Dynamic stability bionic knee joint protector and control method

Abstract

The application discloses a knee joint dynamic stability bionic protector and a control method, and belongs to the field of sports bionics. The protector comprises a high polymer composite shell, a partitioned cushion layer, a bionic multi-axis movable joint assembly and a flexible bionic ligament strip. The protector is internally provided with an angle-pressure integrated sensing module, which can collect knee joint movement and stress data in real time, and through an intelligent micro drive unit, dynamic adjustment of the partitioned damping, rigidity and support angle can be realized. The protector has high adhesion and comfort, can automatically adjust the support mode according to the user's activity state, effectively improves the dynamic stability of the knee joint, prevents injury and assists in rehabilitation. The protector is suitable for knee joint function protection in multiple scenes such as sports protection, clinical rehabilitation and daily health management.

Description

Technical Field

[0001] This invention belongs to the field of bionics, and more specifically relates to a bionic brace and control method for dynamic stability of the knee joint. Background Technology

[0002] With the accelerating aging of society and the popularization of sports, the number of people suffering from knee injuries and functional impairments is showing a significant upward trend. As the core joint for weight-bearing and movement, the stability of the knee joint is crucial for daily life, rehabilitation, and athletic performance. Existing knee braces mostly provide physical support to the knee joint through structural rigidity or single-axis hinges, but they have significant shortcomings in dynamic movement adaptability, biomechanical zone response, and personalized adjustment. Specifically, traditional braces mostly use single rigid materials for their support components, making it difficult to balance comfort and flexibility, and lacking effective adaptation for users of different body types or disease states. While fixed-structure braces can enhance joint stability, they can easily restrict normal physiological joint movement, increase the feeling of a foreign body, and may lead to uneven stress and secondary injuries.

[0003] Furthermore, current market technologies for sensing and intelligent control are relatively lagging. Most knee braces cannot adjust in real time according to changes in the user's movements or stress conditions, making it difficult to achieve dynamic assistance and active protection for the complex biomechanical environment of the knee joint. To address these technical shortcomings, there is an urgent need for a new type of knee brace that integrates zoned support, dynamic bionic assistance, and personalized intelligent control to better meet users' scientific knee protection needs in diverse scenarios such as rehabilitation, prevention, and functional enhancement. Summary of the Invention

[0004] This invention aims to address the shortcomings of existing knee braces in terms of dynamic stability, zoned biomechanical response, personalized adaptation capabilities, and intelligent auxiliary adjustment. It overcomes the technical problems of traditional braces, such as poor comfort, inability to provide dynamic zoned support, and lack of adaptive adjustment to joint movement states. The invention achieves zoned biomimetic support, real-time health monitoring, and intelligent dynamic assistance for the knee joint under different exercise conditions, thereby improving the overall effect of braces on protecting the knee joint function and coordinating exercise physiology for individual users.

[0005] To achieve the above objectives, the present invention employs the following technical solution: It includes an integrated shell (1), which is made of a high-molecular composite material with memory rebound properties. The shell (1) is shaped to follow the contour of the human knee joint, and a partitioned buffer layer (2) surrounds the inner surface. The buffer layer (2) is bonded and fixed to the shell (1). Bionic multi-axis movable joint components (3) are provided on both sides of the shell (1). The bionic multi-axis movable joint components (3) are elastically hinged to the shell (1) and the limiting seats (5) at the upper and lower edges of both ends via flexible bionic ligament strips (4). The flexible bionic ligament strips (4) are made of highly elastic artificial fibers. The center of the bionic multi-axis movable joint components (3) is connected to an intelligent micro-drive unit (6). The drive unit (6) is combined with the bionic multi-axis movable joint assembly (3) through a magnetic quick-plug structure. A replaceable flexible fitting strip (7) is provided in the popliteal fossa inside the shell (1). The two ends of the fitting strip (7) are connected to the slots inside the shell (1). The tightness of the fitting strip (7) is adjusted by the adjustable buckle. An angle-pressure integrated sensing module (8) is embedded between the bionic multi-axis movable joint assembly (3) and the buffer pad layer (2). The sensing module (8) is connected to the side adjustment control (9) through a flexible wire. The adjustment control (9) is placed on the side of the shell (1) and communicates wirelessly or wiredly with the external terminal. The structural connection between the components realizes the synergistic function of mechanical partition support and dynamic bionic assistance. The whole constitutes an intelligent, dynamic and stable bionic knee brace.

[0006] In one embodiment, the flexible bionic ligament strip (4) is elastically connected between the bionic multi-axis movable joint assembly (3) and the limiting seats (5) set at both ends of the shell (1). The highly elastic artificial fiber material can simulate the dynamic tension of the anterior and posterior cruciate ligaments of the human body, generate physiological feedback during the bending and straightening of the joint, and enhance the dynamic stability of the joint.

[0007] In one embodiment, the intelligent micro-drive unit (6) is installed at the center of the bionic multi-axis moving joint assembly (3) through a magnetic quick-plug structure. It can automatically adjust the angle, stiffness or damping of the joint assembly (3) according to the control command, so as to realize intelligent assistance and protection for the dynamic movement of the joint.

[0008] In one embodiment, the replaceable flexible adhesive tape (7) is arranged in the popliteal fossa inside the shell (1) and connected to the slots inside the shell (1) through both ends. The adhesive tape (7) can achieve rapid fine adjustment of tightness through an adjustable buckle structure, thereby improving wearing comfort and personalized adaptability.

[0009] Furthermore, a dynamic stabilization bionic brace control method for the knee joint, the control method being applicable to the aforementioned bionic brace, includes: Initialize adaptive individual feature modeling, automatically collect static physiological parameters of the user's knee joint such as joint contour, tension distribution, and pressure point position through the angle-pressure integrated sensing module, and combine them with the adjustment feedback information of the shell and the fitting strap to establish a personalized physiological-structural coupling parameter model. Continuous dynamic data perception and self-learning processing: The sensing module collects physiological state data of the knee joint in real time under multiple angles and different loads. The lateral adjustment control uses an incremental self-learning algorithm to dynamically update the user model and identify abnormal joint force or movement trajectory characteristics. The partitioned collaborative prediction and adjustment command generation, based on the latest individual model and real-time sensor data, uses a partitioned biomimetic mechanical prediction algorithm to divide the knee joint into biomechanical partitions such as anterior, posterior, medial, and lateral, and predicts the movement trend, force transmission path and potential abnormalities of each partition. The drive unit achieves dynamic assistance through fuzzy adjustment. Based on the zonal prediction results, the drive unit uses a fuzzy control algorithm to autonomously determine the adjustment range and response speed, prioritizing the freedom of functional movement of the knee joint, while actively damping, stiffness enhancement, or relaxation adjustment for abnormal loads or sudden changes.

[0010] In one approach, the initial adaptive individual feature modeling includes: synchronously collecting static physiological parameters of multiple regions of the knee joint through an integrated angle-pressure sensing module, including joint contour, tension distribution and pressure point location, and combining the feedback information from the shell and the fitting strap to obtain the wear interaction feel, thereby establishing a personalized physiological-structural coupling parameter model that includes mechanical and structural features. The model is based on multimodal data fusion, extracts the main influencing factors, and uses a multivariate Gaussian mixture model to cluster and label the features of different regions, thereby improving the user's basic physiological structure and mechanical feature database. In addition, individualized parameters such as the natural range of motion of the knee joint, physiological curve parameters, and pressure threshold are calculated using static data and built-in engineering models. Furthermore, a Bayesian risk minimization strategy is used to estimate the lower confidence bound of abnormally collected data, thereby initializing the individual parameter database.

[0011] In one scheme, the continuous dynamic data perception and self-learning processing includes: multi-dimensional dynamic signals such as joint angle changes, force distribution and acceleration; the side-mounted adjustment control establishes statistical features of the above data through a sliding window mechanism, and uses an incremental self-learning algorithm to dynamically update the user's individual model. The self-learning algorithm integrates multi-timescale change trends and abrupt change feature signals, combined with time-series anomaly detection models such as dynamic Gaussian anomaly scores, to identify joint stress or motion trajectory anomalies, and corrects model parameters through Bayesian adaptive regression. All process data and model results are automatically stored in the individual parameter library.

[0012] In one scheme, the generation of the partitioned collaborative prediction and adjustment instructions includes: the control unit combines the latest individual parameter library and the data from the real-time sensing module to divide the knee joint space into multiple biomechanical partitions such as the anterior, posterior, medial, and lateral partitions, receives the dynamic motion and force characteristic data of each partition, and uses a partitioned biomimetic mechanical prediction algorithm to predict and analyze the motion trend, force transmission, and potential anomalies of each region. Based on the prediction results and combined with the dynamic feedback loop, the intelligent drive unit generates adjustment commands for parameters including damping, stiffness and support angle in real time according to the partition, and realizes synchronous adjustment of each partition, so as to achieve the functions of dynamic multi-point precise assistance and overall biomechanical coordination optimization.

[0013] In one scheme, the fuzzy adjustment of the driving unit realizes dynamic assistance, including: the driving unit uses a fuzzy control algorithm to autonomously determine the auxiliary adjustment range and response speed of each zone based on the zone prediction results, and integrates the expectation-perception difference, motion range, and force gradient fuzzy input, and maps different dynamic states into adjustment commands through membership functions and inference rule bases to realize adaptive adjustment of zone damping, stiffness and support angle. When abnormal joint load or sudden change in motion is detected, the drive unit actively enhances damping, stiffness or relaxation adjustment, and compresses the response window to ensure the freedom of knee joint functional movement and the naturalness of physiological feedback, meeting the user's daily and special movement needs.

[0014] Beneficial effects of this invention: This invention, through the innovative structure of using an integrated polymer composite shell and partitioned cushioning layer, combined with biomimetic multi-axis movable joint components and flexible biomimetic ligament strips, not only achieves multi-directional dynamic biomimetic support and partitioned cushioning protection for the knee joint, but also effectively improves the fit and comfort of wearing it.

[0015] With the help of an integrated angle-pressure sensing module, physiological data of multiple areas of the knee joint can be collected in real time. Through intelligent micro-drive units, fuzzy control algorithms and self-learning optimization models, it can realize intelligent dynamic adjustment of parameters such as zoned damping, stiffness and support angle according to the dynamic force and motion changes of individual users. Attached Figure Description

[0016] Figure 1 This is a structural diagram of the bionic protective gear of the present invention; Figure 2 This is a flowchart of the control method of the present invention.

[0017] In the diagram, 1-integrated shell, 2-buffer pad, 3-joint assembly, 4-ligament strip, 5-limiting seat, 6-drive unit, 7-fitting strip, 8-integrated sensing module, and 9-adjustment control unit. Detailed Implementation

[0018] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0019] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. To facilitate understanding, the invention will now be described more fully with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to make the disclosure of the invention more thorough and complete.

[0020] like Figure 1 As shown, the knee joint dynamic stabilization bionic brace includes the following structural components: The protective gear of this invention features an integrated outer shell 1 made of a high-polymer composite material with memory rebound properties. The outer shell 1 is curved to follow the contour of the human knee joint, and its inner surface is surrounded by a partitioned buffer layer 2. The buffer layer 2 is bonded and fixed to the outer shell 1, achieving multi-angle shock absorption protection for the knee joint. Bionic multi-axis movable joint components 3 are located on both sides of the outer shell 1. These components 3 are elastically hinged to the outer shell 1 and the limiting seats 5 located at its upper and lower edges using flexible bionic ligament strips 4. The ligament strips 4 are made of highly elastic synthetic fibers, simulating the dynamic tension of the anterior and posterior cruciate ligaments of the human body, ensuring physiological feedback during joint flexion and extension. The center of the bionic multi-axis movable joint component 3 is connected to an intelligent micro-drive unit 6. The drive unit 6 is combined with the bionic joint component 3 through a magnetic quick-plug structure, and can automatically adjust the angle, stiffness, or damping under control commands to enhance dynamic protection. A replaceable flexible fitting band 7 is provided on the inner side of the outer shell 1 at the popliteal fossa. The two ends of the fitting band 7 are connected to corresponding slots on the inner side of the outer shell 1. The tightness of the fit can be quickly and finely adjusted through adjustable buckles, which significantly improves the user's wearing comfort. In order to sense the knee status in real time, the present invention embeds an angle-pressure integrated sensing module 8 between the bionic multi-axis movable joint component 3 and the cushioning layer 2, and connects it to the side adjustment control unit 9 through a flexible wire. The control unit 9 is located on the side of the outer shell 1 and communicates with an external terminal wirelessly or by wire. The above-mentioned connection of each component is specially designed to realize the synergistic function of mechanical zone support and dynamic bionic assistance: the outer shell 1 is the main load-bearing structure, each flexible bionic ligament strip 4 is combined with the joint component 3 and the limiting seat 5 in an elastic hinge manner, the drive unit 6 is directly connected to the joint component 3 through a modular interface, the adjustable connection between the fitting band 7 and the outer shell 1 ensures personalized fit, and the sensing module 8 and the adjustment control unit 9 have smooth signal linkage. The whole constitutes an intelligent, dynamic and stable bionic knee brace device, which has both excellent ergonomic performance and innovative sports bionics.

[0021] like Figure 2 As shown, the specific steps of the bionic brace control method for dynamic stability of the knee joint are as follows: Step 1: Initialize adaptive individual feature modeling Upon initial wear, the angle-pressure integrated sensing module 8 automatically collects static physiological parameters of the user's knee joint, such as joint contour, tension distribution, and pressure point location. Combined with adjustment feedback information from the outer shell 1 and the fitting strap 7, a personalized physiological-structural coupling parameter model is established. This model employs a multimodal data fusion method to achieve basic modeling of the wearer's knee joint anatomy and movement habits, providing a dedicated parameter library for subsequent dynamic control strategies.

[0022] When a user first wears the knee joint dynamic stability bionic brace, the initial modeling mode is activated. The angle-pressure integrated sensing module 8 synchronously collects static information of different areas of the knee joint at multiple points, including contour curve data, pressure distribution on the soft tissue surface, and natural joint tension.

[0023] These raw signals are processed through high-precision analog-to-digital conversion and filtering to obtain a matrix in multidimensional space. i represents the partition, j represents different types of sensors, and k is the sampling point. Simultaneously, the displacement and mechanical feedback values ​​of the outer shell 1 and the fitting strap 7 during the adjustment process are read to quantify the individual wearable interactive feel, denoted as the feedback vector. .

[0024] These multimodal data are integrated through a feature alignment network, resulting in a matrix. With vector After normalization, denoising, and principal component analysis (PCA), the main influencing factors were effectively extracted. Let the normalization function be... Then the normalized data Based on this, a multivariate Gaussian mixture model (GMM) is used to cluster the various classification features, with the probability distribution function being... in For feature vectors, As weight, and These represent the mean and covariance, respectively. The clustering results are used to label the basic physiological structures and mechanical features of each region.

[0025] After feature aggregation is completed, the feature database is stored in the individual parameter database. This serves as the basis for personalized configuration of subsequent dynamic control.

[0026] At the same time, by combining the collected static data with the built-in ergonomic model, the natural range of motion of the joint—such as the maximum angle of knee flexion and extension—is calculated. Physiological curve parameters and pressure threshold Personalized parameters are used, and the confidence lower bound of outlier data is estimated using the Bayesian risk minimization method to avoid extreme misjudgments. This series of continuous data processing not only depicts the user's knee joint's basic shape and biomechanical zones, but also makes personalized modifications and presets for the ease of fit adjustment. This ensures that the subsequent dynamic auxiliary adjustment of the protective gear is based on a real and unique physiological foundation, providing a solid guarantee for the device's individual friendliness and feedforward control quality.

[0027] Step 2: Continuous dynamic data perception and self-learning processing During use, the sensor module 8 collects real-time physiological data of the knee joint under multiple angles and different loads. The lateral adjustment control 9 uses an incremental self-learning algorithm to dynamically update the user model and identify abnormal joint force or movement trajectory characteristics. This self-learning algorithm integrates multi-timescale change trends and abrupt change characteristic signals to achieve real-time adaptive adjustment preparation for joint health status and movement requirements.

[0028] During daily activities and exercise, the sensor module 8 of the knee joint dynamic stabilization bionic brace records the real-time angle changes of the joint using a high-frequency sampling stream. Force distribution acceleration per unit time These multidimensional dynamic signals are organized into temporal tensors. The data is continuously transmitted to the side-mounted regulating control 9 at millisecond levels. These physical parameters are analyzed online throughout the entire timeframe, and data over a continuous duration of T is processed using a sliding window mechanism. Establish a statistical description and dynamically extract time-series features such as signal mean, extreme values, rate of change, and their first and second derivatives to construct a joint feature vector. The entire control system employs an incremental self-learning algorithm, continuously updating with newly acquired data. The individual motion-physiological response under external disturbances such as temperature, humidity, and activity intensity is corrected. By using time-series anomaly detection models such as the dynamic Gaussian anomaly fraction method, abrupt changes and slow drifts of physical quantities in each time slice interval can be identified.

[0029] This self-learning mechanism integrates a multi-timescale adaptive incremental weighted filtering strategy. The core idea is to assign different weights to feature changes within different time windows to capture both short-term rapid changes and long-term trends. Specifically, long-term trends can be captured using the exponential moving average formula. get, This is the time decay factor. For abrupt signals, the cumulative error criterion is used; if... This triggers the adaptive dynamic adjustment process. Combined with the accumulated residual... The rate of change can be used to identify sudden abnormal joint loads and non-physiological movement trajectories.

[0030] Meanwhile, the continuously updated individual user model, based on Bayesian adaptive regression, constantly corrects the prior distribution of joint structure and typical dynamic parameters, and iteratively updates the posterior probability. The forgetting factor is used to further reduce the excessive influence of long-term history on the current state, enabling the model to adapt flexibly to new scenarios. In this way, the fusion of multi-timescale and abrupt change characteristic signals allows the control to both keenly capture danger signals and avoid being sluggish or erroneous in response to natural changes. All updates are automatically incorporated into the user's individual parameter library. This provides a data foundation for the next step of intelligent drive adjustment and feedback control. As the amount of data and usage time increase, the model's characterization of users' exercise habits, joint health status, and abnormal risks becomes increasingly accurate, achieving continuous, dynamic, and personalized closed-loop capabilities.

[0031] Step 3: Generation of Regional Collaborative Prediction and Adjustment Instructions Based on the latest individual models and real-time sensor data, the control unit 9 uses a partitioned biomimetic mechanical prediction algorithm to divide the knee joint into anterior, posterior, medial, and lateral biomechanical zones, predicting the movement trend, force transmission path, and potential anomalies of each zone. It innovatively integrates a dynamic feedback loop to generate real-time targeted adjustment commands for the intelligent drive unit 6, such as adjusting damping, stiffness, and support angle. These commands are output synchronously by zone, achieving dynamic multi-point precise assistance.

[0032] During the operation of the dynamic bionic knee brace, real-time collected individual motion-physiological state data and user parameter databases continuously iterated by the self-learning model are used. Together, this provides a solid foundation for zonal biomechanical prediction. The control unit 9 spatially divides the knee joint region into several key biomechanical zones, including the anterior, posterior, medial, and lateral zones. Each zone receives feature streams collected by its sensors via an independent data channel, forming a state vector. At every moment The dynamic prediction module uses an adaptive partition correlation mechanics analysis function based on these partition features and user-defined model parameters. Multi-zone coupling calculations are performed to effectively reflect the mechanical transmission path in the real motion mode.

[0033] In the specific prediction process, a zoned motion trend function is introduced. The system depicts the changing trajectories of forces, angles, and accelerations within the region; it performs cumulative analysis over a time period, and uses a partitioned coupled mechanical model to calculate the force-torque transmission relationship in the associated region in real time. This relationship can be expressed as follows: in Indicates partition For this area The influence transfer function is adaptively estimated from the individual feature database. This represents the specific disturbance in this region at the current moment. Through the above predictions, we can not only detect impending load anomalies but also understand the potential hidden risks arising from the coordinated stress chain across multiple regions.

[0034] The adjustment command generation module is deeply integrated with the dynamic feedback loop. For each zone, the intelligent drive unit 6 calculates the target damping on demand. Stiffness and support angle The adjustment amount is equal to the adjustment command. The specific adjustment command is based on the partition-state mapping and is generated using the following command: in Describe the interval weight transformation. To coordinate the sensitivity coefficient across the entire region, the adjustment commands are not only used for specific risk responses within the region but also guide the overall biomechanical optimization. The drive unit then receives corresponding digital commands from each zone and uses micro-actuators to instantly perform physical controls such as dynamic damping adjustment, stiffness enhancement or release, and support angle correction. The overall output achieves a motion assistance effect of "dynamic multi-point - precise zoning - coordinated update".

[0035] Meanwhile, the zonal prediction-adjustment process continuously uses a dynamic feedback mechanism to automatically compare the implemented adjustment schemes with subsequent actual sensor data. If the current adjustment fails to effectively eliminate the detected abnormal signals, it will be corrected online. and These are core model parameters. In this way, the protective gear can not only provide zone-specific support for local anomalies, but also respond to sudden events or strenuous movements with coordinated optimization across the entire area, achieving a high degree of consistency between intelligent assistance and the user's natural movements.

[0036] Step 4: Fuzzy adjustment of the driving unit to achieve dynamic assistance Based on the zonal prediction results, the drive unit 6 autonomously determines the adjustment range and response speed using a fuzzy control algorithm, prioritizing the freedom of functional knee joint movement while actively damping, stiffness enhancement, or relaxation adjustments for abnormal loads or sudden changes. This fuzzy adjustment method, combined with the expectation-perception differential mechanism, fully caters to the daily and special movement needs of different users, ensuring that physiological feedback closely approximates the natural movement state.

[0037] After the knee joint bionic brace performs real-time zone prediction, the intelligent drive unit 6 immediately enters the autonomous adjustment phase, relying entirely on fuzzy control algorithms to make self-perceptive decisions on auxiliary parameters. The motion intention, actual state, and functional requirements of each zone are abstracted into fuzzy input quantities, such as the expected joint range of motion. Current perceived motion amplitude With the detected force gradient Etc. Based on the "expectation-perception" difference, i.e. By combining fuzzified membership functions such as force and angle, the dynamic state is mapped to a multi-level fuzzy set such as "insufficient, moderate, excessive" and "low risk, medium risk, high risk". (Fuzzy inference rule base) It defines output strategies for dozens of typical cases, such as when "Moderate" and In the "low-risk" state, both the adjustment range and response speed are assigned a lower level to ensure functional freedom of movement; when “Insufficient” and When the risk level is "high", the damping should be increased accordingly. Or increase stiffness Increase the response rate to automatically intervene and provide support.

[0038] The implementation process of the fuzzy control algorithm consists of three progressive steps: fuzzification, inference, and defuzzification. The input quantity is determined through a triangular membership function. The "relaxed / normal / tense" states are categorized into different language variables and represented by functions. Describe them separately. Assume the ideal membership degree for free joint movement is... The current degree of freedom index is To measure the naturalness of movement. Drive adjustment commands. Mapped by fuzzy rules The decision and mapping process can be represented by a generalized Takagi-Sugeno type fuzzy model: in The activation weight for the l-th rule, It is a set of fuzzy parameters. For abnormal and sudden changes detected by the warning, such as load jumps or sudden motion instability, the fuzzy output immediately increases the damping or stiffness adjustment range, while the response window is significantly compressed to actively prevent potential damage in an overdamped or enhanced stiffness state.

[0039] Driven by this fuzzy adjustment algorithm, the intelligent drive unit perfectly matches the real-time movement needs of each user, achieving step-by-step adjustment of the amplitude unit at micro-resolution. Mechanical feedback and neuro-skeletal action are virtually seamlessly coupled, ensuring maximum freedom of individual physiological movement and enhancing proactive protection under abnormal risk conditions. The adjustment process continuously compares real-time sensor feedback with zoned motion data. If the degree of freedom of movement is restricted but there is no abnormal load, it will be judged as "over-adjustment," and stiffness or damping will be automatically relaxed to make the assistive effect closer to the natural movement performance of the joint.

[0040] This application proposes a one-piece shell made of polymer composite material, incorporating a zoned cushioning layer and a zoned biomechanical biomimetic structure. Unlike existing technologies that simply divide the main body of the protective gear into thigh, calf, and knee sections, and utilize traditional support components and passive adjustment mechanisms, this application actively simulates the stress distribution of the human knee joint and its ligaments through the overall biomimetic contour of the shell, the zoned mechanical layout, and the flexible biomimetic ligaments made of highly elastic synthetic fibers. This not only enhances the biomimetic nature and comfort of wearing the garment but also allows for a high degree of adaptability to different individual user structures. In particular, the proposal and implementation of zoned biomechanical biomimetic design overcomes the limitations of existing technologies that rely solely on simple shape fitting and passive support from traditional mechanical structures, endowing the protective gear with active dynamic support and auxiliary capabilities.

[0041] Secondly, at the functional implementation level, this application integrates a biomimetic multi-axis movable joint component with an intelligent micro-drive unit, and is equipped with a magnetic structure for quick insertion, removal, and replacement. This modular and intelligent drive structure can achieve adaptive stiffness and damping adjustment and assistance for the dynamic movement of the knee joint based on the joint motion and force data collected by sensors. Compared with the existing technology that only collects angle data and outputs data display through angle sensors without active adjustment, this application can perform closed-loop control of the protective gear function according to the actual operating state of the joint, significantly improving the intelligent response capability in the knee joint protection and assisted rehabilitation process, and greatly improving the safety, initiative, and personalized service level of the equipment.

[0042] Furthermore, regarding innovation in sensing and control methods, this application proposes an integrated angle-pressure sensing, dynamic self-learning modeling, and zoned collaborative prediction-adjustment integrated control algorithm. Compared to existing technologies that rely on single-point data acquisition and traditional logic processing for data monitoring, this application utilizes multiple intelligent algorithms such as multimodal data fusion, Gaussian mixture clustering, and Bayesian anomaly correction to achieve intelligent closed-loop and dynamic adaptive adjustment between the protective gear and the user, and between different structural zones. In particular, the application of a series of algorithms, including adaptive user modeling, zoned multi-point biomechanical prediction, and fuzzy control, enables the protective gear not only to passively "adapt" to the user's basic body shape but also to "predict" risks in real time and actively adjust support parameters—a capability unmatched by current knee brace technologies.

[0043] In summary, this application represents a breakthrough in the functionality and implementation methods of existing technologies at multiple levels, including structural innovation, intelligent adjustment mechanisms, and control algorithm systems. Its inventiveness lies in the fact that, through the synergy of biomechanical biomimetic design, actively driven structural modules, and an intelligent multimodal self-learning sensing and control system, it achieves a fundamental leap in knee braces from traditional static passive rehabilitation aids to intelligent dynamic, personalized prediction, and active assistance. These innovations not only possess significant novelty and progress but also provide new technical solutions and development paths for the intelligentization of future high-performance rehabilitation aids and sports protective equipment.

[0044] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0045] It should be understood that the above detailed description of the technical solutions of the present invention with reference to preferred embodiments is illustrative and not restrictive. Those skilled in the art can modify the technical solutions described in the embodiments or make equivalent substitutions for some of the technical features based on reading this specification; however, these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A bionic knee joint dynamic stabilization brace, characterized in that: The bionic protective gear includes: an integrated shell (1), which is made of a high-molecular composite material with memory rebound properties. The shell (1) is shaped to follow the contour of the human knee joint. The inner surface is surrounded by a partitioned buffer layer (2), which is bonded to the shell (1). Bionic multi-axis movable joint components (3) are provided on both sides of the shell (1). The bionic multi-axis movable joint components (3) are elastically hinged to the shell (1) and the limiting seats (5) at the upper and lower edges of the shell (1) through flexible bionic ligament strips (4). The flexible bionic ligament strips (4) are made of highly elastic artificial fibers. The center of the bionic multi-axis movable joint components (3) is connected to an intelligent micro-drive unit (6). By combining the magnetic quick-plug structure with the bionic multi-axis movable joint assembly (3), a replaceable flexible fitting strip (7) is provided in the popliteal fossa of the inner side of the shell (1). The two ends of the fitting strip (7) are connected to the slots inside the shell (1). The tightness of the fitting strip (7) is adjusted by the adjustable buckle. An angle-pressure integrated sensing module (8) is embedded between the bionic multi-axis movable joint assembly (3) and the buffer pad layer (2). The sensing module (8) is connected to the side adjustment control (9) through a flexible wire. The adjustment control (9) is placed on the side of the shell (1) and communicates wirelessly or wiredly with the external terminal. The structural connection between the components realizes the synergistic function of mechanical partition support and dynamic bionic assistance. The whole constitutes an intelligent, dynamic and stable bionic knee brace.

2. The knee joint dynamic stabilization bionic brace according to claim 1, characterized in that: The flexible bionic ligament strip (4) is elastically connected between the limiting seats (5) set at both ends of the bionic multi-axis active joint component (3) and the shell (1). This highly elastic artificial fiber material can simulate the dynamic tension of the anterior and posterior cruciate ligaments of the human body, and generate physiological feedback during the bending and straightening of the joint, thereby enhancing the dynamic stability of the joint.

3. The knee joint dynamic stabilization bionic brace according to claim 1, characterized in that: The intelligent micro-drive unit (6) is installed in the center of the bionic multi-axis moving joint assembly (3) through a magnetic quick-plug structure. It can automatically adjust the angle, stiffness or damping of the joint assembly (3) according to the control command, so as to realize intelligent assistance and protection for the dynamic movement of the joint.

4. The knee joint dynamic stabilization bionic brace according to claim 1, characterized in that: The replaceable flexible adhesive tape (7) is arranged in the popliteal fossa inside the shell (1) and connected to the slots inside the shell (1) through both ends. The adhesive tape (7) can achieve rapid fine adjustment of tightness through an adjustable buckle structure, thereby improving wearing comfort and personalized adaptability.

5. A method for controlling the dynamic stability of a knee joint using a bionic brace, wherein the control method is applicable to the bionic brace as described in any one of claims 1-4, characterized in that: The control method includes: Initialize adaptive individual feature modeling, automatically collect static physiological parameters of the user's knee joint such as joint contour, tension distribution, and pressure point position through the angle-pressure integrated sensing module, and combine them with the adjustment feedback information of the shell and the fitting strap to establish a personalized physiological-structural coupling parameter model. Continuous dynamic data perception and self-learning processing: The sensing module collects physiological state data of the knee joint in real time under multiple angles and different loads. The lateral adjustment control uses an incremental self-learning algorithm to dynamically update the user model and identify abnormal joint force or movement trajectory characteristics. The partitioned collaborative prediction and adjustment command generation, based on the latest individual model and real-time sensor data, uses a partitioned biomimetic mechanical prediction algorithm to divide the knee joint into anterior, posterior, medial, and lateral biomechanical partitions, and predicts the motion trend, force transmission path and potential abnormalities of each partition. The drive unit achieves dynamic assistance through fuzzy adjustment. Based on the zonal prediction results, the drive unit uses a fuzzy control algorithm to autonomously determine the adjustment range and response speed, prioritizing the freedom of functional movement of the knee joint, while actively damping, stiffness enhancement, or relaxation adjustment for abnormal loads or sudden changes.

6. The method for controlling the dynamic stability of the knee joint using a bionic brace according to claim 5, characterized in that: The initial adaptive individual feature modeling includes: synchronously collecting static physiological parameters of multiple regions of the knee joint through the angle-pressure integrated sensing module, including joint contour, tension distribution and pressure point position, and obtaining the wearing interaction feel by combining the adjustment feedback information of the shell and the fitting strap, thereby establishing a personalized physiological-structural coupling parameter model that includes mechanical and structural features; The model is based on multimodal data fusion, extracts the main influencing factors, and uses a multivariate Gaussian mixture model to cluster and label the features of different regions, thereby improving the user's basic physiological structure and mechanical feature database. In addition, by using static data and built-in engineering models, the natural range of motion of the knee joint, physiological curve parameters and individualized parameters of pressure threshold are calculated. Furthermore, a Bayesian risk minimization strategy is used to estimate the confidence lower bound of abnormally collected data, thereby initializing the individual parameter database.

7. The method for controlling dynamic stability of the knee joint using a bionic brace according to claim 5, characterized in that: The continuous dynamic data perception and self-learning processing includes: joint angle changes, force distribution and multi-dimensional dynamic signals of acceleration. The side-mounted adjustment control establishes statistical features of the above data through a sliding window mechanism and uses an incremental self-learning algorithm to dynamically update the user's individual model. The self-learning algorithm integrates multi-timescale change trends and abrupt change feature signals, combines them with a dynamic Gaussian anomaly fractional time-series anomaly detection model, identifies joint stress or motion trajectory anomalies, and corrects model parameters through Bayesian adaptive regression. All process data and model results are automatically stored in an individual parameter library.

8. The method for controlling the dynamic stability of the knee joint using a bionic brace according to claim 5, characterized in that: The generation of the partitioned collaborative prediction and adjustment instructions includes: the control unit combines the latest individual parameter library and the data from the real-time sensing module to divide the knee joint space into multiple biomechanical partitions, namely the anterior, posterior, medial, and lateral partitions, and receives the dynamic motion and force characteristic data of each partition respectively. The partitioned biomimetic mechanical prediction algorithm is used to predict and analyze the motion trend, force transmission and potential anomalies of each region. Based on the prediction results and combined with the dynamic feedback loop, the intelligent drive unit generates adjustment commands for parameters including damping, stiffness and support angle in real time according to the partition, and realizes synchronous adjustment of each partition, so as to achieve the functions of dynamic multi-point precise assistance and overall biomechanical coordination optimization.

9. The method for controlling dynamic stability of the knee joint using a bionic brace according to claim 5, characterized in that: The aforementioned fuzzy adjustment of the driving unit realizes dynamic assistance, including: the driving unit uses a fuzzy control algorithm to autonomously determine the auxiliary adjustment amplitude and response speed of each partition based on the partition prediction results, and integrates the expectation-perception difference, motion amplitude, and force gradient fuzzy input, and maps different dynamic states into adjustment commands through membership functions and inference rule bases to realize adaptive adjustment of partition damping, stiffness and support angle. When abnormal joint load or sudden change in motion is detected, the drive unit actively enhances damping, stiffness or relaxation adjustment, and compresses the response window to ensure the freedom of knee joint functional movement and the naturalness of physiological feedback, meeting the user's daily and special movement needs.

Images