Personalized man-machine coupling digital twin model construction method and system

Abstract

The invention provides a personalized man-machine coupling digital twinning model construction method and system, and relates to the technical field of digital twinning, and the method comprises the steps: constructing a man-machine coupling lower limb exoskeleton rehabilitation robot digital twinning model constructed for a target rehabilitation patient; the model is coupled with a rehabilitation patient musculoskeletal system digital twinborn model and a lower limb exoskeleton rehabilitation robot digital twinborn model. The initial rehabilitation parameters are input into the human-machine coupling lower limb exoskeleton rehabilitation robot digital twinborn model for simulation, and simulation physiological feature information of a key muscle group is obtained; and comparing the simulation physiological feature information with evaluation information from experts, and verifying the reliability and effectiveness of the theoretical method provided by the invention.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a method and system for constructing a personalized human-computer coupled digital twin model. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] With the aging population and rising incidence of cardiovascular and cerebrovascular diseases, the rehabilitation needs of patients with lower limb motor dysfunction are increasing. Lower limb exoskeleton rehabilitation robots, as an effective assistive treatment device, can help patients perform repetitive and standardized rehabilitation training, thereby promoting neurological remodeling and muscle function recovery.

[0004] Digital twins, as a core technology for achieving deep integration of cyber-physical systems, construct high-fidelity virtual models of physical entities and utilize real-time data to drive state mapping, process monitoring, and predictive optimization of these entities. Applying digital twin technology to the field of rehabilitation medicine, constructing human-machine coupled digital twin models (i.e., rehabilitation patients and lower limb exoskeleton robot systems), provides a novel technical approach for developing precise, personalized, and adaptive rehabilitation plans.

[0005] However, existing technologies still face many challenges in developing personalized rehabilitation plans based on digital twins: (1) The challenges of constructing and dynamically following human-machine coupled digital twin models Constructing a high-fidelity human-machine coupled digital twin model is the core foundation. The challenge lies in accurately describing the coupling relationship and energy interaction between the complex electromechanical characteristics of the lower limb exoskeleton robot and the physiological characteristics of the rehabilitation patient's musculoskeletal system. Furthermore, this model is not static; the rehabilitation patient's musculoskeletal condition continuously improves with training, while the exoskeleton robot's mechanical performance gradually degrades with use. Existing models often lack effective dynamic update mechanisms, failing to maintain consistency with the physical entity's state throughout the long rehabilitation cycle, leading to decreased accuracy in simulation analysis and solution development based on outdated models.

[0006] (2) The problem of efficiently solving complex digital twin models A high-fidelity digital twin model encompassing all elements of a human-machine system is typically extremely complex, consisting of numerous multi-domain differential-algebraic equations. Directly simulating and solving such models is computationally intensive and time-consuming, severely limiting its application in online analysis and real-time decision-making scenarios requiring rapid response. This computational efficiency bottleneck makes it difficult for digital twin technology to achieve truly "online" applications in clinical rehabilitation, failing to provide therapists with immediate and effective decision support. Summary of the Invention

[0007] To address at least one of the technical problems mentioned above, this invention proposes a personalized human-machine coupled digital twin model construction method and system. Based on the concept of digital twins, it solves the problems of dynamic model following changes and efficient solution in the construction and application services of digital twin models for human-machine coupled lower limb exoskeleton rehabilitation robots.

[0008] The first aspect of this invention provides a method for constructing a personalized human-computer coupled digital twin model, comprising: A human-machine coupled digital twin model of a lower limb exoskeleton rehabilitation robot for target rehabilitation patients is constructed; the model couples the digital twin model of the rehabilitation patient's musculoskeletal system with the digital twin model of the lower limb exoskeleton rehabilitation robot. The initial rehabilitation parameters are input into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation to obtain the simulated physiological characteristic information of key muscle groups; The simulated physiological characteristics information is compared with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

[0009] Furthermore, the construction steps of the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot include: Mathematical models of the mechanical, electrical, and control subsystems of the lower limb exoskeleton rehabilitation robot are obtained. Through the component Connector connection mechanism and energy converter, the models of the subsystems are integrated into a multi-domain unified modeling and simulation environment to construct a digital twin model of the lower limb exoskeleton rehabilitation robot. We acquire magnetic resonance imaging information and clinical perception data of rehabilitation patients, construct a parameterized lower limb muscle model based on the concept of parameterized modeling, and identify muscle constitutive parameters based on Neo-Hookean constitutive model and genetic algorithm to construct an individualized digital twin model of the musculoskeletal system of rehabilitation patients. By analyzing the information flow relationship between the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system, the two are coupled to construct a human-machine coupled lower limb rehabilitation system digital twin model.

[0010] Furthermore, the construction steps of the parameterized lower limb muscle model include: comparing the muscle feature parameters with the Hausdorff distance based on the magnetic resonance imaging information model, and adjusting the feature parameters when the similarity requirement is not met, until a parameterized lower limb muscle model that can represent individual characteristics under different movement states is established.

[0011] Furthermore, the human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model is dynamically followed through the following steps: Acquire real-time sensing signals and performance status data during rehabilitation training, and determine whether the system status has changed through signal analysis; If the state changes, the change time point is accurately located through continuous wavelet transform, and the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system are dynamically updated based on the physical damage model and the genetic algorithm parameter identification strategy. The simulation information of the updated model is compared and verified with the perception information of the physical entity, and the model is corrected when the perception information and simulation information do not match.

[0012] Furthermore, the dynamic updating of the digital twin model of the musculoskeletal system of the rehabilitation patient includes: identifying sensitive parameters of the muscle tissue constitutive model through a genetic algorithm, calculating the changes in sensitive parameters by combining the sensory data of the rehabilitation patient, and correcting the parameters by measuring biological information in a limited and non-destructive manner to ensure the consistency of muscle physiological characteristics. Dynamic updates to the digital twin model of the lower limb exoskeleton rehabilitation robot include: establishing a comprehensive performance damage model based on Arcard adhesive wear theory, nonlinear fatigue damage accumulation theory, or fatigue crack propagation model; mapping the performance degradation of components during operation; and realizing autonomous updates of the digital twin model of the lower limb exoskeleton rehabilitation robot at the physical level.

[0013] Furthermore, before simulation, the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot is efficiently solved through the following steps: Obtain the mass matrix and stiffness matrix of the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot; Based on the impact evaluation method of the maximum error limit in the time domain and the Krylov subspace projection method, the model is adaptively reduced in order. The simulation results of the reduced-order model are compared with the results of the full model to verify whether the reduced-order model meets the application requirements.

[0014] A second aspect of the present invention provides a personalized human-computer coupled digital twin model construction system, comprising: The human-machine coupled digital twin model construction module is used to construct a human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model for a target rehabilitation patient; the model couples the rehabilitation patient's musculoskeletal system digital twin model with the lower limb exoskeleton rehabilitation robot digital twin model; The rehabilitation parameter simulation module is used to input the initial rehabilitation parameters into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation, and to obtain the simulated physiological characteristic information of key muscle groups. The verification module is used to compare the simulated physiological feature information with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

[0015] A third aspect of the present invention provides an electronic device including a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps in the personalized human-machine coupled digital twin model construction method as described in the first aspect of the present invention.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the steps of the personalized human-computer coupled digital twin model construction method as described in the first aspect of the present invention.

[0017] A fifth aspect of the present invention provides a computer program product comprising software code, wherein the program in the software code performs the steps of the personalized human-computer coupled digital twin model construction method described in the first aspect of the present invention.

[0018] Compared with existing technologies, the personalized human-computer coupled digital twin model construction method and system provided by this invention have the following beneficial effects: (1) To address the challenges of constructing and dynamically tracking human-machine coupled digital twin models, this invention constructs an exoskeleton robot model and a patient musculoskeletal model based on parametric modeling and Neo-Hookean constitutive models using a multi-domain unified modeling language (such as Modelica), and analyzes their information flow to achieve coupling. Furthermore, through real-time signal analysis, continuous wavelet transform, physical damage models (such as Archard wear theory), and genetic algorithm parameter identification strategies, dynamic and accurate updates of the digital twin model are achieved. This effectively ensures a high degree of consistency between the virtual model and the physical entity's state throughout the rehabilitation cycle, providing an accurate model foundation for rehabilitation program analysis and improving the reliability and effectiveness of the program.

[0019] (2) To address the challenge of efficiently solving complex digital twin models, this invention utilizes Model Order Reduction (MOR) techniques, such as the influence assessment method based on the maximum error limit in the time domain and the Krylov subspace projection method, to adaptively reduce the order of high-fidelity models. While preserving the dominant dynamic characteristics of the system, this significantly reduces the complexity and computational load of the model, enabling rapid simulation and solution. This significantly improves simulation efficiency, making online analysis and real-time parameter optimization based on digital twins possible, and ensuring the timeliness of rehabilitation service analysis.

[0020] (3) This invention uses the constructed and optimized digital twin model for the simulation and optimization of rehabilitation programs. Through the closed-loop process of "input parameters-simulation-comparison-optimization", it is possible to perform pre-simulation analysis on personalized rehabilitation parameters, predict the rehabilitation effects and potential risks (such as muscle overstretching) under different parameters, thereby assisting therapists in developing more scientific, safe and effective personalized rehabilitation programs, avoiding the risk of secondary injury that may be caused by traditional trial and error methods. Attached Figure Description

[0021] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0022] Figure 1 This is a flowchart of the personalized human-machine coupled digital twin model construction method provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the overall technical route of the method provided in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the technical route for constructing a digital twin model of a human-machine coupled lower limb exoskeleton rehabilitation robot provided in Embodiment 1 of the present invention; Figure 4 This is a flowchart of the dynamic following strategy for the human-machine coupled digital twin model provided in Embodiment 1 of the present invention; Figure 5 This is a flowchart of the efficient solution method for human-machine coupled digital twin models provided in Embodiment 1 of the present invention; Figure 6 This is a schematic diagram of the overall technical route for model-based applications provided in Embodiment 1 of the present invention; Figure 7 This is a schematic diagram of the personalized human-machine coupled digital twin model construction system provided in Embodiment 2 of the present invention. Detailed Implementation

[0023] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0024] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0026] All data acquisition in this embodiment is carried out in accordance with laws and regulations and with user consent, and the data is used legally.

[0027] Example 1 like Figure 1 This embodiment provides a method for constructing a personalized human-computer coupled digital twin model, including: A human-machine coupled digital twin model of a lower limb exoskeleton rehabilitation robot for target rehabilitation patients is constructed; the model couples the digital twin model of the rehabilitation patient's musculoskeletal system with the digital twin model of the lower limb exoskeleton rehabilitation robot. The initial rehabilitation parameters are input into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation to obtain the simulated physiological characteristic information of key muscle groups; The simulated physiological characteristics information is compared with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

[0028] The overall technical solution of the present invention is as follows: Figure 2 As shown, by Figure 2 Therefore, the research approach of this invention is as follows: (1) A digital twin model of the lower limb exoskeleton rehabilitation robot was established using a multi-domain unified modeling language. A descriptive model of the lower limb musculoskeletal system of the rehabilitation patient was established based on modeling theories such as object-oriented thinking and knowledge in fields such as clinical perception data, musculoskeletal composition, and simulation mechanisms. A human-machine coupled digital twin model was established by studying the coupling mechanism between the lower limb exoskeleton rehabilitation robot and the digital twin model of the musculoskeletal system. During all stages of the rehabilitation cycle of the lower limb musculoskeletal system of the rehabilitation patient, the perceived signals were evaluated by signal accumulation state response analysis to determine whether the clinical state of the musculoskeletal system and the performance state of the rehabilitation robot had changed. If changes occurred, the model was updated through the model dynamic update module to achieve dynamic following of changes. Afterwards, the model was verified for application requirements. When the perceived information and simulation information did not match, the model was corrected to improve the consistency between the virtual and real aspects of the human-machine coupled digital twin model.

[0029] (2) The model reduction algorithm is deployed in the cloud in a cloud-edge collaborative manner to improve the computational efficiency and response speed of the model. Specifically, the computational tasks of the model reduction algorithm are distributed to multiple servers in the cloud for parallel computation, thereby accelerating the reduction process. At the same time, the cloud can also utilize high-speed network connections and storage resources to provide more computational and storage space for model reduction, so as to better handle the reduction tasks of large models.

[0030] (3) By dynamically following the efficient simulation of the digital twin model, the characteristic information of key muscle groups such as the soleus muscle is solved, so as to conduct pre-analysis of personalized rehabilitation parameters (and avoid secondary injury to rehabilitation patients), etc., to assist rehabilitation therapists in optimizing personalized rehabilitation parameters, and verify the reliability and effectiveness of the theoretical method of this invention through rehabilitation effect.

[0031] Specifically, the steps for constructing the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot include: Mathematical models of the mechanical, electrical, and control subsystems of the lower limb exoskeleton rehabilitation robot are obtained. Through the component Connector connection mechanism and energy converter, the models of the subsystems are integrated into a multi-domain unified modeling and simulation environment to construct a digital twin model of the lower limb exoskeleton rehabilitation robot. We acquire magnetic resonance imaging information and clinical perception data of rehabilitation patients, construct a parameterized lower limb muscle model based on the concept of parameterized modeling, and identify muscle constitutive parameters based on Neo-Hookean constitutive model and genetic algorithm to construct an individualized digital twin model of the musculoskeletal system of rehabilitation patients. By analyzing the information flow relationship between the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system, the two are coupled to construct a human-machine coupled lower limb rehabilitation system digital twin model.

[0032] In one specific embodiment, utilizing object-oriented principles, the lower limb exoskeleton rehabilitation robot is divided into three subsystems: mechanical, electrical, and control. Based on a multi-domain modeling language, each subsystem is mathematically described using equations. A component connector mechanism is used to establish connection interfaces between components within each subsystem. Through an energy converter, the subsystem models are integrated into a unified multi-domain modeling, simulation, and optimization environment to form a digital twin model of the lower limb exoskeleton rehabilitation robot. The technical approach is as follows: Figure 3 As shown.

[0033] Specifically, the steps for constructing the parameterized lower limb muscle model include: comparing the muscle feature parameters with the Hausdorff distance between the magnetic resonance imaging information model and adjusting the feature parameters when the similarity requirement is not met, until a parameterized lower limb muscle model that can represent individual characteristics under different movement states is established.

[0034] Similarly, utilizing object-oriented principles and based on clinical perception data and domain knowledge (musculoskeletal composition, simulation mechanisms, etc.), a descriptive model of the lower limb musculoskeletal system of rehabilitation patients is established. First, based on parametric modeling, muscle tissue features are parametrically represented using magnetic resonance imaging (MRI) information and information processing and feature reconstruction techniques. Next, the Hausdorff distance between the muscle feature parameters and the MRI-based model is compared. If the similarity requirement is not met, the feature parameters are adjusted until a parametric lower limb muscle model capable of representing individual characteristics under different movement states is established. Simultaneously, the Neo-Hookean model is selected as the constitutive model for muscle tissue, with its strain energy density function being: ; in, and These are shear modulus and bulk modulus, respectively. As the first strain invariant, The volume ratio is used. Then, based on a genetic algorithm, the core parameters in the model are identified, and constitutive models of muscles with different degrees of damage can be constructed by combining the sensory information of the rehabilitation patient. Then, the personalized constitutive model is applied to the established parameterized lower limb muscle tissue model to create a digital twin model of the rehabilitation patient's muscle tissue. Next, individual rehabilitation information is obtained through vital sign parameter calibration. Guided by existing knowledge of human musculoskeletal modeling and model scaling technology, a basic digital twin model of the rehabilitation patient's musculoskeletal system can be established. Finally, the lower limb muscle tissue model is applied to the established basic digital twin model of the musculoskeletal system to create an individualized digital twin model of the rehabilitation patient's musculoskeletal system. This digital twin model can be used to evaluate the dynamic performance of the musculoskeletal system and simulate the dynamic behavior and mechanical characteristics of the rehabilitation patient's musculoskeletal system.

[0035] A reasonable analysis of the correlation between the digital twin models and the assistive function relationship between the lower limb exoskeleton rehabilitation robot and the rehabilitation patient is a key foundation and prerequisite for achieving model coupling. By analyzing the information flow correlation between the digital twin model of the lower limb exoskeleton rehabilitation robot and the lower limb musculoskeletal simulation domain knowledge of the rehabilitation patient, the coupling of the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the lower limb musculoskeletal system of the rehabilitation patient can be achieved, ultimately establishing a basic digital twin model of the lower limb rehabilitation system.

[0036] Specifically, the human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model is dynamically followed through the following steps: Acquire real-time sensing signals and performance status data during rehabilitation training, and determine whether the system status has changed through signal analysis; If the state changes, the change time point is accurately located through continuous wavelet transform, and the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system are dynamically updated based on the physical damage model and the genetic algorithm parameter identification strategy. The simulation information of the updated model is compared and verified with the perception information of the physical entity, and the model is corrected when the perception information and simulation information do not match.

[0037] Specifically, the dynamic updating of the digital twin model of the musculoskeletal system of the rehabilitation patient includes: identifying sensitive parameters of the muscle tissue constitutive model through a genetic algorithm, calculating the changes in sensitive parameters by combining the sensory data of the rehabilitation patient, and correcting the parameters by measuring biological information in a limited and non-destructive manner to ensure the consistency of muscle physiological characteristics. Dynamic updates to the digital twin model of the lower limb exoskeleton rehabilitation robot include: establishing a comprehensive performance damage model based on Arcard adhesive wear theory, nonlinear fatigue damage accumulation theory, or fatigue crack propagation model; mapping the performance degradation of components during operation; and realizing autonomous updates of the digital twin model of the lower limb exoskeleton rehabilitation robot at the physical level.

[0038] In one specific embodiment, the dynamically following digital twin model is a dynamically changing digital model that continuously learns from data and updates its own knowledge to dynamically adjust the model's parameters, thereby improving the model's applicability throughout the entire rehabilitation cycle. The technical route of the human-machine coupled digital twin model dynamic following strategy is as follows: Figure 4 As shown, it mainly covers the following four modules: Status monitoring: Throughout the entire recovery cycle of the lower limb musculoskeletal system in rehabilitation patients, the system assesses and determines whether changes have occurred in the clinical status of the musculoskeletal system and the performance status of the rehabilitation robot through cumulative signal state response analysis. For example, Fourier transform is used to perform time-frequency domain analysis on the sensed information, and the frequency band is used to roughly determine whether changes in the performance status of the rehabilitation robot have occurred.

[0039] Model Dynamic Updates: When changes in performance status are detected, for lower limb exoskeleton rehabilitation robots, continuous wavelet transform can be used to analyze the sensed signals, and the Lipschitz exponent can be used to update the model. It effectively enables precise timing of performance status change signals.

[0040] ; This approach avoids wasting storage and computing resources while accurately pinpointing the performance degradation transition time. Simultaneously, it evaluates the cumulative signal response at various pre-selected sensor locations, estimating the degree, location, and direction of damage. A comprehensive performance damage model is established using Archard adhesive wear theory, nonlinear fatigue damage accumulation theory, and fatigue crack propagation models, combined with mapping data from the physical rehabilitation robot. This model maps the performance degradation of the physical rehabilitation robot caused by fatigue damage and wear of components during operation. Local degradation at material points is quantified by introducing internal state variables, and the evolution equation of the state variables is customized using a void growth model. Through this comprehensive performance damage model, the digital twin model achieves autonomous updates of performance degradation at the physical level. Furthermore, dynamic updates of the physical rehabilitation robot can be achieved at the geometric level using fractal theory and point cloud reconstruction mechanisms. For the lower limb muscle tissue of rehabilitation patients, a genetic algorithm is first used to identify the sensitive parameters of the muscle tissue constitutive model. Then, combined with the patient's sensory data and the Neo-Hookean model (strain energy density function...), the model is updated accordingly. ,in μ Shear modulus κ Bulk modulus As the first strain invariant, J The change in sensitive parameters is calculated based on the volume ratio. Finally, the parameters are corrected by measuring the amount of bioinformation in a limited and non-destructive manner to ensure the consistency of the physiological characteristics of the core muscles related to the development of the rehabilitation program, thereby ensuring the accuracy of the dynamic following characteristics.

[0041] Model Validation: The model's suitability for the application will be determined by comparing the features of the perceived physical entity information with those of the simulation information based on the updated model using algorithms such as the Pearson correlation coefficient. Error thresholds can be set according to different application services to limit the model's applicability.

[0042] Model correction: When there is a mismatch between perceived information and simulation information, model correction is performed. First, the sensitive parameters of the physical entity are identified. Then, the correction values ​​of the sensitive parameters are defined through a genetic algorithm. Finally, the consistency between the virtual and real aspects of the human-machine coupled digital twin model is further improved through parameter adjustment.

[0043] This dynamic, coupled digital twin model can be applied to different stages of rehabilitation for patients, realistically reflecting the clinical status of patients and the performance status of rehabilitation equipment, thereby meeting the simulation service needs of patients at different stages of the rehabilitation cycle.

[0044] Specifically, before simulation, the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot is efficiently solved through the following steps: Obtain the mass matrix and stiffness matrix of the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot; Based on the impact evaluation method of the maximum error limit in the time domain and the Krylov subspace projection method, the model is adaptively reduced in order. The simulation results of the reduced-order model are compared with the results of the full model to verify whether the reduced-order model meets the application requirements.

[0045] In one specific embodiment, the efficient solution technique for the digital twin model of the dynamically following human-machine coupled lower limb exoskeleton rehabilitation robot designed in this invention is as follows: Figure 5 As shown.

[0046] This invention presents an efficient model-solving method based on a reduction-order algorithm. It utilizes the influence evaluation method based on the time-domain maximum error limit and the Krylov subspace projection method to study an adaptive reduction-order method for a digital twin model of a human-machine coupled lower limb exoskeleton rehabilitation robot. This method is encapsulated into a dynamic link library (DLL) to autonomously optimize and reduce the order of the described model, thereby reducing its complexity and computational load. This improves the efficiency of model simulation while preserving the system's necessary behavioral characteristics and dominant effects. During the model reduction process, firstly, the robot's inherent information processing module extracts feature information such as the mass matrix and stiffness matrix. Then, this information is passed to the encapsulated model reduction DLL to reduce the model's order, obtaining a base model with a reduced order. Finally, based on the full model simulation information and the simulation information of the reduced-order base model, the time and accuracy dimensions are used to determine whether the reduced-order base model meets the application service requirements. If it does not meet the application service requirements, the reduction parameters are readjusted, and the reduced-order model is optimized until it meets the requirements.

[0047] The model reduction algorithm is deployed in a cloud-edge collaborative environment. Specifically, the computational tasks of the model reduction algorithm are distributed across multiple servers in the cloud for parallel computation, thereby accelerating the reduction process and improving the model's computational efficiency and response speed. Edge devices facilitate rapid data processing, and pre-processed data is uploaded to the cloud-based human-machine perception database, fully utilizing the cloud's high-speed network connectivity and storage resources to provide greater computational and storage space for model reduction. Information such as the pre-analysis of the rehabilitation plan obtained from the cloud solution is downloaded to the edge device to obtain an optimized rehabilitation plan. The execution parameters of the optimized rehabilitation plan are obtained through analysis, and these parameters are then assigned to the lower limb exoskeleton rehabilitation robot to complete the rehabilitation work.

[0048] The above research can solve the deployment problem of model order reduction algorithms and the problem of co-simulation with cloud edge collaborative technology. From the two aspects of simulation task allocation and model solution matrix optimization, it can realize the rapid simulation of large-scale models and improve the simulation efficiency.

[0049] Based on the research on three aspects—the construction of a digital twin model of a human-machine coupled lower limb exoskeleton rehabilitation robot, the model's dynamic following strategy, and efficient model solving—application service analysis is conducted, including personalized pre-analysis of rehabilitation parameters for the lower limb exoskeleton rehabilitation robot based on the system model (technical route as follows). Figure 6 (As shown). Specifically, the system uses a dynamically tracked digital twin model to efficiently simulate and solve for the characteristic information of key muscle groups such as the soleus muscle, enabling pre-analysis of personalized rehabilitation parameters (which can also prevent secondary injury to the rehabilitation patient) and assisting rehabilitation therapists in optimizing these parameters. First, the initial rehabilitation parameters provided by the physical therapist are loaded into the simulation environment of the human-machine coupled lower limb exoskeleton rehabilitation robot's digital twin model as boundary conditions to obtain clinical information such as energy metabolism and muscle strength of core muscle tissues like the soleus muscle. This simulation information is then compared with the assessment information provided by the therapist. If there is a mismatch, the initial rehabilitation parameters are optimized until an optimal personalized rehabilitation plan for the patient is developed. The accuracy and applicability of the human-machine coupled lower limb exoskeleton rehabilitation robot's digital twin model can also be verified during the development of the personalized rehabilitation plan.

[0050] Example 2 like Figure 7 As shown, this embodiment provides a personalized human-computer coupled digital twin model construction system, including: The human-machine coupled digital twin model construction module is used to construct a human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model for a target rehabilitation patient; the model couples the rehabilitation patient's musculoskeletal system digital twin model with the lower limb exoskeleton rehabilitation robot digital twin model; The rehabilitation parameter simulation module is used to input the initial rehabilitation parameters into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation, and to obtain the simulated physiological characteristic information of key muscle groups. The verification module is used to compare the simulated physiological feature information with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

[0051] Example 3 Embodiment 3 of the present invention provides an electronic device.

[0052] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the personalized human-machine coupled digital twin model construction method as described in Embodiment 1 of the present invention.

[0053] The detailed steps are the same as those provided in Example 1 for constructing a personalized human-machine coupled digital twin model, and will not be repeated here.

[0054] Example 4 Embodiment 4 of the present invention provides a computer-readable storage medium.

[0055] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the personalized human-computer coupled digital twin model construction method as described in Embodiment 1 of the present invention.

[0056] The detailed steps are the same as those provided in Example 1 for constructing a personalized human-machine coupled digital twin model, and will not be repeated here.

[0057] Example 5 Embodiment 5 of the present invention provides a computer program product.

[0058] A computer program product includes software code, wherein the program in the software code performs the steps of the personalized human-computer coupled digital twin model construction method as described in Embodiment 1 of the present invention.

[0059] The detailed steps are the same as those provided in Example 1 for constructing a personalized human-machine coupled digital twin model, and will not be repeated here.

[0060] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0061] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A method for constructing a personalized human-computer coupled digital twin model, characterized in that, include: Construct a human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model for target rehabilitation patients; The model couples a digital twin model of the musculoskeletal system of a rehabilitation patient with a digital twin model of a lower limb exoskeleton rehabilitation robot. The initial rehabilitation parameters are input into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation to obtain the simulated physiological characteristic information of key muscle groups; The simulated physiological characteristics information is compared with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

2. The method as described in claim 1, characterized in that, The steps for constructing the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot include: Mathematical models of the mechanical, electrical, and control subsystems of the lower limb exoskeleton rehabilitation robot are obtained. Through the component Connector connection mechanism and energy converter, the models of the subsystems are integrated into a multi-domain unified modeling and simulation environment to construct a digital twin model of the lower limb exoskeleton rehabilitation robot. We acquire magnetic resonance imaging information and clinical perception data of rehabilitation patients, construct a parameterized lower limb muscle model based on the concept of parameterized modeling, and identify muscle constitutive parameters based on Neo-Hookean constitutive model and genetic algorithm to construct an individualized digital twin model of the musculoskeletal system of rehabilitation patients. By analyzing the information flow relationship between the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system, the two are coupled to construct a human-machine coupled lower limb rehabilitation system digital twin model.

3. The method as described in claim 2, characterized in that, The steps for constructing the parameterized lower limb muscle model include: comparing the muscle feature parameters with the Hausdorff distance based on magnetic resonance imaging information model, and adjusting the feature parameters when the similarity requirement is not met, until a parameterized lower limb muscle model that can represent individual characteristics under different movement states is established.

4. The method as described in claim 1, characterized in that, The human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model is dynamically followed through the following steps: Acquire real-time sensing signals and performance status data during rehabilitation training, and determine whether the system status has changed through signal analysis; If the state changes, the change time point is accurately located through continuous wavelet transform, and the digital twin model of the lower limb exoskeleton rehabilitation robot and the digital twin model of the rehabilitation patient's musculoskeletal system are dynamically updated based on the physical damage model and the genetic algorithm parameter identification strategy. The simulation information of the updated model is compared and verified with the perception information of the physical entity, and the model is corrected when the perception information and simulation information do not match.

5. The method as described in claim 4, characterized in that, Dynamically updating the digital twin model of the musculoskeletal system of the rehabilitation patient includes: identifying sensitive parameters of the muscle tissue constitutive model through a genetic algorithm, calculating the changes in sensitive parameters by combining the patient's sensory data, and correcting the parameters by measuring biological information in a limited and non-destructive manner to ensure the consistency of muscle physiological characteristics. Dynamic updates to the digital twin model of the lower limb exoskeleton rehabilitation robot include: establishing a comprehensive performance damage model based on Arcard adhesive wear theory, nonlinear fatigue damage accumulation theory, or fatigue crack propagation model; mapping the performance degradation of components during operation; and realizing autonomous updates of the digital twin model of the lower limb exoskeleton rehabilitation robot at the physical level.

6. The method as described in claim 1, characterized in that, Before simulation, the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot is efficiently solved through the following steps: Obtain the mass matrix and stiffness matrix of the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot; Based on the impact evaluation method of the maximum error limit in the time domain and the Krylov subspace projection method, the model is adaptively reduced in order. The simulation results of the reduced-order model are compared with the results of the full model to verify whether the reduced-order model meets the application requirements.

7. A personalized human-computer coupled digital twin model construction system, characterized in that, include: The human-machine coupled digital twin model construction module is used to construct a human-machine coupled lower limb exoskeleton rehabilitation robot digital twin model for a target rehabilitation patient; the model couples the rehabilitation patient's musculoskeletal system digital twin model with the lower limb exoskeleton rehabilitation robot digital twin model; The rehabilitation parameter simulation module is used to input the initial rehabilitation parameters into the digital twin model of the human-machine coupled lower limb exoskeleton rehabilitation robot for simulation, and to obtain the simulated physiological characteristic information of key muscle groups. The verification module is used to compare the simulated physiological feature information with the evaluation information from experts to verify the reliability and effectiveness of the human-machine coupled digital twin model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the personalized human-machine coupled digital twin model construction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the personalized human-computer coupled digital twin model construction method as described in any one of claims 1 to 6.

10. A computer program product, comprising software code, characterized in that, The program in the software code executes the steps of the personalized human-computer coupled digital twin model construction method as described in any one of claims 1 to 6.

Images