Intelligent spinal canal puncture surgical robot control method and device and electronic equipment

Abstract

The embodiment of the invention provides an intelligent spinal canal puncture surgical robot control method and device and electronic device.The control method comprises the steps that measured puncture force at the tail end of a surgical robot is measured in real time, a first puncture force set within a preset time range is obtained, and the first puncture force set comprises multiple historical puncture forces within the preset time range; constructing a Gaussian distribution model, adjusting the Gaussian distribution model in real time by using the latest first puncture force set, and predicting the expected puncture force at each time point through the latest adjusted Gaussian distribution model; acquiring a measurement track of the moving of the tail end of the surgical robot, and adjusting the measured puncture force applied to the tail end of the surgical robot at present according to the measurement track, a preset expected track, the measured puncture force measured in real time and the expected puncture force corresponding to the same time point; according to the method provided by the invention, the safety of an operation can be improved, the success rate of the operation is improved, and the operation burden of medical staff is reduced.

Description

Technical Field

[0001] The present application relates to, but is not limited to, the technical field of surgical equipment, and in particular to a control method, device, and electronic device for an intelligent spinal puncture surgical robot. Background Art

[0002] A spinal puncture is a common medical procedure widely used in the diagnosis and treatment of spinal disorders. It is commonly used clinically to collect cerebrospinal fluid (CSF), measure CSF pressure, and administer spinal anesthetics, chemotherapy drugs, or other medications for diagnostic imaging. This procedure can aid in the diagnosis of serious infections and neurological disorders. During a lumbar puncture, a needle is inserted between two lumbar vertebrae (vertebrae). The needle pierces the skin, superficial fascia, deep fascia, supraspinous ligament, interspinous ligament, and ligamentum flavum to reach the epidural space, where medication is injected. However, this procedure is technically demanding, requiring the surgeon to possess extremely precise operating skills and extensive clinical experience.

[0003] Traditional spinal punctures are typically performed manually by doctors. This requires extensive training for anesthesiologists, requiring significant manpower and time. Furthermore, doctors must adjust the force and angle of the puncture in real time to avoid damaging vital nerves and blood vessels. However, human tissue exhibits complex mechanical properties, such as varying hardness at different levels, and tissue displacement due to physiological movements or external interference. These factors increase the difficulty and risk of the procedure. To improve surgical success rates and reduce the burden on doctors, the development of automated, intelligent robotic systems for spinal puncture surgery has become an important research direction. Summary of the Invention

[0004] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0005] The embodiments of the present application provide a method, device, and electronic equipment for controlling an intelligent spinal puncture surgical robot, which can improve the safety of the surgery, increase the success rate of the surgery, and reduce the operational burden on medical staff.

[0006] To achieve the above-mentioned purpose, the first aspect of an embodiment of the present application proposes a control method for an intelligent spinal puncture surgical robot, including: real-time measurement of the measured puncture force of the surgical robot end, and obtaining a first puncture force set within a preset time range, the first puncture force set including multiple historical puncture forces within the preset time range; constructing a Gaussian distribution model, using the latest first puncture force set to adjust the Gaussian distribution model in real time, and predicting the expected puncture force at each time point through the Gaussian distribution model obtained by the latest adjustment; obtaining the measured trajectory of the surgical robot end, and adjusting the measured puncture force currently applied to the surgical robot end through the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

[0007] In some embodiments, constructing the Gaussian distribution model and adjusting the Gaussian distribution model in real time using the latest first puncture force set includes: obtaining a historical position corresponding to each of the historical puncture forces in the latest first puncture force set; Construct a kernel function using the historical positions: ; Update the Gaussian distribution model using the kernel function: ; Among them, x and Respectively represent the two historical positions, Represents x and The Euclidean distance between represents the hyperparameter in the kernel function, exp represents the calculation of the natural logarithm, GP represents the Gaussian distribution, and m(x) is the mean function, which represents the average value of the historical position.

[0008] In some embodiments, before constructing the kernel function using the historical positions, the method further includes: constructing a marginal likelihood function using the historical positions: ; Perform maximum likelihood estimation on the marginal likelihood function to obtain the hyperparameters; wherein X represents a vector composed of the multiple historical positions, K(X,X) represents the covariance matrix of the vector composed of the multiple historical positions, n represents the number of the historical positions, and I is the identity matrix.

[0009] In some embodiments, the measured puncture force currently applied to the end of the surgical robot is adjusted according to the measurement trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point, including: obtaining the real-time measured puncture force, determining the measurement position of the end of the surgical robot; constructing an admittance control model based on the measured puncture force and the expected puncture force, and adjusting the measured puncture force at the measurement position through the admittance control model.

[0010] In some embodiments, before adjusting the measured puncture force at the measurement position by the admittance control model, the method further includes: obtaining a second puncture force set predicted by the Gaussian distribution model, wherein the second puncture force set includes multiple expected puncture forces predicted by the Gaussian distribution model, and the time points of the expected puncture forces correspond one-to-one to the historical puncture forces in the first puncture force set; determining a measurement error by comparing each expected puncture force in the second puncture force set with the corresponding historical puncture force in the first puncture force set; and adjusting the admittance control model constructed by the expected puncture force at the current time point and the measured puncture force by using the measurement error.

[0011] In some embodiments, the measured puncture force represents an estimate of the actual puncture force between the surgical robot tip and the skin layer, and the adjusted admittance control model is: ; in, represents the expected inertia matrix, represents the desired damping matrix, represents the desired stiffness matrix, represents the current measured acceleration of the surgical robot end, represents the current measurement speed of the surgical robot end, x represents the current measurement position of the surgical robot end, represents the target puncture force currently pursued by the surgical robot terminal, represents the measurement error obtained by the first puncture force set and the second puncture force set, Indicates the current error gain coefficient, which is used to dynamically adjust the compensation strength of the measurement error. represents the historical puncture force in the first puncture force set, represents the expected puncture force in the second puncture force set.

[0012] In some embodiments, the measurement error determination process is specifically as follows: sequentially determining the first error between the historical puncture force and the corresponding expected puncture force ; By the corresponding gain coefficient on the measurement position For the first error Adjust and get the second error ; All the second errors are integrated to obtain the measurement error, wherein the measurement error is , from 0 to t represents the time range from the first puncture force set and the second puncture force set.

[0013] To achieve the above-mentioned purpose, the second aspect of the present application proposes an intelligent spinal puncture surgical robot control device, including: a measurement module, used to measure the measured puncture force of the surgical robot end in real time, and obtain a first puncture force set within a preset time range, the first puncture force set including multiple historical puncture forces within the preset time range; a prediction module, used to construct a Gaussian distribution model, and use the latest first puncture force set to adjust the Gaussian distribution model in real time, and predict the expected puncture force at each time point through the Gaussian distribution model obtained by the latest adjustment; an adjustment module, used to obtain the measured trajectory of the surgical robot end, and adjust the measured puncture force currently applied by the surgical robot end through the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

[0014] To achieve the above-mentioned purpose, the third aspect of this application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the intelligent spinal puncture surgical robot control method described in the first aspect when executing the computer program.

[0015] To achieve the above-mentioned purpose, the fourth aspect of the present application proposes a storage medium, which stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the intelligent spinal puncture surgical robot control method described in the first aspect.

[0016] The embodiments of the present application include at least the following beneficial effects: by measuring the measured puncture force of the surgical robot end in real time and predicting the expected puncture force using a Gaussian distribution model, it is possible to more accurately perceive the mechanical changes during the puncture process, thereby achieving precise control of the puncture depth and force, and improving the accuracy and success rate of the puncture. By dynamically adjusting the motion trajectory and operating force of the surgical robot based on the real-time measured puncture force and the expected puncture force predicted by the Gaussian distribution model, the robot can better adapt to the surgical needs of different patients and different tissue characteristics; the puncture force is monitored in real time and adjusted according to the expected trajectory and the measured trajectory, avoiding large deviations in the measured puncture force, reducing surgical risks, effectively improving the safety of the surgery, increasing the success rate of the surgery, and reducing the operational burden on medical staff.

[0017] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be achieved and obtained through the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0019] Figure 1 An optional flowchart of the intelligent spinal puncture surgical robot control method provided in an embodiment of the present application; Figure 2 An optional flow chart of the Gaussian process provided in the embodiment of the present application; Figure 3 An optional flowchart of the admittance process provided in an embodiment of the present application; Figure 4 A schematic diagram of an optional flow chart of the adaptive admittance process provided in an embodiment of the present application; Figure 5 A schematic diagram of an optional flow chart for calculating measurement error provided in an embodiment of the present application; Figure 6 An optional flowchart of the surgical robot control process provided in an embodiment of the present application; Figure 7 Another optional flow chart of the Gaussian process provided in the embodiment of the present application; Figure 8 A schematic diagram of an optional structure of the intelligent spinal puncture surgical robot control device provided in an embodiment of the present application; Figure 9A schematic diagram of an optional hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0021] In the description of this application, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number, and "above", "below", "within", etc. are understood to include the number.

[0022] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and the like in the specification, claims, or accompanying drawings are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0023] Among the related technologies, surgery has high technical requirements, and the doctor needs to have extremely high fine operation skills and rich clinical experience during the operation.

[0024] Based on this, the embodiments of the present application provide an intelligent spinal puncture surgical robot control method, device and electronic equipment, which can improve the safety of the operation, increase the success rate of the operation and reduce the operational burden of medical staff.

[0025] The intelligent spinal puncture surgical robot control method, device and electronic equipment provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the intelligent spinal puncture surgical robot control method in the embodiments of the present application is described.

[0026] The intelligent spinal puncture surgical robot control method provided in the embodiment of the present application relates to the field of computer technology. The intelligent spinal puncture surgical robot control method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server side, and can also be software running in a terminal or a server side. In some embodiments, the terminal can be a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the intelligent spinal puncture surgical robot control method, etc., but is not limited to the above forms.

[0027] The embodiments of the present application are further described below with reference to the accompanying drawings.

[0028] like Figure 1 and Figure 6 As shown, Figure 1 This is an optional flow chart of a method for controlling an intelligent spinal puncture surgical robot provided in an embodiment of the present application. The method for controlling an intelligent spinal puncture surgical robot can be executed by a server, or by a terminal, or by a server in conjunction with a terminal. The method for controlling an intelligent spinal puncture surgical robot includes but is not limited to the following steps S110 to S140: Step S110, measuring the puncture force of the surgical robot end in real time, and obtaining a first puncture force set within a preset time range, the first puncture force set including a plurality of historical puncture forces within the preset time range; Step S120: constructing a Gaussian distribution model, adjusting the Gaussian distribution model in real time using the latest first puncture force set, and predicting the expected puncture force at each time point using the newly adjusted Gaussian distribution model; Step S130, obtaining the measured trajectory of the surgical robot end, and adjusting the measured puncture force currently applied by the surgical robot end through the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

[0029] It is understandable that by measuring the puncture force at the end of the surgical robot in real time and predicting the expected puncture force using a Gaussian distribution model, it is possible to more accurately perceive the mechanical changes during the puncture process, thereby achieving precise control of the puncture depth and force, and improving the accuracy and success rate of the puncture. By dynamically adjusting the motion trajectory and operating force of the surgical robot based on the real-time measured puncture force and the expected puncture force predicted by the Gaussian distribution model, the robot can better adapt to the surgical needs of different patients and different tissue characteristics; by monitoring the puncture force in real time and adjusting the measured puncture force based on the expected trajectory and the measured trajectory, it can avoid large deviations in the measured puncture force, reduce surgical risks, effectively improve surgical safety, increase the success rate of surgery, and reduce the operational burden on medical staff.

[0030] A six-dimensional force sensor is integrated at the end of the surgical robot. The six-dimensional force sensor can measure the puncture force components (Fx, Fy, Fz) in three directions and the torque components (Mx, My, Mz) in three directions during the puncture process in real time, providing the surgical robot with comprehensive mechanical information. After receiving the mechanical data from the six-dimensional force sensor, the surgical robot first filters the mechanical data.

[0031] Optionally, a weighted moving average filtering algorithm is used to filter the mechanical component data in each direction. The specific formula of the weighted moving average filtering algorithm is: ; in, represents the filtering output result at sampling time k, represents the actual sampling value at sampling time k, N represents the window length of the filtering process, Represents the weight of the input sample value, which satisfies: .

[0032] In the process of testing the surgical robot and verifying the data algorithm, the method proposed in the embodiment of the present application allows analysis and comparison of different experimental data to determine the optimal sampling value weight and window length.

[0033] After preprocessing the force sampling values, the various force data obtained from real-time measurements are synthesized to obtain the measured puncture force. The peak resistance characteristics and force-position information change rate of the measured puncture force are calculated, and the above data characteristics are finely marked on the time axis to identify penetration events between different skin layers.

[0034] Specifically, the embodiment of the present application constructs a skin layer penetration recognition algorithm, which detects penetration events by accumulating puncture force errors to identify the different skin layers encountered by the surgical robot end during the acupuncture process, and then adjusts the various parameters and puncture force during the implementation of this method. The update formula of the puncture force cumulative error is specifically: ; in, It represents the cumulative error value (i.e., cusum value) at sampling time k, which indicates the degree of deviation between the current measured puncture force and the mean of all puncture force data within the sampling window with the current measured puncture force as the latest data. represents the measured puncture force at sampling time k, represents the mean of the puncture force data.

[0035] Specifically, the core idea of the formula of the skin layer penetration identification algorithm is to obtain the measured puncture force value through actual measurement and mean puncture force , the measured puncture force over any time window should be smooth, so It can also be called the theoretical value of the puncture force on the current skin layer. At any time k, when the measured puncture force does not reach the theoretical value hour, If it is a negative value, it can be identified that no skin layer penetration event occurs. At this time, the cumulative value max(0, ) is 0, and when When is greater than or equal to zero, the surgical robot end has penetrated the skin layer, and the above algorithm is used for accumulation; for example, assuming the theoretical skin penetration force is 5N, that is, =5, when the measured puncture force does not reach 5N, the max value (0, )=0, cusum value is 0, when it reaches 5N, cusum value=max(0, ), will continue to rise, which means that penetration occurs.

[0036] In the process of obtaining and measuring the puncture force, the position of the end of the surgical robot is obtained in real time, and the puncture force data is matched with the position data in real time. The historical position at the time point is used as input, and the historical puncture force corresponding to the first puncture force set is used as output. A Gaussian distribution model is constructed to predict the expected puncture force in the future relative to the first puncture force set.

[0037] The Gaussian distribution model, also known as the normal distribution model, is a widely used probability distribution model in probability theory and statistics. It describes the probability characteristics of random variables distributed around a certain mean in a symmetrical bell-shaped curve. The core concept of the Gaussian distribution model is that the distribution of most natural phenomena and random variables can be described by a mean and a standard deviation, and the distribution of these variables exhibits a bilaterally symmetrical bell-shaped curve. During acupuncture surgery, the Gaussian distribution model can be used to model and predict puncture force. During acupuncture surgery, the variation in puncture force is affected by multiple factors, including the mechanical properties of soft tissue, needle geometry, and puncture path. Using the Gaussian distribution model, historical puncture force data can be modeled and future puncture force changes can be predicted, thereby helping surgical robots dynamically adjust puncture strategies. During surgery, real-time puncture force often has a certain degree of uncertainty. The Gaussian distribution model can naturally generate uncertainty estimates, which is crucial for surgical safety. Through the Gaussian distribution model, the uncertainty range of the force can be provided while predicting the puncture force, thereby helping doctors or robots better cope with complex and dynamic surgical environments.

[0038] In some embodiments of the present application, Figure 2 As shown, Figure 1 Step S120 in the embodiment includes but is not limited to the following steps S210 to S230: Step S210, obtaining the historical position corresponding to each historical puncture force in the latest first puncture force set; Step S220: construct a kernel function based on historical positions: ; Step S230: Use the kernel function to update the Gaussian distribution model: ; Among them, x and Represent two historical positions respectively. Represents x and The Euclidean distance between represents the hyperparameters in the kernel function, exp represents the calculation of the natural logarithm, GP represents the Gaussian distribution, and m(x) is the mean function, which means to obtain the average value of the historical positions.

[0039] Specifically, the basic form of the Gaussian distribution model is: .

[0040] like Figure 7As shown in the figure, the Gaussian process uses the RBF kernel function as the kernel function. Since the RBF kernel function is smooth, continuous and unbounded, and has good locality and smoothness, the RBF kernel function can effectively capture the nonlinear relationship between the puncture force and the position. represents a hyperparameter within the kernel function that controls how the similarity between points in the input space decays with their distance. A value of indicates that the similarity between input points decays slowly, that is, farther points still have higher similarity, which makes the generated model tend to be smoother, while smaller A value of indicates that the similarity between input points decays faster, that is, only very close points have high similarity, which causes the generated function to be more localized and more volatile, making the model tend to fit local details.

[0041] It can be seen that the RBF kernel function The prediction effect of the Gaussian distribution model on the expected puncture force is determined. In some optional embodiments of the present application, the marginal likelihood function is constructed to predict the expected puncture force. Optimize, specifically, define the position x and puncture force y used in the training data, construct the marginal likelihood function, and maximize the marginal likelihood function. The specific formula is: ; Where X represents a vector composed of multiple historical positions, K(X,X) represents the covariance matrix of the vector composed of multiple historical positions, and n represents the number of historical positions. is the noise term, represents the measurement error, I is the identity matrix, and the optimal kernel function hyperparameters are obtained by maximizing the above marginal likelihood function.

[0042] The puncture force is predicted using the debugged Gaussian distribution model, and the expected data mean and expected data variance of the predicted data are calculated using the Gaussian distribution model. The specific formula is: ; in, represents the expected data mean, represents the expected data variance, represents the expected puncture force obtained by prediction, Y represents the vector composed of the expected puncture force, and K() represents the covariance matrix obtained.

[0043] The expected data mean and expected data variance can reflect the accuracy and effectiveness of the Gaussian distribution model's prediction. Based on the expected data mean and expected data variance, the various parameters in the Gaussian distribution model are fine-tuned, such as the kernel function parameters and noise level, to obtain the Gaussian distribution model again. After multiple iterations of debugging, the Gaussian distribution model with the best expected data mean and expected data variance is selected as the target to reduce the prediction error of the expected puncture force during the movement of the surgical robot terminal.

[0044] The trained Gaussian distribution model can quickly generate prediction functions based on historical data. During each puncture, newly generated sensor data is incorporated into the Gaussian distribution model's training set. By continuously updating the model, the surgical robot can adapt to the patient's tissue characteristics, anatomical structure, and potential variables that may occur during the procedure, achieving adaptive learning and improving prediction accuracy and robustness.

[0045] After predicting the puncture force of the surgical robot end through the Gaussian distribution model, during the movement of the surgical robot end, Figure 1 In step S130 , the surgical robot obtains the measured puncture force of the distal end in real time and adjusts the distal end's movement process in real time through an admittance control algorithm.

[0046] Admittance control is a robotic control strategy designed to adapt the robot's motion in response to externally applied forces or torques, thereby enabling the system to exhibit specific dynamic interaction characteristics. Its core concept is to calculate the desired motion adjustment based on external forces to achieve compliant and safe physical interaction.

[0047] Admittance control is widely used in medical robots. When a robot encounters resistance, it slows down or stops instead of forcibly advancing, thereby reducing the risk of damage to patient tissue. It can also respond to external force or torque input, making the robot compliant when in contact. The admittance control algorithm is usually expressed as: .

[0048] in, represents the desired inertia matrix of the robot; represents the desired damping matrix of the robot; represents the desired stiffness matrix of the robot; Indicates the actual acceleration of the robot tool end; Indicates the actual speed of the robot tool end; Indicates the actual position of the robot tool end; represents the desired acceleration of the robot tool end; represents the desired velocity of the robot tool end; represents the desired position of the robot tool end; Indicates the actual puncture force of the robot Expected puncture force The difference between the two is used to track the desired puncture force during control.

[0049] In general applications, the measured puncture force in the embodiment of the present application can be used as the actual puncture force in the admittance control process. To control the admittance process, such as Figure 3 As shown, in some embodiments of the present application, Figure 1 Step S130 in the embodiment includes but is not limited to the following steps S310 to S320: Step S310, obtaining real-time measured puncture force and determining the measurement position of the end of the surgical robot; Step S320: constructing an admittance control model based on the measured puncture force and the expected puncture force, and adjusting the measured puncture force at the measurement position using the admittance control model.

[0050] During the puncture operation, the acceleration and movement speed of the surgical robot end are very small. Therefore, in the algorithm formula of the above admittance process, the terms related to the actual acceleration and actual speed can be ignored. The algorithm of the admittance process of the surgical robot end can be simplified as follows: ; The required stiffness Directly setting it to zero can reduce the force steady-state error to zero, but this will cause the surgical robot end to lose its rigid response to external forces. This will make the system unable to effectively resist external interference and may cause the system to become too compliant, reducing its response speed, thereby affecting the dynamic performance of the system. Therefore, the reference trajectory It is given by: ; in is the environmental stiffness. From this formula, we can see that the reference trajectory By environmental location , environmental stiffness and the required puncture force However, the formula for the admittance process often requires accurate environmental information, which is often inaccurate or unknown in the actual operation of the robot. Therefore, traditional admittance control has difficulty achieving precise force control during the contact process. The embodiment of the present application simulates the precise compliance force control operation process performed by the medical team on the patient and proposes an adaptive admittance process.

[0051] like Figure 4As shown, Figure 3 In step S320, the adaptive admittance process includes but is not limited to the following steps S410 to S430: Step S410: Obtain a second puncture force set predicted by the Gaussian distribution model, wherein the second puncture force set includes a plurality of expected puncture forces predicted by the Gaussian distribution model, and the time points of the expected puncture forces correspond one-to-one to the historical puncture forces in the first puncture force set; Step S420, determining a measurement error by comparing each expected puncture force in the second puncture force set with the corresponding historical puncture force in the first puncture force set; Step S430 : adjusting the admittance control model constructed by the expected puncture force and the measured puncture force at the current time point according to the measurement error.

[0052] First, suppose and Respectively and The estimated value of , then the above equation can be expressed as: ; Substituting the estimated values of the environmental parameters into the puncture force equation and taking the difference, the puncture force can be obtained Estimated value of Estimated value of the measurement error between and measured puncture force: ; ; Simplifying the formula, we can get: ; in, , , .

[0053] From the above formula, we can see that in the admittance control model, the goal of the algorithm can be set to the parameter and When the time length t of the window approaches infinity, Infinitely close to ,and Infinitely close to , thereby achieving the desired penetration force target.

[0054] Adaptive control can be achieved by adjusting the target output of the puncture force according to the rate of change of the environmental stiffness coefficient. Assuming that the time interval for measuring the puncture force is infinitesimal, the accumulation process of the measurement error can be expressed as Figure 5 As shown in steps S510 to S530: Step S510: determining the first error between the historical puncture force and the corresponding expected puncture force ; Step S520, using the gain coefficient at the corresponding measurement position For the first error Adjust and get the second error ; Step S530: Integrate all the second errors to obtain the measurement error, where the measurement error is , from 0 to t represents the time range from the first puncture force set and the second puncture force set.

[0055] In short, according to the integration theory, the measurement errors in the first puncture force set and the second puncture force set are integrated, and the calculation formula can be expressed as: ; in, , is a preset constant.

[0056] Substituting the above integral theory into the admittance control model, the following new adaptive admittance control expression can be obtained: ; in, represents the expected inertia matrix, represents the desired damping matrix, represents the desired stiffness matrix, Indicates the current measured acceleration of the surgical robot end. represents the current measurement speed of the surgical robot end, x represents the current measurement position of the surgical robot end, Indicates the target puncture force currently pursued by the surgical robot end. represents the measurement error obtained from the first puncture force set and the second puncture force set, Indicates the current error gain coefficient, which is used to dynamically adjust the compensation strength of the measurement error. represents the historical puncture force in the first puncture force set, represents the desired puncture force in the second puncture force set.

[0057] The gain coefficient can be controlled during the debugging of the surgical robot and control algorithm. In a rigid environment, the system needs to adjust the integral term more carefully to avoid excessive force response, while in a softer environment, a stronger integral effect is required to overcome the error caused by environmental deformation. When the environment is relatively rigid (i.e., the environment is relatively rigid), the integral gain coefficient Ki will be small and the integral effect will be relatively weak; when the environment is relatively rigid (i.e., the environment is relatively soft), Ki will be large and the integral effect will be strong. An adaptive change rate is established based on the puncture force error, the accumulated force error, and the force error coefficient to achieve adaptive dynamic adjustment.

[0058] In addition, reference Figure 8 The present application also provides an intelligent spinal canal puncture surgical robot control device 800, comprising: The measurement module 801 is used to measure the puncture force of the surgical robot end in real time and obtain a first puncture force set within a preset time range, wherein the first puncture force set includes multiple historical puncture forces within the preset time range; Prediction module 802 is used to construct a Gaussian distribution model, adjust the Gaussian distribution model in real time using the latest first puncture force set, and predict the expected puncture force at each time point using the newly adjusted Gaussian distribution model; The adjustment module 803 is used to obtain the measured trajectory of the surgical robot end, and adjust the measured puncture force currently applied by the surgical robot end through the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

[0059] It is understandable that the specific implementation of the intelligent spinal puncture surgical robot control device 1200 is basically the same as the specific embodiment of the above-mentioned model training method, and will not be repeated here.

[0060] In addition, the embodiment of the present application also proposes a surgical robot, which is used to execute the above control method. Figure 6 As shown, first, the six-dimensional force sensor is used to collect the relevant force data of the end position and needle movement, the current gravity is measured in real time by the pressure sensor, the gravity error is eliminated by the self-weight bias, and the sensor data is subjected to weighted sliding average filtering. After filtering to remove noise, the key features are extracted for model training; then the Gaussian distribution model is used to predict the needle trajectory, and the adaptive admittance control technology is used to adjust the needle movement in real time to cope with the changes in tissue resistance; the needle insertion is executed by robot drive, and the actual trajectory is recorded and compared with the predicted trajectory to achieve visual monitoring; when it is detected that the needle has successfully penetrated the target, the puncture process is completed, and the data of this insertion is incremented to the training data set to continuously optimize the model accuracy.

[0061] If the surgical robot is not yet completed, it will re-execute the steps to adjust the puncture force at the distal end. The number of judgment cycles is set to five. If the judgment fails more than five times, the program will terminate and wait for the physician's confirmation to prevent safety accidents. It will also perform follow-up actions, such as exiting along the original path or being dragged out by the physician, and record an error log. The number of judgment failures is not shown in the figure. The entire program execution process is automatic. Once the puncture is in progress, the only two conditions that can trigger the shutdown are exceeding the safety limit force or an external emergency stop button.

[0062] Raw force data is collected using an ATI six-dimensional force sensor. After filtering to remove noise, the data is divided into multiple tissue interval slices and organized as training samples. Key features such as peak resistance and force conversion rate are then extracted from each slice, and a Gaussian process model is constructed based on a Gaussian kernel function and set hyperparameters. The model is trained using the training data, and its performance is verified using test data, outputting accuracy metrics such as the predicted mean and predicted variance. Finally, the system supports iterative loops, continuously optimizing the model's accuracy and generalization capabilities by updating the training data. This method achieves efficient processing and accurate prediction of force sensor data through a complete process of data preprocessing, feature extraction, model construction, and verification.

[0063] Secondly, the surgical robot's control system designed a real-time environmental elastic coefficient estimator. Based on the dynamic characteristics of the needle-tissue interaction during the puncture process, it dynamically senses changes in soft tissue mechanical parameters. Then, an adaptive law was constructed that takes the error between the desired force and the actual force as input. By dynamically adjusting the controller parameters, the uncertainty of the environmental stiffness is compensated, thereby achieving rapid and accurate realization of the desired contact force. Specifically, the actual contact force is compared with the desired contact force, and the resulting force error is converted into a displacement transformation. This is then converted into a position movement instruction for the robotic arm through inverse kinematics, thereby achieving rapid and accurate tracking and response of the desired force.

[0064] In addition, refer to Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes: The processor 901 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application. The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 902 and are called by the processor 901 to execute the intelligent spinal canal puncture surgical robot control method of the embodiment of this application, for example, to execute the above-described Figure 1 Steps S110 to S130 of the method, Figure 2 Steps S210 to S230 of the method, Figure 3 Steps S310 to S320 of the method, Figure 4 Steps S410 to S430 of the method, Figure 5 Steps S510 to S530 of the method; Input / output interface 903, used to implement information input and output; Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.); Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 ); The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0065] The embodiment of the present application also provides a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned intelligent spinal canal puncture surgical robot control method, for example, to execute the above-described Figure 1 Steps S110 to S130 of the method, Figure 2 Steps S210 to S230 of the method, Figure 3 Steps S310 to S320 of the method, Figure 4 Steps S410 to S430 of the method, Figure 5 Method steps S510 to S530.

[0066] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0067] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0068] It will be understood by those skilled in the art that Figures 1 to 5 The technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or a combination of certain steps, or different steps.

[0069] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0070] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0071] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0072] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0074] The units described above as separate components may or may not be physically separate, and 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 these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0077] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for controlling an intelligent spinal puncture surgical robot, characterized in that: include: measuring the measured puncture force of the surgical robot end in real time, and obtaining a first puncture force set within a preset time range, wherein the first puncture force set includes a plurality of historical puncture forces within the preset time range; Constructing a Gaussian distribution model, adjusting the Gaussian distribution model in real time using the latest first puncture force set, and predicting the expected puncture force at each time point using the latest adjusted Gaussian distribution model; Obtain the measured trajectory of the surgical robot end, and adjust the measured puncture force currently applied by the surgical robot end based on the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

2. The intelligent spinal puncture surgical robot control method according to claim 1, characterized in that: The constructing of the Gaussian distribution model and adjusting the Gaussian distribution model in real time using the latest first puncture force set includes: Obtaining a historical position corresponding to each of the historical puncture forces in the latest first puncture force set; Construct a kernel function using the historical positions: ; Update the Gaussian distribution model using the kernel function: ; Among them, x and Respectively represent the two historical positions, Represents x and The Euclidean distance between represents the hyperparameter in the kernel function, exp represents the calculation of the natural logarithm, GP represents the Gaussian distribution, and m(x) is the mean function, which represents the average value of the historical position.

3. The intelligent spinal puncture surgical robot control method according to claim 2, characterized in that: Before constructing the kernel function using the historical positions, the method further includes: Construct the marginal likelihood function using the historical positions: ； Performing maximum likelihood estimation on the marginal likelihood function to obtain the hyperparameter; Wherein, X represents a vector composed of multiple historical positions, K(X,X) represents the covariance matrix of the vector composed of multiple historical positions, and n represents the number of historical positions.

4. The intelligent spinal puncture surgical robot control method according to claim 1, characterized in that: The step of adjusting the measured puncture force currently applied by the surgical robot end using the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point includes: Obtaining real-time measured puncture force and determining the measurement position of the distal end of the surgical robot; An admittance control model is constructed based on the measured puncture force and the expected puncture force, and the measured puncture force is adjusted at the measurement position by using the admittance control model.

5. The intelligent spinal puncture surgical robot control method according to claim 4, characterized in that: Before adjusting the measured puncture force at the measuring position by the admittance control model, the method further includes: Obtaining a second puncture force set predicted by the Gaussian distribution model, wherein the second puncture force set includes a plurality of expected puncture forces predicted by the Gaussian distribution model, and time points of the expected puncture forces correspond one-to-one to the historical puncture forces in the first puncture force set; determining a measurement error by comparing each of the expected puncture forces in the second puncture force set with the corresponding historical puncture forces in the first puncture force set; The admittance control model constructed by the expected puncture force and the measured puncture force at the current time point is adjusted according to the measurement error.

6. The intelligent spinal puncture surgical robot control method according to claim 5, characterized in that: The measured puncture force represents an estimated value of the actual puncture force between the surgical robot tip and the skin layer. The adjusted admittance control model is: ； in, represents the expected inertia matrix, represents the desired damping matrix, represents the desired stiffness matrix, represents the current measured acceleration of the surgical robot end, represents the current measurement speed of the surgical robot end, x represents the current measurement position of the surgical robot end, represents the target puncture force currently pursued by the surgical robot terminal, represents the measurement error obtained by the first puncture force set and the second puncture force set, Indicates the current error gain coefficient, which is used to dynamically adjust the compensation strength of the measurement error. represents the historical puncture force in the first puncture force set, represents the expected puncture force in the second puncture force set.

7. The intelligent spinal puncture surgical robot control method according to claim 6, characterized in that: The specific process of determining the measurement error is as follows: Sequentially determine a first error between the historical puncture force and the corresponding expected puncture force ; By the gain coefficient corresponding to the measurement position For the first error Adjust and get the second error ; All the second errors are integrated to obtain the measurement error, wherein the measurement error is , from 0 to t represents the time range from the first puncture force set and the second puncture force set.

8. An intelligent spinal puncture surgical robot control device, characterized in that: include: a measurement module, configured to measure the puncture force of the distal end of the surgical robot in real time and obtain a first puncture force set within a preset time range, wherein the first puncture force set includes a plurality of historical puncture forces within the preset time range; a prediction module, configured to construct a Gaussian distribution model, adjust the Gaussian distribution model in real time using the latest set of the first puncture forces, and predict the expected puncture force at each time point using the latest adjusted Gaussian distribution model; An adjustment module is used to obtain the measured trajectory of the surgical robot end, and adjust the measured puncture force currently applied by the surgical robot end through the measured trajectory, the preset expected trajectory, the measured puncture force obtained by real-time measurement, and the expected puncture force corresponding to the same time point.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the intelligent spinal canal puncture surgical robot control method according to any one of claims 1 to 7 when executing the computer program.

10. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the intelligent spinal puncture surgical robot control method according to any one of claims 1 to 7 is implemented.

Images