Arm control method, device and terminal for surgical robot

Abstract

The application relates to the technical field of communication networks, and provides a mechanical arm control method and device for a surgical robot and a terminal. The method is applied to a control platform and comprises the following steps: acquiring a first state value set corresponding to a mechanical arm of a surgical robot during operation, the first state value set comprising a plurality of first state value sequences corresponding to the mechanical arm and respective first weights; sampling the plurality of first state value sequences to obtain a plurality of second state value sequences; determining respective second weights of the plurality of second state value sequences according to an importance sampling density and a posterior probability corresponding to the first state value set; determining a target network slice according to the plurality of second state value sequences and all the second weights; and controlling the mechanical arm to operate through the target network slice. The target network slice determined by the method has the advantages of large bandwidth and low time delay, and thus the operating efficiency of the mechanical arm can be effectively improved when the mechanical arm is controlled to operate according to the target network slice.

Description

Technical Field

[0001] This application relates to the field of communication network technology, specifically to a robotic arm control method, device, and terminal for surgical robots. Background Technology

[0002] Teleoperational surgical robots based on 5G (5th Generation Mobile Communication Technology) are a new type of robot proposed to address the complexity and limitations of traditional surgery. 5G's high bandwidth effectively acquires three-dimensional (3D) high-definition images, improving the clarity of surgical site images, while its low latency effectively eliminates the problem of end effector judder in robotic arms. This allows the end effector to surpass the human hand and perform delicate operations in confined spaces that were previously inaccessible. Furthermore, 5G network slicing is a functional definition implemented through a series of custom software programs. These programs define features including geographic coverage, duration, capacity, speed, latency, reliability, security, and availability.

[0003] In controlling the operation of the robotic arm of a surgical robot, it is necessary to configure a superior 5G network slice to meet the technical requirements of high bandwidth and low latency, so as to improve the operating efficiency of the robotic arm. Summary of the Invention

[0004] This application provides a method, apparatus, and terminal for controlling a robotic arm in a surgical robot, addressing the technical problem of low operational efficiency caused by limitations in existing network slicing methods. By resampling the state value sequence of the robotic arm, a target network slice is determined based on the resampled state value sequence. During the resampling process, the state value sequence is redistributed from low-density regions to high-density regions, solving the degradation problem of the state value sequence and significantly optimizing it. This allows for the determination of the optimal network slice corresponding to the robotic arm's operation, i.e., the target network slice. This target network slice possesses advantages such as high bandwidth and low latency. Therefore, controlling the robotic arm's operation based on this target network slice effectively improves the robotic arm's operational efficiency.

[0005] In a first aspect, embodiments of this application provide a robotic arm control method for a surgical robot, applied to a control platform, the method comprising:

[0006] Obtain a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences.

[0007] The plurality of first state value sequences are sampled to obtain a plurality of second state value sequences;

[0008] Based on the importance sampling density and the posterior probability corresponding to the first set of state values, a second weight is determined for each of the plurality of second state value sequences, wherein the posterior probability is determined based on the plurality of first state value sequences and all first weights;

[0009] Based on the multiple second state value sequences and all second weights, a target network slice is determined, and the operation of the robotic arm is controlled through the target network slice.

[0010] In one embodiment, determining the target network slice based on the plurality of second state value sequences and all second weights includes: for each second state value sequence, determining the information entropy corresponding to the second state value sequence based on the second state value sequence and the second weights of the second state value sequence; sorting all information entropies greater than a first preset entropy threshold in descending order to obtain an entropy sequence; determining the target information entropy that satisfies the second preset entropy threshold from the entropy sequence; and determining the network slice corresponding to the target information entropy as the target network slice.

[0011] In one embodiment, determining the target information entropy that satisfies a second preset entropy threshold from the entropy sequence includes: determining the absolute value of the difference between a first information entropy and a second information entropy in the entropy sequence, wherein the first information entropy is less than the second information entropy and the first information entropy is adjacent to the second information entropy; and determining the first information entropy as the target information entropy when the ratio between the absolute value of the difference and the absolute value of the first information entropy is greater than the second preset entropy threshold.

[0012] In one embodiment, determining the network slice corresponding to the target information entropy as the target network slice includes: obtaining a set of element information in the network slice corresponding to the target information entropy; and removing redundancy from the set of element information to obtain the target network slice.

[0013] In one embodiment, determining the target network slice based on the plurality of second state value sequences and all second weights includes: constructing a second state value set based on the plurality of second state value sequences and all second weights; determining the second state value set as a new first state value set; and repeatedly performing the above-described step of sampling the plurality of first state value sequences to obtain a plurality of second state value sequences; determining the second weights of each of the plurality of second state value sequences based on the importance sampling density and the posterior probability corresponding to the first state value set, until a target state value set corresponding to a preset number of times is determined; and determining the target network slice based on the target state value set.

[0014] In one embodiment, sampling the plurality of first state value sequences to obtain a plurality of second state value sequences includes: determining a plurality of target first weights among all the first weights that are greater than a preset weight threshold; and determining the first state value sequence of each of the plurality of target first weights as the plurality of second state value sequences.

[0015] Secondly, embodiments of this application provide a robotic arm control device for a surgical robot, comprising:

[0016] The state observation module is used to acquire a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences.

[0017] The particle filter calculation module is used to sample the plurality of first state value sequences to obtain a plurality of second state value sequences; determine the second weight of each of the plurality of second state value sequences according to the importance sampling density and the posterior probability corresponding to the first state value set, wherein the posterior probability is determined based on the plurality of first state value sequences and all first weights; and determine the target network slice according to the plurality of second state value sequences and all second weights.

[0018] The network slice management module is used to control the operation of the robotic arm through the target network slice.

[0019] Thirdly, embodiments of this application provide a terminal, including a memory, a transceiver, and a processor;

[0020] A memory for storing computer programs; a transceiver for sending and receiving data under the control of the processor; and a processor for reading the computer programs from the memory and executing the steps of implementing the robotic arm control method for a surgical robot as described in the first aspect.

[0021] Fourthly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the steps of the robotic arm control method for a surgical robot described in the first aspect.

[0022] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the robotic arm control method for a surgical robot described in the first aspect.

[0023] The robotic arm control method, apparatus, and terminal for surgical robots provided in this application embodiment acquire a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences. The multiple first state value sequences are sampled to obtain multiple second state value sequences. A second weight for each of the multiple second state value sequences is determined based on the importance sampling density and the posterior probability corresponding to the first state value set, wherein the posterior probability is determined based on the multiple first state value sequences and all first weights. A target network slice is determined based on the multiple second state value sequences and all second weights, and the robotic arm operation is controlled through the target network slice. This method determines the target network slice by resampling the state value sequences of the robotic arm and then using the resampled state value sequences. During the resampling of the state value sequence, the state value sequence is redistributed from low-density regions to high-density regions, which solves the degradation problem of the state value sequence and optimizes the state value sequence to a large extent. This allows the determination of the optimal network slice corresponding to the robotic arm of the surgical robot during operation, i.e., the target network slice. The target network slice has advantages such as large bandwidth and low latency. In this way, controlling the operation of the robotic arm according to the target network slice can effectively improve the operating efficiency of the robotic arm. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram illustrating a scenario where the control platform provided in this application interacts with other devices for data exchange.

[0026] Figure 2 This is a schematic diagram illustrating a scenario where other devices access the control platform as provided in the embodiments of this application;

[0027] Figure 3 This is a schematic diagram of the microservice framework structure corresponding to the control platform provided in the embodiments of this application;

[0028] Figure 4 This is a flowchart illustrating the robotic arm control method for a surgical robot provided in an embodiment of this application;

[0029] Figure 5 This is a schematic diagram of the structure of the robotic arm control device for a surgical robot provided in an embodiment of this application;

[0030] Figure 6 This is a schematic diagram of the terminal structure provided in the embodiments of this application;

[0031] Figure 7 This is a schematic diagram of the control platform provided in the embodiments of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] In related technologies, the selection of which 5G network slice to use in the application of Sampling Importance Resampling Filter (SIR) particle filter-based surgical teleoperation robots has always been a challenge. Current methods include manual allocation, model allocation, and service testing, but the following problems persist:

[0034] 1. Regarding the manual allocation method, the network slice resources required by this method (hereinafter referred to as network slices) are far greater than the network slice resources required by the actual needs of the surgical robot.

[0035] 2. Regarding the model allocation method, this method directly applies the network slice instance information in the model when tested in a real application environment. This can easily lead to problems such as jamming and malfunctions in the end effector of the surgical robot's robotic arm because the actual application scenario requires a large number of network slice resources while the model allocates too few resources.

[0036] 3. For business testing methods, on-site testing requires maintenance personnel and medical institution personnel to conduct on-site testing, and the network slices required for the surgical robot can only be selected after long-term calculation results are calculated. This is very costly in terms of human resources and time.

[0037] In other words, all of the above applications will have certain limitations.

[0038] Among them, SIR particle filtering is an algorithm widely used in surgical robots for simultaneous localization and mapping (SALM) of the end effector of the robotic arm. It can estimate relevant parameters from a large amount of data containing non-deterministic factors to eliminate the influence of various uncertain factors. Therefore, SIR particle filtering-based surgical robots have a wide range of applications.

[0039] Network slicing divides a physical network into multiple logical networks, each of which is a network slice. Different network slices can provide different network characteristics such as speed, latency, and reliability; the network slices can be isolated from each other, and the overload of one network slice will not affect the performance of other network slices.

[0040] Figure 1 This is a schematic diagram illustrating a scenario where the control platform provided in this embodiment interacts with other devices. Figure 1 The control platform includes a robotic arm control device for the surgical robot; other equipment may include: a surgical robot control console, a surgical robot operating table, and a network system, which may include: wireless network equipment, core network equipment, and transmission network equipment.

[0041] The surgical robot control console is located in the surgical robot control area, and the surgical robot operating table is located in the surgical robot operating area. The surgical robot control console and the surgical robot operating table can form a master-slave surgical robot.

[0042] During data interaction between the control platform and other devices, the control platform can transmit data with the surgical robot console via the A interface to send various control / scheduling commands to the surgical robot and receive data information from the surgical robot. The control platform can also transmit data with the surgical robot's operating table via the Z interface, enabling the surgical robot's robotic arm to execute various time-delayed backend commands, specifically performing surgical teleoperations such as anesthesia, tissue incision, sharp and blunt dissection, retraction and traction, and suturing. Furthermore, the control platform can transmit data with the network system via the Application Programming Interface (API), enabling it to issue network slice instance information (i.e., network slices) during the network slice testing phase and issue the final network slice instance information (i.e., the target network slice) during the network slice confirmation phase.

[0043] In summary, the control platform can interact with the surgical robot console, surgical robot operating table, and network system to remotely control the surgical robot's robotic arm to perform corresponding operations.

[0044] Optionally, the control platform can be connected to other devices via wireless communication technology, which may include, but is not limited to, one of the following: fourth-generation mobile communication technology (4G) and 5G.

[0045] Optionally, the network slice instance value corresponding to the wireless network device may include the number of CPU cores, memory, hard disk and other factors.

[0046] The network slice instance values ​​corresponding to core network devices can include: port, bandwidth, latency, and other factors.

[0047] The network slice instance value corresponding to the transmission network equipment can include: the number of Assigned Amount Units (AAUs), the number of Distributed Units (DUs), the number of Centralized Units (CUs), and other factors. Among them, centralized units refer to general-purpose equipment that can be cloudified and handles non-real-time services; distributed units refer to dedicated equipment that cannot be cloudified and handles real-time services.

[0048] It should be noted that different network slice instance values ​​correspond to different network systems.

[0049] Considering the architecture of the control platform, network system, and master-slave surgical robot, the control platform can be deployed on a cloud-based x86 virtual machine based on Ubuntu 16.04.1 LTS Xenial Xerus and the ROS Kinetic Kame framework. Data exchange between the control platform and the surgical robot console uses the Internet Protocol (IP) bearer, as does data exchange between the control platform and the surgical robot operating table. The IP network corresponding to this IP bearer is determined through pre-configuration of relevant data; the specific scheme is as follows... Figure 2 As shown, the Figure 2 This is a schematic diagram illustrating a scenario where other devices access the control platform as provided in the embodiments of this application.

[0050] exist Figure 2In the above, it is assumed that the network system is a 5G network system. (1) The surgical robot console is connected to the core network equipment through a dedicated line. The main control module and the main control area input / output gateway in the surgical robot console can realize the input and output of various control / scheduling commands.

[0051] (2) The control platform may include a state observation module, a particle filter calculation module, a network slice management module and a prediction generation module.

[0052] The state observation module is used to acquire relevant information about the surgical robot's robotic arm during operation. Optionally, this relevant information may include: position information, speed information, and acceleration information of the robotic arm's end effector.

[0053] The particle filtering calculation module is used to perform particle generation, sampling, resampling, weight normalization calculation, and weight entropy calculation. The particles can include a sequence of state values ​​and the weights corresponding to that sequence of state values.

[0054] The network slice management module is used to send network slice instance information to the Communication Service Management Functions (CSMF) module in the 5G network system, and determine the optimal network slice instance (i.e., the target network slice) based on the weight entropy calculation results generated by the particle filter calculation module.

[0055] The prediction module is used to convert the confirmed particle calculation information (such as state value sequence) into information that can be recognized by the main control module in the surgical robot console, and control the robotic arm of the surgical robot to perform corresponding operations based on the information.

[0056] (3) The robotic arm of the surgical robot can be routed to the control console of the surgical robot through the 5G base station, transmission network equipment and core network equipment in the 5G network system, and reuse the N2 interface.

[0057] In this way, from Figure 2 It can be seen that during the data interaction between the control platform and other devices, after processing multiple particles, the control platform can automatically and reasonably select the target network slice for the robotic arm during operation, thereby effectively ensuring the smooth operation of the surgical robot under the 5G network.

[0058] Each particle is used to characterize the state value sequence corresponding to the robotic arm.

[0059] also, Figure 3 This is a schematic diagram of the microservice framework structure corresponding to the control platform provided in the embodiments of this application. Figure 3As can be seen, the control platform uses the Spring Cloud microservice framework, which can include functional module layers such as the presentation layer, interface layer, programming analysis layer, service implementation layer, and protocol layer.

[0060] The presentation layer can include: web interface presentation, network slice data analysis presentation, and test process management presentation, which are used to provide web management capabilities in the control platform and present data involved in the calculation process and network slice testing process.

[0061] The interface layer can include: managing application programming interfaces (APIs), business APIs, and third-party APIs, providing various internal and external APIs and API management capabilities.

[0062] The programming analysis layer can include: network slice orchestration management, service orchestrator execution engine, and Flink big data analysis engine. It is used to provide management functions such as orchestration, modification, and deletion of various surgical operation business test scenario samples, execution functions of test samples, and data analysis functions based on Flink big data analysis engine.

[0063] The service implementation layer may include: business microservice management, distributed business microservice management group, and particle filter computing service.

[0064] The protocol layer connects all layers and modules via APIs. It provides various interfaces, application protocol flow refactoring, and network link management. Internet protocols within this protocol layer and other functional module layers may include: Software Router (ROS) protocol, Simple Network Management Protocol (SNMP), General Packet Radio Service (GPRS) Tunneling Protocol (GTP), and Hypertext Transfer Protocol (HTTP), among others.

[0065] based on Figure 3The Spring Cloud microservice framework adopted by the control platform has the following advantages: (1) Modular design facilitates customization: Each functional module and 5G testing function is defined by fine-grained "microservices", which makes it easy for users to customize and orchestrate testing capabilities at the granularity of "microservices" according to mobile business scenarios. (2) Lightweight API facilitates expansion: The interface of the electronic device is based on the Internet protocol and interacts with the API that can be flexibly called, which reduces the overhead of device configuration and data processing internally and supports the same interface that provides open capabilities externally. (3) Independence facilitates rapid upgrade and iteration: Since microservices can be deployed independently, business testing functions can be developed and iterated quickly. At the same time, the testing and protocol stack service capabilities of the electronic device can be quickly deployed and elastically scaled up / down based on the virtualization platform.

[0066] It should be noted that the execution subject involved in the embodiments of this application can be a robotic arm control device for surgical robots or a control platform.

[0067] The embodiments of this application will be further described below using the control platform as an example.

[0068] Figure 4 This is a schematic flowchart illustrating a robotic arm control method for a surgical robot provided in an embodiment of this application. (Refer to...) Figure 4 This application provides a robotic arm control method for a surgical robot, which may include:

[0069] 401. Obtain the set of first state values ​​corresponding to the robotic arm of the surgical robot during operation.

[0070] The first state value set includes multiple first state value sequences corresponding to the robotic arm and the first weights of each of the multiple first state value sequences, and the sum of all the first weights is 1.

[0071] It should be noted that a surgical robot is a nonlinear dynamic system.

[0072] The state-space model of a state system can be represented by formula (1). express.

[0073] in, This indicates that the end effector of the surgical robot's robotic arm is in The state value at the given time; Indicates the end effector of the robotic arm at The state value at the given time; Indicates the end effector of the robotic arm at The process noise value at time t; State transition equations for nonlinear dynamic systems; Indicates the end effector of the robotic arm at The observation at time t; Indicates the end effector of the robotic arm at The measurement noise value at that time; The observation equations of a nonlinear dynamic system; Indicates the end effector of the robotic arm at The network slice instance value used at the current time; This represents the equation for selecting the target network slice.

[0074] Assume the execution process of the above nonlinear dynamic system is as follows: According to the existing interpretation of nonlinear filtering problems, the objective of the equation of the nonlinear dynamic system is to recursively estimate the posterior probability corresponding to the state of the nonlinear system based on noisy observations. This process can be expressed by formula (2).

[0075] express.

[0076] in, This represents the posterior probability corresponding to the state of a nonlinear system. Indicates the robotic arm end effector from 0 to The sequence of state values ​​corresponding to each running time; Indicates the number of robotic arm end effectors from 1 to... The sequence of observations corresponding to the time of execution; Indicates the number of robotic arm end effectors from 1 to... The sequence of observations corresponding to the time of execution; Indicates the robotic arm end effector from 0 to The sequence of state values ​​corresponding to each running time.

[0077] It should be noted that, Let represent the conditional posterior probability distribution of factor x after factor y has occurred. It is a measured value, which can be obtained through measurement and calculation. In this case, we can let... The conditional probability observation constant Formula (2) can be used .

[0078] Since this nonlinear system is a nonlinear non-Gaussian system, the posterior probability in formula (2) is... If there is no solution, the control platform can first obtain multiple first state value sequences corresponding to the robotic arm, that is, multiple first state value sequences of the robotic arm end effector and the first weight of each of these multiple first state value sequences, and then construct the first state value set corresponding to the robotic arm during operation.

[0079] The first set of state values ​​can be used express, This represents the number of the first state value sequence, N≥2; express The first state value sequence in the first state value sequence A sequence of first state values; Indicates the robotic arm end effector from 0 to The corresponding time of the runtime A sequence of first state values; Indicates the first A sequence of first state values The corresponding first weights, all first weights satisfy the normalization condition, that is .

[0080] At this point, the posterior probability of the state of the nonlinear system can be expressed by formula (3). This indicates that the nonlinear system satisfies a posterior probability distribution, where, This represents the Dirac function equation.

[0081] In summary, it can be seen from formula (3) that the posterior probability corresponding to the set of first state values ​​is determined based on multiple sequences of first state values ​​and all first weights.

[0082] 402. Sample multiple first state value sequences to obtain multiple second state value sequences.

[0083] Since it is difficult to extract a valid sequence of first state values ​​from the posterior probability distribution, an importance sampling method can be used. This involves introducing an importance sampling density to sample the sequence of first state values ​​in the set of first state values, thereby obtaining multiple sequences of second state values ​​and effectively solving the sampling difficulty problem.

[0084] In some embodiments, the control platform samples multiple first state value sequences to obtain multiple second state value sequences, which may include: the control platform determining multiple target first weights that are greater than a preset weight threshold among all first weights; the control platform determining the first state value sequences of each of the multiple target first weights as multiple second state value sequences.

[0085] The preset weight threshold can be set by the control platform before it leaves the factory or it can be user-defined; no specific limitation is made here.

[0086] After acquiring the first weights of multiple first state value sequences, the control platform first determines target first weights greater than a preset weight threshold from all first weights. There are multiple target first weights. Then, the control platform determines the first state value sequences of these multiple target first weights as multiple second state value sequences for subsequent determination of the target network slice. The target network slice is determined based on the resampled state value sequences. During the resampling process of the state value sequences, the sequences are redistributed from low-density regions to high-density regions, solving the degradation problem of the state value sequences and greatly optimizing them. This allows for the determination of the optimal network slice corresponding to the robotic arm of the surgical robot during operation, i.e., the target network slice. This target network slice has advantages such as high bandwidth and low latency. Therefore, controlling the robotic arm based on this target network slice effectively improves the operating efficiency of the robotic arm.

[0087] 403. Determine the second weights of multiple second state value sequences based on the importance sampling density and the posterior probability corresponding to the first state value set.

[0088] During the sampling process of the first state value sequence in the aforementioned first state value set after introducing importance sampling density, it is also necessary to determine the weight of the sampled first state value sequence, that is, to determine the second weight of each of the multiple second state value sequences. At this time, the weight of the first state value sequence in formula (1) can be determined. The expected value is then calculated, and this process can be described using formula (4).

[0089] express.

[0090] in, Expressing expectations; This represents the importance density function of factor x after factor y occurs. Indicates the importance sampling density; , representing the first in a sequence of multiple second state values The second weight of the second state value sequence.

[0091] The control platform can be approximately derived from the above formula (4). , Represents an approximate expectation; Represents the sequence of multiple second state values. The second weight of the second state value sequence.

[0092] Furthermore, since all second weights satisfy the normalization condition, i.e. ,so,

[0093] Formula (5) = Let it be a computable recursive function; then, according to the conditional probability method, the recursive function... Decomposition yields formula (6).

[0094] Next, substitute formula (6) into formula (5) and consider Markov. Assuming that the orders are mutually independent, we can obtain formula (7).

[0095] = ;

[0096] = .

[0097] in, Indicates that the end effector of the robotic arm starts from... arrive The corresponding time of the runtime A sequence of first state values; Indicates the robotic arm end effector from 0 to The corresponding time of the runtime A sequence of first state values.

[0098] In this way, we can obtain the second weights for each of the multiple second state value sequences.

[0099] 404. Based on multiple second state value sequences and all second weights, determine the target network slice and control the operation of the robotic arm through the target network slice.

[0100] In some embodiments, the control platform determines the target network slice based on multiple second state value sequences and all second weights, which may include: the control platform constructing a second state value set based on multiple second state value sequences and all second weights; the control platform determining the second state value set as a new first state value set; and repeating the above steps 402-403 until the target state value set corresponding to the preset number of times is determined; the control platform determining the target network slice based on the target state value set.

[0101] The preset number of times can be set by the control platform before it leaves the factory or it can be user-defined; no specific limitation is made here.

[0102] In steps 402-403 above, the control platform has acquired multiple second state value sequences and their respective second weights, thereby constructing a second state value set. Since the possibility of particle degradation exists in each of the multiple second state value sequences is considered, this second state value set can be determined as a new first state value set; and steps 402-403 are repeated until the target state value set corresponding to reaching a preset number of times is determined, thereby determining the target network slice.

[0103] At this point, the target state value set is available. express, ≥ ≥2.

[0104] in, Indicates from time 0 to The moment A sequence of target state values, Indicates the first A sequence of target state values The corresponding weights.

[0105] The target state value set is subject to a full probability distribution: , and They are the same.

[0106] In some embodiments, the control platform determines the target network slice based on multiple second state value sequences and all second weights, which may include: for each second state value sequence, the control platform determines the information entropy corresponding to the second state value sequence according to the second weight of the second state value sequence; the control platform sorts all information entropies greater than a first preset entropy threshold in descending order to obtain an entropy sequence; the control platform determines the target information entropy that meets the second preset entropy threshold from the entropy sequence; and the control platform determines the network slice corresponding to the target information entropy as the target network slice.

[0107] Among them, information entropy is a weight entropy.

[0108] The second preset entropy threshold is available. Optionally, the first preset entropy threshold and the second preset entropy threshold... It can be set by the control platform before it leaves the factory, or it can be user-defined; there are no specific limitations here.

[0109] For example, the second preset entropy threshold The value range is [0, 1], and the typical value is 0.9.

[0110] After determining multiple second state value sequences, the control platform first determines the information entropy corresponding to each second state value sequence based on the sequence and its second weight. This yields multiple information entropies. Then, the control platform sorts the information entropies greater than a first preset entropy threshold in descending order to obtain an entropy sequence. Next, the control platform determines the target information entropy from this entropy sequence that satisfies the second preset entropy threshold. Finally, the control platform identifies the network slice corresponding to the target information entropy as the target network slice.

[0111] Optionally, the control platform determines the information entropy corresponding to the second state value sequence based on the second state value sequence and the second weight of the second state value sequence, which may include: the control platform determining the information entropy corresponding to the second state value sequence based on the entropy formula;

[0112] The entropy formula is as follows: ;

[0113] Indicates the first The information entropy corresponding to the second state value sequence, i.e., the first... j The network slice in the _ ... k The information entropy of the second weight obtained from the second surgical procedure; Indicates the first The network slice corresponding to the second state value sequence; express The moment A sequence of target state values, Indicates the first A sequence of target state values The corresponding weights.

[0114] To achieve adaptive evaluation and selection of the optimal 5G network slice (i.e., the target network slice), it is set to be performed in the same environment. k The same surgical procedure is repeated, and the information entropy value is introduced to measure the effectiveness of each network slice. Therefore, the control platform can determine the information entropy with higher accuracy based on the above entropy formula, in order to subsequently determine the target network slice.

[0115] In some embodiments, the control platform determines the target information entropy that satisfies the second preset entropy threshold from the entropy sequence, which may include: the control platform determining the absolute value of the difference between the first information entropy and the second information entropy in the entropy sequence, wherein the first information entropy is less than the second information entropy and the first information entropy is adjacent to the second information entropy; the control platform determines the first information entropy as the target information entropy when the ratio between the absolute value of the difference and the absolute value of the first information entropy is greater than the second preset entropy threshold.

[0116] The control platform is based on formula (8). Determine the absolute value of the difference, where, Represents the first information entropy. This represents the second information entropy; then, the control platform follows formula (9). The control platform determines the ratio between the absolute value of the difference and the absolute value of the first information entropy. Then, it compares the ratio with a second preset entropy threshold. If the ratio is greater than the second preset entropy threshold, it indicates that the cumulative average has decreased significantly. At this time, the first information entropy can be determined as the target information entropy, and then the target network slice corresponding to the target information entropy can be determined.

[0117] In some embodiments, the control platform determines the network slice corresponding to the target information entropy as the target network slice, which may include: the control platform obtaining a set of feature information in the network slice corresponding to the target information entropy; the control platform removing redundancy from the set of feature information to obtain the target network slice.

[0118] The feature information set may include Individual element information, ≥2; This set of element information is available = express, Indicates the first Information on each element.

[0119] After acquiring the set of element information from the network slice corresponding to the target information entropy, the control platform requires a certain degree of network resource redundancy to consider the importance of nonlinear systems and teleoperation surgery. Therefore, based on the test of the optimized network slice, a redundancy coefficient was added. And use formula (10). Redundancy is removed from the prime information set to obtain the target network slice, which can then be used... express, Indicates the redundancy-free th Information on each element.

[0120] Among them, the redundancy coefficient It can be set by the control platform before it leaves the factory, or it can be user-defined; there are no specific limitations here.

[0121] For example, redundancy coefficient The value range is [1, 2], and it is usually 1 or 2.

[0122] In this embodiment, a first set of state values ​​corresponding to the robotic arm of a surgical robot during operation is obtained; multiple first state value sequences are sampled to obtain multiple second state value sequences; a second weight is determined for each of the multiple second state value sequences based on the importance sampling density and the posterior probability corresponding to the first state value set; a target network slice is determined based on the multiple second state value sequences and all second weights, and the operation of the robotic arm is controlled by the target network slice. This method determines the target network slice by resampling the state value sequence of the robotic arm. During the resampling process, the state value sequence is redistributed from low-density regions to high-density regions, solving the degradation problem of the state value sequence and greatly optimizing the state value sequence to determine the optimal network slice corresponding to the robotic arm of the surgical robot during operation, i.e., the target network slice. This target network slice has advantages such as large bandwidth and low latency. Therefore, controlling the operation of the robotic arm based on the target network slice can effectively improve the operating efficiency of the robotic arm.

[0123] The robotic arm control device for a surgical robot provided in the embodiments of this application will be described below. The robotic arm control device for a surgical robot described below can be referred to in correspondence with the robotic arm control method for a surgical robot described above.

[0124] Figure 5 The structural schematic diagram of the robotic arm control device for a surgical robot provided in the embodiments of this application may include:

[0125] The state observation module 501 is used to acquire a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences.

[0126] The particle filter calculation module 502 is used to sample the multiple first state value sequences to obtain multiple second state value sequences; determine the second weight of each of the multiple second state value sequences according to the importance sampling density and the posterior probability corresponding to the first state value set, wherein the posterior probability is determined based on the multiple first state value sequences and all first weights; and determine the target network slice according to the multiple second state value sequences and all second weights.

[0127] The network slice management module 503 is used to control the operation of the robotic arm through the target network slice.

[0128] Optionally, the particle filter calculation module 502 is specifically used to determine the information entropy corresponding to each second state value sequence based on the second state value sequence and the second weight of the second state value sequence; sort all information entropies greater than a first preset entropy threshold in descending order to obtain an entropy sequence; determine the target information entropy that meets the second preset entropy threshold from the entropy sequence; and determine the network slice corresponding to the target information entropy as the target network slice.

[0129] Optionally, the particle filter calculation module 502 is specifically used to determine the absolute value of the difference between the first information entropy and the second information entropy in the entropy sequence, wherein the first information entropy is less than the second information entropy and the first information entropy is adjacent to the second information entropy; if the ratio between the absolute value of the difference and the absolute value of the first information entropy is greater than the second preset entropy threshold, the first information entropy is determined as the target information entropy.

[0130] Optionally, the particle filter calculation module 502 is specifically used to obtain the set of element information in the network slice corresponding to the target information entropy; and to remove redundancy from the set of element information to obtain the target network slice.

[0131] Optionally, the particle filter calculation module 502 is specifically used to construct a second state value set based on the plurality of second state value sequences and all second weights; determine the second state value set as a new first state value set; and repeatedly perform the above steps of sampling the plurality of first state value sequences to obtain a plurality of second state value sequences; determining the second weights of each of the plurality of second state value sequences based on the importance sampling density and the posterior probability corresponding to the first state value set, until the target state value set corresponding to the preset number of times is determined; and determine the target network slice based on the target state value set.

[0132] Optionally, the particle filter calculation module 502 is specifically used to determine multiple target first weights that are greater than a preset weight threshold among all the first weights; and to determine the first state value sequence of each of the multiple target first weights as the multiple second state value sequences.

[0133] The terminal involved in the embodiments of this application may be a device that provides voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. The name of the terminal device may differ in different systems; for example, in a 5G system, the terminal device may be called a User Equipment (UE).

[0134] Figure 6 This is a schematic diagram of the terminal structure according to an embodiment of this application, with reference to... Figure 6This application embodiment also provides a terminal, which may include: a memory 610, a transceiver 620, and a processor 630;

[0135] The memory 610 is used to store computer programs; the transceiver 620 is used to send and receive data under the control of the processor 630; the processor 630 is used to read the computer program in the memory 610 and perform the following operations:

[0136] Obtain a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences.

[0137] The plurality of first state value sequences are sampled to obtain a plurality of second state value sequences;

[0138] Based on the importance sampling density and the posterior probability corresponding to the first set of state values, a second weight is determined for each of the plurality of second state value sequences, wherein the posterior probability is determined based on the plurality of first state value sequences and all first weights;

[0139] Based on the multiple second state value sequences and all second weights, a target network slice is determined, and the operation of the robotic arm is controlled through the target network slice.

[0140] Among them, Figure 6 In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (processor 630) and memory (memory 610). The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 620 can be multiple elements, including transmitters and receivers, providing a unit for communicating with various other devices over a transmission medium. For different user equipment, the user interface 640 can also be an interface capable of connecting external or internal devices as needed.

[0141] The processor 630 is responsible for managing the bus architecture and general processing, while the memory 610 can store the data used by the processor 630 when performing operations.

[0142] The processor 630 executes any of the methods described in the embodiments of this application according to the obtained executable instructions by calling a computer program stored in the memory 610. The processor and the memory may also be physically separated.

[0143] Optionally, the processor 630 is also used to perform the following operations:

[0144] For each second state value sequence, the information entropy corresponding to the second state value sequence is determined according to the second weight of the second state value sequence; all information entropies greater than the first preset entropy threshold are sorted in descending order to obtain an entropy sequence; the target information entropy that satisfies the second preset entropy threshold is determined from the entropy sequence; and the network slice corresponding to the target information entropy is determined as the target network slice.

[0145] Optionally, the processor 630 is also used to perform the following operations:

[0146] In the entropy sequence, determine the absolute value of the difference between the first information entropy and the second information entropy, wherein the first information entropy is less than the second information entropy and the first information entropy is adjacent to the second information entropy; if the ratio between the absolute value of the difference and the absolute value of the first information entropy is greater than the second preset entropy threshold, determine the first information entropy as the target information entropy.

[0147] Optionally, the processor 630 is also used to perform the following operations:

[0148] Obtain the set of element information in the network slice corresponding to the target information entropy; remove redundancy from the set of element information to obtain the target network slice.

[0149] Optionally, the processor 630 is also used to perform the following operations:

[0150] Based on the plurality of second state value sequences and all second weights, a second state value set is constructed; the second state value set is determined as a new first state value set; and the above sampling of the plurality of first state value sequences is repeated to obtain a plurality of second state value sequences; based on the importance sampling density and the posterior probability corresponding to the first state value set, the second weights of each of the plurality of second state value sequences are determined until a target state value set corresponding to a preset number of times is determined; based on the target state value set, the target network slice is determined.

[0151] Optionally, the processor 630 is also used to perform the following operations:

[0152] Determine multiple target first weights that are greater than a preset weight threshold among all the first weights; determine the first state value sequence of each of the multiple target first weights as the multiple second state value sequences.

[0153] Figure 7 An example is a schematic diagram of the physical structure of a control platform, such as... Figure 7As shown, the control platform may include a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 can call a computer program in the memory 730 to execute steps of a robotic arm control method for a surgical robot, such as: acquiring a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation, the first set of state values ​​including multiple first state value sequences corresponding to the robotic arm and first weights for each of the multiple first state value sequences; sampling the multiple first state value sequences to obtain multiple second state value sequences; determining second weights for each of the multiple second state value sequences based on importance sampling density and posterior probabilities corresponding to the first set of state values, wherein the posterior probabilities are determined based on the multiple first state value sequences and all first weights; determining a target network slice based on the multiple second state value sequences and all second weights, and controlling the operation of the robotic arm through the target network slice.

[0154] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0155] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the robotic arm control method for a surgical robot provided in the above embodiments, such as: obtaining a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation, the first set of state values ​​including a plurality of first state value sequences corresponding to the robotic arm and a first weight of each of the plurality of first state value sequences; sampling the plurality of first state value sequences to obtain a plurality of second state value sequences; determining a second weight of each of the plurality of second state value sequences according to the importance sampling density and the posterior probability corresponding to the first set of state values, wherein the posterior probability is determined based on the plurality of first state value sequences and all first weights; determining a target network slice according to the plurality of second state value sequences and all second weights, and controlling the operation of the robotic arm through the target network slice.

[0156] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program. The computer program is used to cause a processor to execute the steps of the robotic arm control method for a surgical robot provided in the above embodiments. For example, it includes: acquiring a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation, the first set of state values ​​including multiple first state value sequences corresponding to the robotic arm and first weights for each of the multiple first state value sequences; sampling the multiple first state value sequences to obtain multiple second state value sequences; determining second weights for each of the multiple second state value sequences based on importance sampling density and the posterior probability corresponding to the first state value set, wherein the posterior probability is determined based on the multiple first state value sequences and all first weights; determining a target network slice based on the multiple second state value sequences and all second weights, and controlling the operation of the robotic arm through the target network slice.

[0157] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0158] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0159] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for controlling a robotic arm in a surgical robot, characterized in that, Applied to a control platform, the method includes: Obtain a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences. The plurality of first state value sequences are sampled to obtain a plurality of second state value sequences; Based on the importance sampling density and the posterior probability corresponding to the first set of state values, the second weights of each of the plurality of second state value sequences are determined, including: ； in, Represents the sequence of multiple second state values. A sequence of second state values ​​in The second weight at time t, Represents the sequence of multiple second state values. A sequence of second state values ​​in The second weight at time t, This represents the posterior probability of factor x after factor y has occurred. This represents the sampling density indicating the importance of factor x after factor y occurs. Indicates the robotic arm end effector from 0 to The sequence of observations corresponding to the time of execution. Indicates the end effector of the robotic arm from 0 to The corresponding time of the runtime A sequence of first state values Indicates that the end effector of the robotic arm starts from... arrive The corresponding time of the runtime A sequence of first state values; Indicates the end effector of the robotic arm at The observation at time t; The posterior probability is determined based on the plurality of first state value sequences and all first weights; Based on the multiple second state value sequences and all second weights, a target network slice is determined, and the operation of the robotic arm is controlled through the target network slice.

2. The robotic arm control method for a surgical robot according to claim 1, characterized in that, The step of determining the target network slice based on the plurality of second state value sequences and all second weights includes: For each second state value sequence, the information entropy corresponding to the second state value sequence is determined based on the second state value sequence and its second weight. Sort all information entropies greater than the first preset entropy threshold in descending order to obtain the entropy sequence; Determine the target information entropy that satisfies the second preset entropy threshold from the entropy sequence; The network slice corresponding to the target information entropy is determined as the target network slice.

3. The robotic arm control method for a surgical robot according to claim 2, characterized in that, Determining the target information entropy that satisfies the second preset entropy threshold from the entropy sequence includes: In the entropy sequence, determine the absolute value of the difference between the first information entropy and the second information entropy, wherein the first information entropy is less than the second information entropy, and the first information entropy and the second information entropy are adjacent; If the ratio between the absolute value of the difference and the absolute value of the first information entropy is greater than the second preset entropy threshold, the first information entropy is determined as the target information entropy.

4. The robotic arm control method for a surgical robot according to claim 2 or 3, characterized in that, The step of determining the network slice corresponding to the target information entropy as the target network slice includes: Obtain the set of element information in the network slice corresponding to the target information entropy; Redundancy is removed from the set of element information to obtain the target network slice.

5. The robotic arm control method for a surgical robot according to any one of claims 1-3, characterized in that, The step of determining the target network slice based on the plurality of second state value sequences and all second weights includes: Based on the plurality of second state value sequences and all second weights, construct a set of second state values; The second set of state values ​​is determined as a new set of state values; and the above sampling of the multiple first state value sequences is repeated to obtain multiple second state value sequences; the second weight of each of the multiple second state value sequences is determined according to the importance sampling density and the posterior probability corresponding to the first state value set, until the target state value set corresponding to the preset number of times is determined; The target network slice is determined based on the target state value set.

6. The robotic arm control method for a surgical robot according to any one of claims 1-3, characterized in that, The step of sampling the plurality of first state value sequences to obtain a plurality of second state value sequences includes: Determine multiple target first weights that are greater than a preset weight threshold from among all the first weights; The first state value sequence of each of the multiple target first weights is determined as the multiple second state value sequences.

7. A robotic arm control device for a surgical robot, characterized in that, The robotic arm control method for a surgical robot as described in claim 1 includes: The state observation module is used to acquire a first set of state values ​​corresponding to the robotic arm of the surgical robot during operation. The first set of state values ​​includes multiple first state value sequences corresponding to the robotic arm and a first weight for each of the multiple first state value sequences. The particle filter calculation module is used to sample the plurality of first state value sequences to obtain a plurality of second state value sequences; determine the second weight of each of the plurality of second state value sequences according to the importance sampling density and the posterior probability corresponding to the first state value set, wherein the posterior probability is determined based on the plurality of first state value sequences and all first weights; and determine the target network slice according to the plurality of second state value sequences and all second weights. The network slice management module is used to control the operation of the robotic arm through the target network slice.

8. A terminal, characterized in that, Includes memory, transceiver, and processor; Memory, used to store computer programs; Transceiver, used to send and receive data under the control of the processor; A processor for reading a computer program from the memory and executing the steps of the robotic arm control method for a surgical robot as described in any one of claims 1 to 6.

9. A control platform, comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the robotic arm control method for a surgical robot according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the robotic arm control method for a surgical robot as described in any one of claims 1 to 6.

Images