affiliated medical institution. The system includes: the type code of the surgery to be performed; the associated identifier of the dedicated network slice, covering the slice allocation number, network resource reservation identifier, and 5G core network logical transmission path ID; key result data for transmission performance verification, specifically the average round-trip latency of the logical connection path, the actual calculated packet loss rate, and the transmission performance qualification result; the device identifier and collaborative confirmation instruction information of the control end, including the hardware identifier of the control end and the generation timestamp of the collaborative confirmation instruction; and the time window parameters for channel activation, specifying the effective time of the activation request and the expected duration. After receiving the channel activation request, the network management platform locks the allocated dedicated network slice resources based on the identifier information, officially activating the 5G network bidirectional transmission channel between the surgical robot and the control end, ensuring stable transmission performance during the surgery.

[0028] In this embodiment of the invention, a connection establishment request containing device identification and network service quality level is initiated by the surgical robot, enabling the network management platform to accurately identify the device and match the exclusive transmission requirements of the surgical scenario. By triggering the allocation of dedicated network slice resources, a dedicated transmission path with strong isolation and stable bandwidth is provided for remote surgical data transmission, avoiding transmission interference caused by competition for public network resources. Through specialized analysis of the transmission and round-trip latency and packet loss rate of channel test data packets, the transmission performance of the logical connection path can be accurately verified to meet the stringent requirements of real-time performance and reliability for surgical operations. Finally, through collaborative confirmation between the surgical robot and the control terminal, a dual verification of the transmission channel stability is achieved, ensuring that the established 5G network bidirectional transmission channel has the core characteristics of low latency, low packet loss, and high stability. This lays a solid network foundation for the synchronous transmission of control command data streams and surgical image data streams, ensuring the continuity and accuracy of remote surgical control from the source.

[0029] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1 involves initializing the data stream transmission configuration, i.e., setting the operating parameters of the surgical image acquisition device and determining the frame rate and resolution standards of the data to be transmitted, to obtain the initialization configuration result. Specifically, this includes: setting the operating parameters of the surgical image acquisition device according to the type of surgery being performed, such as the need for detailed vascular observation in laparoscopic minimally invasive surgery or the requirement for bone structure imaging in orthopedic surgery; adjusting the sensitivity parameters according to the lighting conditions of the surgical area; reducing sensitivity to reduce noise when the surgical light is bright enough, and appropriately increasing sensitivity to ensure image clarity when the lighting is dim; and controlling the exposure time within a specified range. At the microsecond level, motion blur caused by excessive exposure is avoided, while white balance is calibrated to restore the true colors of the surgical area tissue. Combining the preset bandwidth limit of the 5G network's bidirectional transmission channel with the real-time response requirements of surgical operations, the frame rate and resolution standards of the data to be transmitted are determined. For surgeries requiring rapid motion feedback, the frame rate is set to 30 frames per second to ensure the continuity of capturing robotic arm movements. For surgeries requiring high-definition observation, the resolution is 1080P to balance image detail and data transmission volume. All parameters are pre-adapted to the subsequent transmission protocol format, resulting in an initial configuration that is both compatible and targeted.

[0030] Step 2.2: Based on the initial configuration results, the surgical image data acquired by the surgical robot is encapsulated in real time, and the control command data from the control terminal is parsed and its validity verified. Two types of standard data streams conforming to transmission specifications are formed, specifically including: Based on the initial configuration results, a layered encapsulation structure is used to process continuous image frames acquired by the surgical robot in real time. The header of each image frame encapsulates a data identifier to distinguish image streams from different surgical stages, the middle encapsulates the original image data, and the tail adds a millisecond-precision timestamp and a cyclic redundancy check field. The timestamp ensures the traceability of transmission timing, and the check field is used by the receiving end to verify the integrity of the data. At the same time, according to the preset control command protocol format, the command data from the control end is split into fields to extract core parameters such as the direction of movement of the robotic arm, joint angles, and operating speed. Combined with the physical movement limits of the surgical robot and the safety boundaries of the surgical area, the validity of the parameter value range is verified, and erroneous commands that exceed the movement limits of the robotic arm or may touch normal tissue are eliminated. At the same time, the integrity of the command format is verified, and missing necessary fields are filled in. Finally, two types of standard data streams with a unified structure and conforming to transmission specifications are formed: a control command data stream and a surgical image data stream.

[0031] Step 2.3: Based on the real-time requirements of the surgical operation, priority is assigned to the two types of standard data streams. The control command data stream is marked as having the highest transmission priority, and the surgical image data stream is marked as having the second highest transmission priority. Specifically, this involves: combining the safety execution logic of remote surgery, deeply analyzing the impact weight of the two types of data streams on the surgical outcome; the control command data stream directly drives the real-time movements of the surgical robot's robotic arm, and if its transmission delay exceeds 50 milliseconds, it may cause the robotic arm's movements to become disconnected from the expert's operational intentions, leading to the risk of accidental contact with surgical instruments. Therefore, it is marked as having the highest transmission priority and is marked with a dedicated identifier to ensure priority response during transmission scheduling; the surgical image data stream provides visual feedback on the surgical area to the expert and is the basis for generating control commands. However, slight transmission delays, such as those within 100 milliseconds, will not significantly affect the expert's judgment and operation. Therefore, it is marked as having the second highest transmission priority and is distinguished using a differentiated identifier. The priority marking information is bound to the standard data streams for transmission, ensuring that critical data receives priority protection in network resource competition.

[0032] Step 2.4: Through the 5G network bidirectional transmission channel, the two types of data streams are transmitted synchronously according to the set priority strategy, and the timing consistency between the data is maintained during the transmission process. Specifically, this includes: relying on the 5G network bidirectional transmission channel, enabling a priority-based queue scheduling mechanism, allocating an independent transmission queue for the highest priority control command data stream, which enjoys bandwidth priority and avoids resource competition with other data streams, allocating a parallel transmission queue for the second highest priority surgical image data stream, and utilizing the multi-connectivity characteristics of the 5G network to achieve synchronous sending and receiving of the two types of data streams; before transmission, the clocks of the surgical robot and the control end are calibrated through the network time protocol, and a unified reference timestamp is added to the two types of data streams. After receiving the data, the receiving end first caches it in the corresponding priority buffer, sorts the data according to the timestamp, removes data with timing errors caused by network jitter, and then processes it sequentially according to time order to ensure that the control commands issued by the expert match the corresponding surgical images in real time, maintain the timing consistency between the data, and ensure the precise coordination of operation actions and visual feedback.

[0033] Step 2.5: Continuously monitor the 5G network transmission quality via the surgical robot to obtain monitoring feedback results. Based on these results, dynamically adjust the data stream compression ratio parameters. These compression ratio parameters are a set of control parameters used to dynamically balance data volume and transmission quality, ensuring the integrity and real-time performance of critical surgical data transmission. Specifically, this includes continuously acquiring real-time transmission indicators of the 5G network's bidirectional transmission channel at a sampling frequency of once every 10 milliseconds, including the round-trip latency of the control command data stream, the one-way transmission latency of the surgical image data stream, the data packet loss rate, and the current available bandwidth utilization, forming a continuous network transmission quality monitoring feedback sequence. Based on this feedback sequence, dynamically optimize the data stream compression ratio parameters using adaptive adjustment calculations. The compression ratio parameter range... The compression ratio is set to 10% to 80%. When the transmission latency is detected to be lower than the preset threshold, the packet loss rate is less than 0.1%, and the bandwidth utilization is less than 60%, a low compression ratio of 10% to 30% is adopted. Lossless compression calculation is performed to preserve the subtle texture of the surgical image and the complete parameters of the control command. When the transmission latency is detected to increase, the packet loss rate exceeds 0.5%, or the bandwidth utilization is higher than 85%, the compression ratio is gradually adjusted to 50% to 80%. Lossy compression calculation is used to reduce the data volume within an acceptable range. At the same time, the step size of the compression ratio adjustment is set not to exceed 10% to avoid drastic changes in image quality or loss of command information caused by parameter mutations. The data volume and transmission quality are dynamically balanced to ensure that the critical surgical data can still maintain the integrity and real-time transmission when the network status fluctuates.

[0034] In this embodiment of the invention, the surgical image acquisition parameters and the frame rate and resolution standards for data transmission are clearly defined through initial configuration, establishing a unified benchmark for data transmission and avoiding transmission chaos or data distortion caused by parameter mismatch. Simultaneously, surgical image data is encapsulated in real time, and control command data is parsed and validated, transforming non-standardized data into a standard data stream conforming to transmission specifications and filtering invalid commands to ensure transmission effectiveness and security. Furthermore, data stream priorities are divided based on the real-time requirements of the surgery, with the control command data stream marked as the highest priority to ensure priority processing in transmission resource contention and avoid operational delays. Then, based on the priority strategy, the two types of data streams are transmitted synchronously while maintaining timing consistency, ensuring synchronized transmission of control commands and surgical images and improving the coordination of remote surgical operations. Finally, by continuously monitoring network transmission quality and dynamically adjusting data stream compression ratio parameters, data volume and transmission quality are flexibly balanced when network conditions fluctuate, avoiding image lag or control command transmission interruptions, comprehensively ensuring the integrity and real-time nature of critical surgical data transmission.

[0035] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: Activate the network latency monitoring mechanism, configure the monitoring period parameters and latency sampling frequency, and establish a latency data buffer to store historical latency data. Specifically, this includes: activating the network latency monitoring mechanism based on the dynamic fluctuation characteristics of 5G network latency affected by factors such as base station load and gateway pressure, and the millisecond-level latency response requirements of remote surgery; setting the monitoring period parameter to 10 milliseconds, which is shorter than the minimum latency redundancy allowed clinically, while avoiding excessive network resource consumption due to an excessively short period; setting the latency sampling frequency to 10 times per monitoring period to ensure high-density capture of subtle latency changes; simultaneously establishing a first-in-first-out circular latency data buffer, with a buffer capacity planned to store latency data from the most recent 100 complete monitoring periods. Each data set includes the original value of the round-trip latency of control commands, the original value of the one-way transmission latency of surgical images, the corresponding collection timestamp, and the data stream type identifier. This historical data will provide a foundation for subsequent anomaly identification and latency trend analysis.

[0036] Step 3.2: After the monitoring mechanism is activated, based on the configured monitoring cycle parameters, the round-trip latency of the control command data stream and the one-way transmission latency of the surgical image data stream are collected synchronously during data transmission. Specifically, after the monitoring mechanism is activated, latency collection is carried out synchronously during the bidirectional transmission of the control command data stream and the surgical image data stream, strictly following a 10-millisecond monitoring cycle. For the control command data stream, the initial timestamp is recorded the instant the command is issued from the control end, and the termination timestamp is recorded when the surgical robot fully receives the command and returns confirmation information containing the command verification result. The difference in timestamps represents the round-trip latency of the control command data stream, directly reflecting the efficiency of the complete link for command transmission and feedback. For surgical image data streams, the sending timestamp is recorded the instant the image frame is encapsulated and sent from the surgical robot. When the control terminal successfully receives the image frame and completes data verification, the receiving timestamp is recorded. The difference between the two timestamps represents the one-way transmission latency of the surgical image data stream, reflecting the real-time nature of visual feedback. Simultaneous acquisition of both types of latency data can comprehensively cover the uplink and downlink status of the bidirectional transmission channel, avoiding network status misjudgments caused by single-direction latency monitoring.

[0037] Step 3.3 involves real-time preprocessing of the collected multi-channel latency data. This involves calculating the average latency value after removing abnormal latency sampling points. The average latency value is used as an evaluation indicator for the current network latency. Specifically, this includes performing real-time preprocessing on 10 sets of round-trip latency data for control commands and 10 sets of one-way transmission latency data for surgical images collected in each monitoring cycle. First, based on the normality of the data distribution, abnormal latency sampling points are identified using the 3σ principle. The mean and standard deviation of the latency data for each data stream are calculated, and values ​​exceeding the mean ± 3 times the standard deviation are removed. These outliers are typically caused by network instantaneous latency. Jitter and sudden interference can reduce the impact of extreme data on the evaluation results by removing them. Then, the average values ​​of the remaining valid data are calculated to obtain the average round-trip delay of control commands and the average one-way transmission delay of surgical images. Considering that the control command data stream has the highest transmission priority and a higher weight in terms of its impact on surgical safety, it is assigned a weight of 0.6, and the surgical image data stream is assigned a weight of 0.4. The comprehensive average delay value is obtained by weighted summation. This value takes into account the transmission characteristics of the two types of data streams and can objectively reflect the overall state of the current network latency. Therefore, it is determined as the core evaluation indicator of the current network latency.

[0038] Step 3.4 involves comparing and analyzing the current network latency evaluation metrics with the preset latency threshold in real time to obtain the latency status judgment result. Specifically, the preset latency threshold is set based on the clinical operating procedures for the current surgery (e.g., the safety requirements for latency of different surgical procedures in the remote surgical technique operation guidelines) and the physical response limits of the surgical robot's robotic arm: For neurosurgical microsurgery, due to its extremely high precision requirements (millimeter-level positioning), the preset latency threshold is set to 40 milliseconds; for laparoscopic minimally invasive surgery, which needs to balance operational continuity and visual clarity, the preset latency threshold is set to 50 milliseconds; for orthopedic implant surgery, due to the larger range of instrument movements, the threshold can be relaxed to 60 milliseconds. Milliseconds; Referring to current clinical operating procedures for performing surgery and considering the response limits of the surgical robot's robotic arm, a preset latency threshold is determined. For example, the preset latency threshold for laparoscopic minimally invasive surgery is set at 50 milliseconds. This value has been clinically verified to ensure that exceeding it may cause lag in the robotic arm's movements. This threshold is extracted from a preset parameter library and compared in real time with the calculated current network latency evaluation index. The comparison process adopts a cycle-by-cycle real-time verification mode. After each monitoring cycle, the evaluation index is immediately compared with the threshold. If the comprehensive average latency value is less than or equal to 50 milliseconds, the latency status is determined to be within the threshold. If the comprehensive average latency value is greater than 50 milliseconds, the latency status is determined to be beyond the threshold. The resulting latency status judgment result will be transmitted in real time in the form of a status identifier to ensure rapid response to changes in network latency.

[0039] Step 3.5: Based on the latency status judgment result, trigger the corresponding processing flow. When the latency status judgment result exceeds the threshold, activate the dynamic adjustment mechanism; when the latency status judgment result does not exceed the threshold, maintain the existing transmission strategy, specifically including: receiving the transmitted latency status judgment result and triggering a differentiated processing flow based on the result; when the latency status judgment result exceeds the threshold, immediately activate the dynamic adjustment mechanism, which will link subsequent network probe deployment, dynamic correction of latency threshold, and hierarchical data stream transmission to specifically address the problem of increased network latency, avoiding continuous latency exceeding the standard, which could lead to asynchrony between control commands and images, thereby causing surgical risks; when the latency status judgment result does not exceed the threshold, maintain the existing transmission strategy, including the current data stream priority configuration, dynamic compression ratio parameters, and transmission queue scheduling rules, without making additional parameter adjustments. This design can avoid the interruption of control command transmission or image stuttering caused by accidental triggering of adjustments due to fixed thresholds, as in existing technologies, ensuring the continuity and smoothness of surgical data transmission when the network status is stable.

[0040] In this embodiment of the invention, a network latency monitoring mechanism is activated, with reasonable monitoring period parameters and latency sampling frequency configured. A latency data buffer is established to store historical latency data, providing standardized support for latency assessment. After the monitoring mechanism is activated, the round-trip latency of the control command data stream and the one-way transmission latency of the surgical image data stream are collected synchronously, comprehensively reflecting the true performance of the 5G network's bidirectional transmission channel and avoiding misjudgments of network status caused by single latency data. The collected multi-channel latency data is preprocessed in real time. After removing abnormal latency sampling points, the average latency value is calculated and used as an evaluation indicator of the current network latency, effectively improving the accuracy of latency assessment. This evaluation indicator is compared and analyzed in real time with a preset latency threshold to quickly obtain the latency status judgment result, ensuring timely response to network latency changes. Based on the judgment result, a corresponding processing flow is triggered. When the latency exceeds the threshold, a dynamic adjustment mechanism is activated; when it does not exceed the threshold, the existing transmission strategy is maintained, achieving flexible adaptation to network status and ensuring the stable and real-time transmission of remote surgical data.

[0041] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1 involves triggering a dynamic network probe deployment process via the surgical robot to identify key deployment nodes in the 5G network transmission path. These key nodes include base station access points, core network gateways, and edge computing nodes. Specifically, when network latency exceeds a preset threshold, the surgical robot automatically triggers the dynamic network probe deployment process based on network topology awareness. This process incorporates optimized path detection logic, sending low-overhead probe packets to the 5G network transmission link to track the forwarding trajectory of data packets. During transmission, the probe packets record the response time and identification information of the nodes along the route. Combining this with the link characteristics of remote surgical data transmission, key deployment nodes that have a core impact on latency fluctuations are selected. The base station access point, as the first relay link for data entering and leaving the network, has its signal strength and the number of accessing users directly affecting the initial transmission efficiency. The core network gateway bears the core responsibility of routing and forwarding, and load changes can cause forwarding latency fluctuations. Edge computing nodes are responsible for processing surgical data locally, and their resource occupancy status is directly related to data processing latency. These three nodes together constitute the core links of the latency-sensitive link. Finally, through analysis of node response characteristics and latency contribution, the unique network identifier and topology location of each key deployment node are determined.

[0042] Step 4.2: Based on the location information of the identified key deployment nodes, a probe deployment request is sent to the network management platform to activate the network probe function on the designated node. This includes: collecting precise network topology and physical location parameters of the identified key deployment nodes, including latitude and longitude coordinates, network hierarchy, and device port identifiers; clarifying the network probe's collection requirements, including latency, data type, collection frequency, and data transmission format; structuring and integrating this information with the probe deployment request according to the standardized communication protocols supported by the network management platform, ensuring the request content is complete and conforms to the platform's parsing specifications; after sending the request to the network management platform, verifying the network access status of the key nodes to confirm they are in normal working order, and then checking the node's CPU utilization, memory usage, bandwidth availability, and other resource conditions to ensure the node has redundant resources to support the probe function; after successful verification, issuing a precise activation command to the designated node, which triggers the node's built-in probe module and transmission module to work together, embedding the probe function into the existing data transmission process without consuming excessive network resources, ultimately entering a real-time data collection ready state.

[0043] Step 4.3: Collect real-time latency data of each key node using the network probe function, and record the corresponding data collection timestamps to form a time-stamped latency dataset. Specifically, the activated network probe follows the time-division multiplexing acquisition principle, accurately capturing multi-dimensional latency data of each key node at a high-frequency acquisition frequency of once every 10 milliseconds. Inbound latency records the time taken from receiving data from the node port to entering the processing queue; outbound latency records the time taken from processing completion to leaving the node port; forwarding latency records the forwarding time taken after routing decisions and data verification within the node. During the acquisition process, the network time protocol is used to maintain millisecond-level synchronization with the clock system of the surgical robot, adding accurate acquisition timestamps to each set of acquired latency data to ensure time sequence consistency. The node's unique identifier, inbound latency, outbound latency, forwarding latency, and corresponding timestamps are associated and bound, and organized in chronological order to form a structured time-stamped latency dataset. The dataset also includes a data validity identifier to distinguish between normal acquired data and abnormal interference data.

[0044] Step 4.4: Based on the latency dataset and the geometric positional relationships of each node, the base station access point, core network gateway, and edge computing node are used as vertices, and the vertices are connected in sequence according to the transmission path to construct a dynamic polygon latency distribution map. Specifically, this includes: establishing a two-dimensional network topology coordinate system with the physical location of the surgical robot as the origin; calculating the precise coordinates of the nodes in the coordinate system based on the latitude and longitude coordinates of each key deployment node and the network hierarchy, ensuring that the positional relationships are consistent with the actual 5G network transmission links; extracting the latest effective latency data from the latency dataset; and using the base station access point, core network gateway, and edge computing node as independent vertices, with each vertex simultaneously carrying... Location information and corresponding real-time total latency values ​​are generated. The total latency is obtained by summing the inbound latency, outbound latency, and forwarding latency. According to the actual data transmission path order, the data sent by the surgical robot enters the network through the base station access point, is then transmitted to the edge computing node for local processing, and is subsequently routed to the control end through the core network gateway. Each vertex is connected by line segments in sequence. The length of the line segment is positively correlated with the transmission latency between the corresponding nodes. The greater the latency, the longer the line segment. As the latency data of each node is dynamically updated, the length of the line segment will be adjusted synchronously, and the overall shape of the polygon will also change in real time. Finally, a dynamic polygon latency distribution map that can intuitively reflect the link latency distribution and dynamic fluctuations is constructed.

[0045] Step 4.5 involves topological verification of the polygonal delay distribution map to ensure the closure of the polygonal structure and the correctness of vertex connections, resulting in an effective delay distribution map that can be used for area change rate calculation. Specifically, this includes: using a verification method based on network topology map comparison to perform comprehensive topological verification of the constructed polygonal delay distribution map. First, it verifies whether all key deployment nodes are completely included in the distribution map according to the actual transmission path, ensuring no vertices are missing or redundant / invalid. Next, it verifies the closure of the polygonal structure, confirming that the starting vertex (base station access point) and the ending vertex (core network gateway link node returned from the control terminal) can form an effective connection, conforming to the round-trip link characteristics of data transmission, avoiding the problem of non-closed graphs caused by link breaks. Simultaneously, it verifies the connection relationships between each vertex, comparing them with the routing rules of the actual 5G network transmission path, eliminating topological deviations caused by incorrect node identification or disordered connection order. Furthermore, it verifies whether the delay data associated with each vertex is within a reasonable range, eliminating vertex deviations corresponding to abnormal data caused by acquisition interference. After multi-dimensional verification, an effective delay distribution map with a standardized structure, accurate topology, and reliable data is obtained.

[0046] In this embodiment of the invention, by triggering a dynamic deployment process for network probes, key deployment nodes such as base station access points, core network gateways, and edge computing nodes in the 5G network transmission path are accurately located, enabling targeted latency monitoring and avoiding resource waste and data redundancy caused by indiscriminate monitoring. Then, based on the location information of the key nodes, a probe deployment request is sent to the network management platform to achieve dynamic activation and on-demand deployment of the network probes, adapting to the differences in transmission paths in different remote surgical scenarios, ensuring precise binding between probes and key nodes, and improving the standardization and efficiency of deployment through network management platform scheduling. Subsequently, real-time latency data of each key node is collected through the network probes. Synchronous recording of data collection timestamps forms a time-stamped latency dataset, giving each data set a clear temporal attribute and accurately reflecting the latency status of each node at different times. Then, based on the latency dataset and the geometric positional relationship of the nodes, a dynamic polygon latency distribution map is constructed with key nodes as vertices and connected in sequence according to the transmission path. This transforms the scattered node latency data into a structured and visualized link latency configuration, intuitively presenting the latency distribution characteristics of the transmission link and helping to quickly locate latency bottleneck nodes. Finally, the polygon latency distribution map is topologically verified to ensure structural closure and the correctness of vertex connections, eliminating invalid distribution maps and ensuring the accuracy of area change rate calculation.

[0047] In a preferred embodiment of the present invention, step 5 above may include: Step 5.1: Obtain the vertex coordinate sequence of the effective latency distribution map; based on the vertex coordinate sequence, divide the polygon latency distribution map into multiple triangular sub-regions. Specifically, this includes: extracting the complete vertex coordinate sequence from the effective latency distribution map. This sequence strictly follows the order of data transmission paths and includes the precise coordinates of key deployment nodes such as base station access points, edge computing nodes, and core network gateways. All coordinate values ​​correspond to their actual positions in the two-dimensional network topology coordinate system. The optimized Delaunay triangulation calculation method is then activated to construct an initial vertex set based on this coordinate sequence. The initial vertex set is automatically constructed following the empty circle criterion, meaning that the circumcircle of any triangle will not contain other vertices, thus avoiding the generation of elongated or distorted triangles. During the partitioning process, the boundary vertices of the polygons are processed first, through pre-... Constraints are set to ensure that the edges of the generated triangles are either the boundary edges of the original polygon or the diagonals inside the polygon, and that all internal diagonals do not intersect. Specifically, starting from the first vertex of the coordinate sequence (base station access point), the positional relationship between each subsequent vertex and the existing triangle is checked in turn. First, the first vertex is connected to the third vertex to form an initial triangle. Then, based on the empty circle criterion, the position of the fourth and subsequent vertices is determined. Existing triangles are split or adjusted as needed, and new triangle units are added step by step until all vertices are completely incorporated into the triangle structure. Finally, multiple non-overlapping and seamlessly connected triangular sub-regions are formed, and their set completely covers all areas of the original polygon, without omitting any part or having any redundant sub-regions that exceed the range of the original polygon.

[0048] Step 5.2: Calculate the area of ​​each triangular sub-region based on the triangle face accumulation and addition calculation method, i.e., select a common vertex of the polygon; connect the common vertex with the remaining non-adjacent vertices to form a triangular division, and calculate the area value of each divided triangle to obtain the area value of each triangular sub-region. Specifically, this includes: when using the triangle face accumulation and addition calculation method, selecting the first vertex of the corresponding base station access point in the coordinate sequence as the common vertex. This selection is consistent with the starting point of the data transmission path to ensure that the division logic is consistent with the network transmission link; connecting the common vertex with other non-adjacent vertices in the sequence one by one, i.e., connecting the common vertex with the third vertex, the fifth vertex, etc., in sequence to form a regular triangular division structure. Each triangle consists of a common vertex and two non-adjacent subsequent vertices; for each divided triangle, based on the coordinate values ​​of its three vertices, calculate the actual length of the base and the corresponding height of the triangle by the difference between the horizontal and vertical coordinates, and combine the core calculation principle of triangle area in plane geometry to accurately solve the area value of each sub-region; to avoid error accumulation, all area values ​​are retained to three decimal places.

[0049] Step 5.3: Sum the area values ​​of each triangular sub-region to obtain the current total area of ​​the polygon delay distribution map; simultaneously, retrieve the total area of ​​the polygon delay distribution map from the previous moment, i.e., the historical total area, from the delay data buffer. This involves iterating through all the obtained triangular sub-region area values ​​and summing them sequentially according to the basic logic of numerical accumulation. The sum obtained is the current total area of ​​the polygon delay distribution map at the current moment. This total area indirectly reflects the overall delay level of the network link through geometric quantization. The larger the area, the higher the total delay of each key node, and the lower the network transmission efficiency. At the same time, according to the storage rules of the established delay data buffer, retrieve the data in descending order of timestamp and extract the total area of ​​the polygon delay distribution map stored in the previous monitoring period immediately adjacent to the current monitoring period, i.e., the historical total area; ensure that the time interval between the current total area and the historical total area is strictly equal to the preset 10-millisecond monitoring period, providing a unified and standard time series data pair for calculating the area change per unit time.

[0050] Step 5.4: Calculate the ratio of the difference between the current total area and the historical total area. This ratio is used as the area change rate of the polygonal latency distribution map. Specifically, this involves: first, calculating the absolute difference between the current total area and the historical total area. If the current total area is greater than the historical total area, the difference is positive, indicating an overall upward trend in network latency, potentially due to increased base station load or gateway congestion. If the current total area is less than the historical total area, the difference is negative, indicating an overall downward trend in network latency and an improving network transmission status. Dividing this absolute difference by the historical total area yields the ratio of the current total area to the historical total area. This ratio is formally defined as the area change rate of the polygonal latency distribution map. The absolute value of the area change rate directly corresponds to the fluctuation range of network latency. A larger absolute value indicates more drastic changes in network latency per unit time, while a smaller absolute value indicates a more stable network state. This indicator transforms the dynamic changes of the graph into quantifiable numerical values.

[0051] Step 5.5, generate a network state adjustment factor based on the value of the area change rate; use the adjustment factor to perform proportional correction on the preset delay threshold to form an optimized delay threshold applicable to the current network state, specifically including: based on the clinical surgical safety requirements and a large amount of network test data, establish a mapping relationship between the area change rate and the network state adjustment factor; when the absolute value of the area change rate is less than 5%, it indicates that the network state is in a stable range and will not affect surgical operations, and the adjustment factor is set to 0.05; when the absolute value of the area change rate is between 5% and 15%, it indicates that the network has medium fluctuations and the threshold needs to be slightly adjusted to adapt to the changes, and the adjustment factor is set to 0.1; when the absolute value of the area change rate is greater than 15%, it indicates that the network fluctuates violently and may cause the delay to exceed the standard, and the adjustment factor is set to 0.2 to strengthen the adaptation effect; multiply the adjustment factor by the preset delay threshold to obtain the threshold correction amount. If the area change rate is positive, it indicates that the network delay increases, and the optimized delay threshold is obtained by adding the correction amount to the preset delay threshold to avoid mis-triggering the adjustment; if the area change rate is negative, it indicates that the network delay decreases, and the optimized delay threshold is obtained by subtracting the correction amount from the preset delay threshold to maintain strict control over the delay. The optimized delay threshold can dynamically match the fluctuation state of the 5G network.

[0052] It should be noted that: The adjustment factor is a network state adaptation type proportional adjustment factor, which is specifically divided into three categories, corresponding to different network fluctuation scenarios. The value of the stable range adjustment factor is 0.05, which is applicable to the scenario where the absolute value of the area change rate is less than 5%. At this time, the network state is stable, the delay fluctuation is gentle, and it will not affect the real-time performance and safety of surgical operations; the value of the medium fluctuation adjustment factor is 0.1, which is applicable to the scenario where the absolute value of the area change rate is between 5% and 15%. At this time, the network has moderate fluctuations, the delay changes significantly but does not exceed the clinical safety bottom line, and the threshold needs to be slightly corrected to adapt to the changes; the value of the violent fluctuation adjustment factor is 0.2, which is applicable to the scenario where the absolute value of the area change rate is greater than 15%. At this time, the network state is unstable, the delay changes sharply, and if not adjusted in time, it may cause the control instructions and images to be out of sync, and the threshold correction intensity needs to be strengthened to ensure surgical safety.

[0053] The adjustment factors were generated based on the dynamic fluctuation characteristics of 5G networks and the clinical safety requirements of remote surgery, and were derived through extensive experimental data calibration and verification. First, a 5G network test environment simulating remote surgery was built, simulating real-world scenarios such as base station load, gateway pressure, and edge node resource consumption. Over 100,000 sets of sample data covering the entire network range were collected, including area change rate, surgical response status, and safety feedback. Based on clinical guidelines, the fluctuation impact boundaries were clearly defined, and the network state was divided into stable, moderately fluctuating, and severely fluctuating ranges, corresponding to risk incidence rates of less than 0.1%, 1%-3%, and more than 5%, respectively. Orthogonal experimental design was used to test the correction effects of different factor ratios in each range. Values ​​were calibrated by comparing indicators such as response speed and false trigger rate. For example, a value of 0.05 in the stable range balances flexibility and stability, 0.1 in the moderately fluctuating range provides the best fit, and 0.2 in the severely fluctuating range enables rapid response to extreme scenarios. Finally, after three months of clinical simulation and real-world application verification, the factor values ​​were fine-tuned to ensure accurate matching of network states and dynamic optimization of preset latency thresholds.

[0054] In this embodiment of the invention, by obtaining the vertex coordinate sequence of the effective time delay distribution map, the complex polygonal time delay distribution map is divided into multiple triangular sub-regions, transforming the calculation of the area of ​​irregular polygons into the calculation of the area of ​​regular triangles, thus reducing the solution complexity. A triangle partitioning method is adopted, where common vertices are connected to the remaining non-adjacent vertices, ensuring that the polygon is completely decomposed into independent sub-regions. By accumulating the areas of each triangular sub-region, the accuracy of the total area calculation is improved, and systematic errors are avoided. The current total area is obtained by accumulating the sub-region areas. The historical total area is retrieved from the time delay data buffer, establishing a temporal correlation between the current and historical network states, capturing the dynamic change trajectory of time delay, and providing data for calculating the area change rate. The system provides complete time-series data pairs; it calculates the ratio of the difference between the current and historical total area to obtain the area change rate, transforming the dynamic changes in polygon shape into a quantifiable indicator that accurately reflects the amplitude and trend of network latency fluctuations, solving the problem that existing technologies cannot quantify the degree of fluctuation; it generates a network state adjustment factor based on the area change rate, achieving precise binding between the adjustment factor and the latency fluctuation amplitude, and proportionally corrects the preset latency threshold through the adjustment factor, optimizing the fixed threshold into a dynamically optimized latency threshold that fits the current network state, effectively solving the defect that the fixed threshold cannot adapt to the dynamic fluctuations of 5G networks, avoiding unnecessary transmission adjustments when network fluctuations are small, and responding promptly when fluctuations are severe.

[0055] In a preferred embodiment of the present invention, step 6 above may include: Step 6.1: Analyze the numerical characteristics of the optimized latency threshold and determine the hierarchical transmission strategy based on the different ranges of the numerical characteristics. Specifically, this includes: after obtaining the optimized latency threshold, comprehensively analyzing its numerical characteristics, including the specific threshold value, the direction and magnitude of the difference from the historical optimized threshold, and the corresponding area change rate range, while also associating it with real-time status parameters such as the available bandwidth and packet loss rate of the current 5G network; combining remote surgery clinical safety guidelines and the robotic arm response limits, clarifying the criteria for dividing the three core strategy ranges: 30 milliseconds is the optimal latency upper limit for a lag-free robotic arm response, 50 milliseconds is the maximum safe latency threshold allowed clinically; when the optimized latency threshold is less than or equal to 30 milliseconds... For networks with excellent performance and an absolute area change rate of less than 5%, a Level 1 transmission strategy is established to maintain the priority of routine data transmission and the basic image compression rate. When the optimization latency threshold is between 30 and 50 milliseconds, corresponding to a moderate network performance and an absolute area change rate of 5% to 15%, a Level 2 transmission strategy is established to increase the priority of critical command transmission and moderately improve the image compression rate. When the optimization latency threshold is greater than 50 milliseconds, corresponding to a poor network performance and an absolute area change rate greater than 15%, a Level 3 transmission strategy is established to strengthen the bandwidth guarantee for critical commands and improve the image compression rate, ensuring that the tiered transmission strategy is accurately matched with network latency fluctuations and clinical safety requirements.

[0056] Step 6.2: Based on a hierarchical transmission strategy, the control command data stream is divided into critical command sub-streams and non-critical command sub-streams. Simultaneously, the surgical image data stream undergoes multi-level compression processing to obtain a compressed image data stream. Specifically, this includes: based on the hierarchical transmission strategy, the control command data stream is divided into sub-streams according to the degree of impact of the commands on surgical safety; the critical command sub-stream contains core commands that directly determine the accuracy of the surgical operation and patient safety, such as robotic arm start / stop, steering, force adjustment, and surgical instrument switching. These commands are filtered using built-in function identifiers and priority tags and marked as having the highest transmission priority; the non-critical command sub-stream contains equipment operating status... Auxiliary commands such as feedback, parameter query, and log recording, which do not affect real-time surgical operations, are marked as the second highest transmission priority. Simultaneously, differentiated multi-level compression processing is performed for different transmission strategies. Under the first-level strategy, the compression ratio is set to 1.5:1, balancing transmission efficiency and image quality. Under the second-level strategy, the compression ratio is set to 2:1, preserving core information of the surgical field while controlling bandwidth usage. Under the third-level strategy, the compression ratio is set to 3:1, intelligently identifying key areas such as surgical incisions and lesions, prioritizing the preservation of clarity in these areas, and moderately compressing non-critical areas to ensure that the compressed image data stream reduces bandwidth usage without affecting the expert's judgment of the surgical scenario.

[0057] Step 6.3: Dynamically allocate transmission bandwidth based on network latency conditions. This involves allocating guaranteed bandwidth resources for critical command sub-streams and shared bandwidth resources for non-critical command sub-streams and compressed image data streams. Specifically, this includes: real-time acquisition of network latency-related parameters such as the current total available bandwidth and bandwidth fluctuation range, and performing dynamic bandwidth allocation based on the sub-stream partitioning results. Guaranteed bandwidth resources are exclusively used by critical command sub-streams and are not shared with other sub-streams. Under the first-level strategy, 30% of the total available bandwidth is allocated; under the second-level strategy, this is increased to 40%; and under the third-level strategy, it is further increased to 50%, ensuring that regardless of network fluctuations... Critical commands can obtain stable transmission channels; shared bandwidth resources are used to carry non-critical command sub-streams and compressed image data streams, allocated at an initial ratio of 30% for non-critical command sub-streams and 70% for image data streams. This ratio is set based on the transmission requirements of the two types of data, with image data streams requiring more bandwidth due to their larger data volume; the total bandwidth allocation is always dynamically adapted to the total available bandwidth of the 5G network in real time. When the available bandwidth of the network increases, the guaranteed bandwidth and shared bandwidth are expanded proportionally to avoid resource waste; when the available bandwidth decreases, priority is given to ensuring sufficient allocation of guaranteed bandwidth, and then the shared bandwidth is compressed proportionally.

[0058] Step 6.4 involves hierarchical transmission using a pre-defined data scheduling mechanism. Key instruction substreams are sent first, followed by non-key instruction substreams and compressed image data streams in sequence. Specifically, this includes: activating a pre-defined priority queue-based scheduling mechanism, deeply integrated with the 5G network QoS service quality mechanism to prioritize medical data transmission; placing key instruction substreams in the highest priority queue using a preemptive scheduling rule, ensuring that even if other queues are transmitting data, a new key instruction is immediately interrupted for priority processing, guaranteeing the total transmission time from send to receive is within 10 milliseconds to meet the clinical requirements of real-time robotic arm response; placing non-key instruction substreams in the second-highest priority queue, and compressed image data streams in the ordinary priority queue. Once the highest priority queue has no data transmission, data is transmitted in the order of second-highest and ordinary priority; data within the same priority queue is transmitted strictly according to the order of timestamp generation to avoid disordered instruction or image frame transmission that could disrupt surgical logic; and setting a transmission timeout retransmission mechanism for each queue.

[0059] Step 6.5 involves real-time monitoring of the transmission quality of each sub-stream during data transmission to obtain monitoring results. Based on these results, the bandwidth allocation ratio is dynamically adjusted to form a hierarchical transmission processing control command data stream. Specifically, this includes: real-time collection of core indicators for each sub-stream throughout the data transmission process; for critical command sub-streams, focusing on packet loss rate and end-to-end transmission latency, verifying packet loss using data checksums, and calculating latency by comparing sending and receiving timestamps; for non-critical command sub-streams, monitoring data arrival integrity by calculating the ratio of received to sent data packets; and for the compressed image data stream, monitoring frame rate and sharpness loss rate by counting frames and assessing pixel contrast in key areas. Analyze image clarity; when the packet loss rate of critical command substreams exceeds 1% or the transmission latency reaches 80% of the optimized latency threshold, reduce the proportion of image data streams in the shared bandwidth by 10% and supplement it to the guaranteed bandwidth; when the integrity of non-critical command substreams is below 95%, increase their proportion in the shared bandwidth by 5%; when the frame rate of image data streams drops by more than 5 frames / second or the clarity loss rate of critical areas exceeds 10%, temporarily reduce the compression rate by one level and restore 5% of the shared bandwidth; based on these real-time monitoring results, dynamically adjust the bandwidth allocation ratio to form a closed-loop control of transmission quality, and finally output a hierarchical transmission processing control command data stream that is stable in transmission, has clear priorities, and is adapted to the current network status.

[0060] In this embodiment of the invention, a hierarchical transmission strategy is determined by analyzing the numerical characteristics of the optimized latency threshold, enabling the transmission strategy to accurately adapt to the current network state. The control command data stream is divided into critical command sub-streams and non-critical command sub-streams, while the surgical image data stream undergoes multi-level compression processing. This clarifies the data transmission priority and reduces the bandwidth consumption of the image data. Transmission bandwidth is dynamically allocated based on network latency conditions, allocating guaranteed bandwidth resources to critical command sub-streams and shared bandwidth resources to non-critical command sub-streams and the compressed image data stream, achieving efficient utilization of bandwidth resources. A pre-set data scheduling mechanism is used for hierarchical transmission, prioritizing the transmission of critical command sub-streams to ensure real-time response to core surgical operations. During transmission, the transmission quality of each sub-stream is monitored in real time, and the bandwidth allocation ratio is dynamically adjusted to ensure transmission stability, effectively avoiding control command lag or image stuttering issues, improving the reliability and security of remote surgical data transmission, and adapting to the dynamic fluctuation characteristics of 5G networks.

[0061] In a preferred embodiment of the present invention, step 7 above may include: Step 7.1 involves abstracting key tissues and organs within the surgical area into polygonal boundaries. An obstacle constraint model is then established based on these polygonal boundaries to determine the starting and target coordinates of the robotic arm's movement. Specifically, this includes: utilizing high-resolution CT and MRI medical image data acquired preoperatively, extracting the contour features of key tissues and organs within the surgical area through edge detection and threshold segmentation, including vulnerable structures such as important blood vessels, nerve bundles, and vital organs; sampling feature points on the extracted contours, selecting one contour feature point every 0.5 mm, and using a least-squares fitting method to connect the discrete feature points into a continuous, smooth polygon. Polygonal boundaries are used to precisely map the spatial range of the corresponding tissue or organ. Using these polygonal boundaries as core constraints, a three-dimensional obstacle constraint model is constructed. The model clearly defines the spatial coordinate range, volume, and relative position of each obstacle. Combined with the pre-set operation procedures in the surgical plan, the coordinates of the starting point of the robotic arm movement, i.e., the actual spatial position of the current robotic arm end effector, and the coordinates of the target point, i.e., the key surgical operation positions such as lesion treatment and instrument interaction, are determined to ensure that both the starting point and the target point are within the safe area defined by the obstacle constraint model, and that the line connecting the two points does not cross any polygonal obstacle boundary.

[0062] The specific training and optimization process of the obstacle constraint model is as follows: First, preoperative data preparation is performed by collecting high-resolution CT and MRI medical images of the surgical area. Image preprocessing techniques are used to remove data noise and enhance tissue contour contrast to ensure that key structures are clearly identifiable. Based on clinical surgical safety guidelines, critical tissues and organs requiring special avoidance are selected, including important blood vessels, nerve bundles, and vital organs that are prone to damage. Feature points are extracted from the contours of the selected tissues and organs, and spatial coordinate points on the contours are collected at a density interval of 0.5 mm to form a discrete set of feature points. The least squares fitting method is used to connect the discrete feature points into continuous smooth polygonal boundaries. Each polygonal boundary accurately maps to the actual spatial range of the corresponding tissue or organ. The number of vertices of the polygons is dynamically adjusted according to the contour complexity to ensure boundary fit. A three-dimensional surgical space coordinate system is established with the center of the operating table as the origin. The coordinates of the feature points of all polygonal boundaries are transformed into this coordinate system to clarify the three-dimensional spatial position, volume, and relative positional relationship of each polygonal obstacle. Finally, a safe distance threshold is set. Based on clinical surgical experience, a 5 mm safety buffer area is added to each polygonal obstacle to form a complete obstacle constraint model. The model also marks the allowed and prohibited areas for robotic arm movement.

[0063] Model training focuses on calibrating the model against simulated surgical scenarios and real-world data. First, a physical simulation environment consistent with clinical surgery is built, recreating the anatomical structure of the surgical area, the robotic arm's motion parameters, and the operational procedures. The initially constructed obstacle constraint model is loaded into the physical simulation environment for training. The robotic arm is driven to perform obstacle avoidance tests along a preset path, collecting data such as the actual trajectory of the robotic arm's end effector, the minimum distance to obstacles, and the obstacle avoidance success rate. Based on the test data, model deviations are analyzed. If the distance between the robotic arm and obstacles is less than the safety threshold or the obstacle avoidance path is unreasonable, the coordinates of the feature points on the corresponding polygon boundaries are adjusted to optimize the boundary fitting accuracy. Subsequently, surgical image data from multiple patients are introduced to expand the model's scenario adaptability. For differences in anatomical structures, the shape parameters of the polygon boundaries and the safety distance threshold are adjusted to ensure the model is suitable for diverse surgical scenarios. The optimized model is then connected to an actual surgical robot for small-scale clinical simulation validation, collecting obstacle avoidance response data and positional deviation data of the robotic arm in a real surgical environment. Based on the validation results, the model parameters were further fine-tuned, such as correcting boundary offsets caused by tissue deformation and optimizing the dynamic adaptation logic of the safety distance. This resulted in an obstacle constraint model that had been validated in multiple scenarios and achieved the required accuracy. The trained model can accurately match the anatomical structure of the surgical area, providing a reliable spatial constraint basis for the robotic arm's obstacle avoidance path planning, ensuring the safety and precision of surgical operations.

[0064] Step 7.2 involves planning the motion trajectory based on the robot obstacle avoidance path calculation method and calculating the optimal path according to the obstacle constraint model. The optimal path consists of a series of continuous line segments, forming an optimal path sequence to ensure that the path trajectory avoids all polygonal obstacle areas. Specifically, this includes: planning the motion trajectory based on the spatial search principle of robot obstacle avoidance path planning; firstly, integrating the path length cost and obstacle avoidance safety cost, where the path length cost is calculated using Euclidean distance, and the obstacle avoidance safety cost is obtained by weighting the shortest distances from nodes to the boundaries of polygonal obstacles; using the obstacle constraint model as the search environment, setting the starting point of the robotic arm's movement as the search start point and the target point as the search end point. The distance threshold between adjacent path nodes is set to 1 mm, and the obstacle safety distance threshold is set to 5 mm to ensure that the robotic arm maintains a sufficient safety gap with obstacles during movement. Adjacent nodes are generated iteratively, and the total cost function value of each node is calculated, which is the sum of the actual movement cost from the starting point to the node and the estimated cost from the node to the destination. The node with the smallest total cost is always selected as the next search node. The search continues iteratively until the target point is reached, forming an optimal path sequence consisting of a series of continuous line segments. The endpoint of each line segment is a key node of the path. All line segments do not intersect with the boundary of the polygonal obstacle, and the overall path length is the shortest with the fewest turning times, balancing movement efficiency and obstacle avoidance safety.

[0065] Step 7.3 transforms the optimal path sequence into robotic arm joint motion parameters. Inverse kinematics is then performed based on the principles of robotic arm kinematics to generate corresponding joint angle control commands. Specifically, this involves: extracting the three-dimensional spatial coordinates of each key node in the optimal path sequence and using these coordinates as the desired position and orientation parameters of the robotic arm's end effector. Based on the structural characteristics of the surgical robot's robotic arm, including the number of joints, link length, and joint range of motion, a forward kinematics mechanism is established to clarify the mathematical mapping relationship between the end effector's position and orientation and the angles of each joint. Inverse kinematics is then derived based on the forward kinematics mechanism, i.e., by using the desired position and orientation parameters of the end effector to inversely deduce the angle value required for each joint. Joint motion limit constraints are introduced during the solution process to automatically eliminate invalid solutions exceeding the maximum range of motion of the joints. Simultaneously, a numerical iteration method is used to optimize the solution accuracy, ensuring that the calculation error for each joint angle is controlled within 0.1 degrees. The obtained joint angles are then arranged sequentially according to the path node order to generate corresponding joint angle control commands, achieving a precise conversion from spatial path coordinates to robotic arm joint movements.

[0066] Step 7.4 involves smoothing the joint angle control commands to eliminate abrupt changes in the motion trajectory and forming a control command sequence that matches the motion characteristics of the surgical robot arm. Specifically, this includes: frame-by-frame parsing of the generated joint angle control command sequence; identifying abrupt changes where the angle change rate exceeds the maximum allowable angular acceleration of the robot arm by calculating the joint angle change rate of adjacent commands. These abrupt changes can cause impacts and stuttering during robot arm movement, affecting operational stability; smoothing the command segments containing the abrupt changes by fitting the angle change trends of adjacent valid commands to generate continuous transitional joint angle commands, ensuring the continuity of both the first and second derivatives of the joint angles; simultaneously, adjusting the angle change rate of the commands based on the dynamic characteristics of the surgical robot arm, including joint inertia and damping coefficients, to ensure that the angular velocity and angular acceleration of the joint movement are controlled within the design limits of the robot arm; after optimization, abrupt changes in the trajectory are completely eliminated, forming a control command sequence that highly matches the motion characteristics of the robot arm, ensuring the smoothness and continuity of the robot arm's movement.

[0067] Step 7.5: Drive the surgical robot's robotic arm to perform surgical operations using a sequence of control commands. Monitor the position accuracy of the robotic arm's end effector in real time during the movement to obtain the monitoring results. Fine-tune the control command parameters based on the monitoring results. Specifically, this includes: sending the optimized control command sequence to the surgical robot via a high-speed communication bus to drive each joint to move precisely according to the command requirements, thereby driving the robotic arm's end effector to perform surgical operations such as cutting, suturing, and hemostasis; during the movement, collect the actual spatial position coordinates of the end effector in real time at a period of 10 milliseconds, and calculate the deviation value between the actual position and the expected position, i.e., the position accuracy data; the preset position accuracy threshold is 0.1 mm. When the detected deviation value exceeds this threshold, calculate and fine-tune the corresponding joint angle control command parameters using proportional, integral, and derivative control based on the direction, magnitude, and trend of the deviation, such as increasing or decreasing the angle value of a specific joint in the direction of the deviation; continuously cycle through the monitoring and fine-tuning process to ensure that the robotic arm's end effector always moves within the preset accuracy range, maximizing the accuracy of the surgical operation and avoiding surgical risks caused by robotic arm movement deviations.

[0068] In this embodiment of the invention, by abstracting key tissues and organs in the surgical area into polygonal boundaries and establishing an obstacle constraint model, the coordinates of the starting and target points of the robotic arm's movement are clearly defined, providing precise spatial constraints for obstacle avoidance path planning and effectively mitigating the risk of accidental contact with surgical instruments. Optimal paths are planned within the obstacle constraint model, forming a continuous sequence of optimal paths to ensure that the robotic arm's trajectory avoids all polygonal obstacle areas, balancing path safety and movement efficiency. The optimal path sequence is converted into robotic arm joint motion parameters, and corresponding joint angle control commands are generated through inverse kinematics solving, achieving precise conversion from path planning to actual robotic arm movements. Smoothness optimization is applied to the joint angle control commands to eliminate abrupt changes in the motion trajectory, ensuring that the control command sequence matches the robotic arm's motion characteristics and improving operational coherence and stability. The robotic arm is driven to perform surgical operations through the control command sequence, while the accuracy of the robotic arm's end-effector position is monitored in real time and the control command parameters are fine-tuned, ensuring the precision of the surgical operation and further enhancing the safety and reliability of remote surgery, adapting to the operational needs of complex surgical procedures.

[0069] like Figure 2 As shown, embodiments of the present invention also provide an embodied robot intelligent control system based on human-computer interaction data, including: A module is established to create a two-way 5G network transmission channel between the control unit and the surgical robot. The transmission module is used to synchronously transmit two types of data streams between the control end and the surgical robot end through the two-way transmission channel of the 5G network: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. The monitoring module is used to continuously monitor real-time network latency during the transmission of two types of data streams and determine whether the current network latency exceeds a preset latency threshold, wherein the preset latency threshold can be dynamically adjusted according to the network status. The deployment module is used to dynamically deploy multiple network probes on the bidirectional transmission channel of the 5G network when the current network latency is determined to exceed a preset latency threshold. The network probes are deployed at the base station access point, the core network gateway, and the edge computing node. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed. The calculation module is used to calculate the area change rate of the polygon delay distribution map; to form a network state adjustment factor based on the area change rate; and to dynamically correct the preset delay threshold through the adjustment factor to obtain the optimized delay threshold. The processing module is used to perform hierarchical transmission processing on the surgical image data stream and the control command data stream according to the numerical characteristics of the optimized delay threshold, so as to obtain the hierarchical transmission processing control command data stream. The execution module is used to form a sequence of control commands that matches the motion characteristics of the surgical robot arm based on the hierarchical transmission and processing of the control command data stream; and to drive the surgical robot arm to perform corresponding surgical operations through the sequence of control commands.

[0070] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent control method for embodied robots based on human-computer interaction data, characterized in that, The method includes: Establish a two-way 5G network transmission channel between the control terminal and the surgical robot terminal; Through the two-way transmission channel of the 5G network, two types of data streams are transmitted synchronously between the control end and the surgical robot end: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. During the transmission of the two types of data streams, the real-time network latency is continuously monitored, and it is determined whether the current network latency exceeds the preset latency threshold, which can be dynamically adjusted according to the network status. When the current network latency is determined to exceed the preset latency threshold, multiple network probes are dynamically deployed on the bidirectional transmission channel of the 5G network. The network probes are deployed at the base station access point, the core network gateway, and the edge computing node. Based on the real-time latency data collected by the network probes, a dynamically changing polygonal latency distribution map is constructed. Calculate the area change rate of the polygonal time delay distribution map; form a network state adjustment factor based on the area change rate; dynamically correct the preset time delay threshold using the adjustment factor to obtain the optimized time delay threshold; Based on the numerical characteristics of the optimized delay threshold, the surgical image data stream and the control command data stream are processed in a hierarchical manner to obtain the hierarchical control command data stream. Based on the hierarchical transmission and processing of the control command data stream, a control command sequence matching the motion characteristics of the surgical robot arm is formed; the control command sequence drives the surgical robot arm to perform corresponding surgical operations.

2. The intelligent control method for embodied robots based on human-computer interaction data according to claim 1, characterized in that, Establish a two-way 5G network transmission channel between the control terminal and the surgical robot terminal, including: The surgical robot initiates a connection establishment request to the network management platform; the connection establishment request includes the device identifier of the surgical robot and the required network service quality level. Based on the connection establishment request, the execution result of the connection establishment request is obtained; based on the execution result of the connection establishment request, the allocation process of dedicated network slice resources is triggered, and a logical connection path with the control end is established at the same time. Through the logical connection path, the surgical robot sends a channel test data packet to the control terminal, and after the transmission is completed, it initiates a confirmation channel data packet reception waiting mechanism. Based on the surgical robot, the round-trip time and packet loss rate of the channel test data packets are analyzed to obtain the analysis results. Based on the analysis results, it is used to verify whether the transmission performance of the logical connection path meets the requirements of surgical operation. Once the transmission performance is verified to meet the surgical requirements, a stable two-way transmission channel for the 5G network is established through collaborative confirmation between the surgical robot and the control unit.

3. The intelligent control method for embodied robots based on human-computer interaction data according to claim 2, characterized in that, Through a two-way transmission channel of the 5G network, two types of data streams are synchronously transmitted between the control terminal and the surgical robot terminal: a control command data stream generated by the control terminal and a surgical image data stream acquired by the surgical robot terminal, including: The initial configuration result is obtained by performing the initial configuration of data stream transmission, that is, setting the working parameters of the surgical image acquisition device and determining the frame rate and resolution standard of the data to be transmitted. Based on the initial configuration results, the surgical image data collected by the surgical robot is encapsulated in real time, and the control command data from the control terminal is parsed and its validity is verified to form two types of standard data streams that conform to the transmission specifications. Based on the real-time requirements of surgical operations, priority is assigned to the two types of standard data streams: the control command data stream is marked as the highest transmission priority, and the surgical image data stream is marked as the second highest transmission priority. Through the two-way transmission channel of the 5G network, the two types of data streams are transmitted synchronously according to the set priority strategy, and the timing consistency between the data is maintained during the transmission process. The 5G network transmission quality is continuously monitored by the surgical robot to obtain monitoring feedback results. Based on the monitoring feedback results, the compression ratio parameters of the data stream are dynamically adjusted. The compression ratio parameters are a set of control parameters used to dynamically balance data volume and transmission quality to ensure the integrity and real-time transmission of critical surgical data.

4. The intelligent control method for embodied robots based on human-computer interaction data according to claim 3, characterized in that, During the transmission of both types of data streams, real-time network latency is continuously monitored, and it is determined whether the current network latency exceeds a preset latency threshold. This preset latency threshold can be dynamically adjusted based on network conditions, including: Start the network latency monitoring mechanism, configure the monitoring period parameters and latency sampling frequency, and establish a latency data buffer to store historical latency data; After the monitoring mechanism is activated, based on the configured monitoring cycle parameters, the round-trip latency of the control command data stream and the one-way transmission latency of the surgical image data stream are collected synchronously during the data transmission process. The collected multi-channel latency data is preprocessed in real time, and the average latency value is calculated after removing abnormal latency sampling points; the average latency value is used as the evaluation index of the current network latency. The current network latency evaluation metrics are compared and analyzed in real time with preset latency thresholds to obtain latency status judgment results. The corresponding processing flow is triggered based on the delay status judgment result. When the delay status judgment result exceeds the threshold, the dynamic adjustment mechanism is started; when the delay status judgment result does not exceed the threshold, the existing transmission strategy is maintained.

5. The intelligent control method for embodied robots based on human-computer interaction data according to claim 4, characterized in that, When the current network latency is determined to exceed the preset latency threshold, multiple network probes are dynamically deployed on the bidirectional transmission channel of the 5G network. Network probes are deployed at base station access points, core network gateways, and edge computing nodes. Based on real-time latency data collected by network probes, a dynamically changing polygonal latency distribution map is constructed, including: The network probe dynamic deployment process is triggered by the surgical robot to identify key deployment nodes in the 5G network transmission path. Key deployment nodes include base station access points, core network gateways and edge computing nodes. Based on the location information of the identified key deployment nodes, a probe deployment request is sent to the network management platform to activate the network probe function on the specified nodes. The network probe function collects real-time latency data of each key node and records the corresponding data collection timestamps to form a time-stamped latency dataset. Based on the latency dataset and the geometric positional relationship of each node, the base station access point, core network gateway, and edge computing node are used as vertices, and the vertices are connected in the order of the transmission path to construct a dynamic polygon latency distribution map. Topological verification is performed on the polygon time delay distribution map to ensure the closure of the polygon structure and the correctness of the vertex connections, thereby obtaining an effective time delay distribution map that can be used to calculate the rate of change of area.

6. The intelligent control method for embodied robots based on human-computer interaction data according to claim 5, characterized in that, Calculate the rate of change of the area of ​​the polygonal time-delay distribution map; The network state adjustment factor is determined based on the area change rate. The optimized latency threshold is obtained by dynamically correcting the preset latency threshold through adjustment factors, including: Obtain the vertex coordinate sequence of the effective time delay distribution map; divide the polygonal time delay distribution map into multiple triangular sub-regions based on the vertex coordinate sequence; The area of ​​each triangular sub-region is calculated based on the triangular face accumulation method. That is, a common vertex of the polygon is selected; the common vertex is connected to the remaining non-adjacent vertices to form a triangular division, and the area value of each division triangle is calculated to obtain the area value of each triangular sub-region. The area values ​​of each triangular sub-region are summed to obtain the current total area of ​​the polygon time delay distribution map; at the same time, the total area of ​​the polygon time delay distribution map at the previous moment is obtained from the time delay data buffer, that is, the historical total area. Calculate the ratio of the difference between the current total area and the historical total area; use this ratio as the area change rate of the polygon time-delay distribution map; A network state adjustment factor is generated based on the magnitude of the area change rate; the preset delay threshold is proportionally corrected by the adjustment factor to form an optimized delay threshold suitable for the current network state.

7. The intelligent control method for embodied robots based on human-computer interaction data according to claim 6, characterized in that, Based on the numerical characteristics of the optimized latency threshold, the surgical image data stream and the control command data stream are subjected to hierarchical transmission processing to obtain the hierarchically processed control command data stream, including: Analyze the numerical characteristics of the optimized latency threshold, and determine the hierarchical transmission strategy based on the different ranges of the numerical characteristics; Based on a hierarchical transmission strategy, the control command data stream is divided into critical command sub-streams and non-critical command sub-streams; at the same time, the surgical image data stream is subjected to multi-level compression processing to obtain a compressed image data stream. The transmission bandwidth is dynamically allocated based on network latency, that is, guaranteed bandwidth resources are allocated to critical instruction sub-streams, and shared bandwidth resources are allocated to non-critical instruction sub-streams and compressed image data streams. The data is transmitted in a hierarchical manner through a preset data scheduling mechanism, prioritizing the transmission of critical instruction sub-streams, and then transmitting non-critical instruction sub-streams and compressed image data streams in sequence. The transmission quality of each sub-stream is monitored in real time during data transmission to obtain monitoring results. Based on the monitoring results, the bandwidth allocation ratio is dynamically adjusted to form a control command data stream after hierarchical transmission processing.

8. The intelligent control method for embodied robots based on human-computer interaction data according to claim 7, characterized in that, Based on the hierarchical transmission and processing of the control command data stream, a control command sequence that matches the motion characteristics of the surgical robot's robotic arm is formed. The surgical robot's robotic arm is driven by a sequence of control commands to perform corresponding surgical operations, including: Key tissues and organs within the surgical area are abstracted into polygonal boundaries. An obstacle constraint model is then established based on these polygonal boundaries to determine the coordinates of the starting and target points of the robotic arm's movement. The motion trajectory is planned based on the robot obstacle avoidance path calculation method, and the final path is calculated according to the obstacle constraint model. The final path consists of a series of continuous line segments, that is, the final path sequence is formed to ensure that the path trajectory avoids all polygon obstacle areas. The final path sequence is converted into joint motion parameters of the robotic arm, and inverse kinematics is solved based on the kinematics principle of the robotic arm to generate corresponding joint angle control commands. Smoothness optimization is performed by using joint angle control commands to eliminate abrupt changes in the motion trajectory and form a control command sequence that matches the motion characteristics of the surgical robot arm. The surgical robot arm is driven to perform surgical operations by a sequence of control commands. The positional accuracy of the end effector of the robotic arm is monitored in real time during the movement to obtain the monitoring results. The control command parameters are then fine-tuned based on the monitoring results.

9. An embodied robot intelligent control system based on human-computer interaction data, wherein the system implements the method as described in any one of claims 1 to 8, characterized in that, include: A module is established to create a two-way 5G network transmission channel between the control unit and the surgical robot. The transmission module is used to synchronously transmit two types of data streams between the control end and the surgical robot end through the two-way transmission channel of the 5G network: the control command data stream generated by the control end and the surgical image data stream collected by the surgical robot end. The monitoring module is used to continuously monitor real-time network latency during the transmission of two types of data streams and determine whether the current network latency exceeds a preset latency threshold, wherein the preset latency threshold can be dynamically adjusted according to the network status. The deployment module is used to dynamically deploy multiple network probes on the bidirectional transmission channel of the 5G network when the current network latency is determined to exceed a preset latency threshold. Network probes are deployed at base station access points, core network gateways, and edge computing nodes. Based on real-time latency data collected by network probes, a dynamically changing polygonal latency distribution map is constructed. The calculation module is used to calculate the rate of change of the area of ​​the polygon time-delay distribution map; A network state adjustment factor is formed based on the area change rate; the preset delay threshold is dynamically corrected by the adjustment factor to obtain the optimized delay threshold; The processing module is used to perform hierarchical transmission processing on the surgical image data stream and the control command data stream according to the numerical characteristics of the optimized delay threshold, so as to obtain the hierarchical transmission processing control command data stream. The execution module is used to form a sequence of control commands that matches the motion characteristics of the surgical robot arm based on the hierarchical transmission and processing of the control command data stream; and to drive the surgical robot arm to perform corresponding surgical operations through the sequence of control commands.

10. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

Images