Intelligent lower limb walking aid and control method

Abstract

The invention relates to the technical field of medical equipment, in particular to an intelligent lower limb walking aid and a control method. The intelligent lower limb walking aid comprises a lower limb power assisting device which is worn on the lower limbs of a patient and provides joint driving force to assist walking movement; the treadmill platform is used for providing a controllable motion surface and supporting the patient to carry out gait training; the body trunk auxiliary device is installed on the treadmill platform and used for supporting the trunk part of the patient, relieving the burden of the upper limbs and keeping the body balanced; the lower limb power assisting device, the treadmill platform and the body trunk assisting device are all electrically connected with the central control unit, and the central control unit is used for collecting gait parameters of a patient, the speed of the treadmill and the driving state of the lower limb power assisting device in real time and controlling the body trunk assisting device based on a multi-modal sensor fusion algorithm. The speed of the treadmill and the assisting power output of the lower limb assisting device are dynamically adjusted, and self-adaptive synchronization of the speed of the treadmill and the gait rhythm of the patient is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical equipment, and in particular to an intelligent lower limb walking aid and a control method. BACKGROUND

[0002] Lower limb rehabilitation equipment refers to a type of medical instrument or comprehensive training system specially designed to assist patients in recovering or significantly improving their lower limb motor function, effectively enhancing muscle strength, significantly improving joint range of motion, comprehensively improving balance ability, and walking skills. This type of equipment plays a crucial role in the field of rehabilitation medicine, aiming to help patients gradually recover normal lower limb function and improve their quality of life.

[0003] The common lower limb walking auxiliary equipment on the market usually needs to be worn on the patient's leg to provide necessary support when performing walking training. However, due to the weakness of the patient's lower limbs, they often need to exercise within a certain space range and usually need the assistance of family members or professional caregivers to ensure safety. This exercise method is relatively cumbersome and inconvenient in actual operation, limiting the patient's self-exercise ability and rehabilitation effect. SUMMARY

[0004] To solve the above technical problems existing in the prior art, the present application provides an intelligent lower limb walking aid and its advanced control method. The intelligent walking aid integrates advanced sensors, intelligent control systems, and mechanical support structures, which can monitor the patient's walking state in real time and provide personalized assistance according to the patient's actual needs. This not only greatly improves the safety and comfort of the patient during use, but also significantly improves the efficiency and effectiveness of rehabilitation training, allowing patients to perform lower limb function recovery training in a more autonomous and convenient environment.

[0005] To achieve the above purpose, the embodiments of the present application provide the following technical solutions: In a first aspect, in one embodiment provided by the present application, an intelligent lower limb walking aid is provided, which comprises: a lower limb assistance device worn on the lower limbs of a patient to provide joint driving force to assist walking movement; a treadmill platform for providing a controllable movement surface to support the patient in gait training; a body trunk assistance device installed on the treadmill platform for supporting the patient's trunk portion, reducing the burden on the upper limbs and maintaining body balance; The central control unit, lower limb assist device, treadmill platform, and body trunk assist device are all electrically connected to the central control unit. The central control unit is used to collect the patient's gait parameters, treadmill speed, and driving status of the lower limb assist device in real time, and dynamically adjusts the treadmill speed and the assist output of the lower limb assist device based on a multimodal sensor fusion algorithm to achieve adaptive synchronization between the treadmill speed and the patient's gait rhythm.

[0006] As a further embodiment of the present invention, the lower limb assistive device includes a mounting belt, a hip joint drive block, and a knee joint drive block. The mounting belt is used to be fitted around the patient's waist. The hip joint drive block is mounted on the mounting belt and is used to drive the patient's hip joint. The knee joint drive block is mounted on the hip joint drive block and is used to drive the patient's knee joint. A first restraint strap is also provided on the hip joint drive block for fitting the hip joint drive block onto the patient's thigh. A second restraint strap is provided on the knee joint drive block for fitting the knee joint drive block onto the patient's lower leg.

[0007] As a further embodiment of the present invention, both the mounting belt and the hip joint drive block are equipped with electromyography sensors, force sensors, angle sensors, and inertial measurement units.

[0008] As a further embodiment of the present invention, the body trunk assist device includes a support, a telescopic seat, and a restraint assembly. The support is fixedly installed on the treadmill platform, the telescopic seat is installed on the support, and the restraint assembly is disposed at the output end of the telescopic seat. The restraint assembly is used to provide auxiliary support for the patient's upper limbs.

[0009] As a further embodiment of the present invention, the restraint assembly may include a shoulder support pad and at least one fixing strap.

[0010] As a further aspect of the present invention, the central control unit includes: a gait phase recognition unit, a cooperative control algorithm unit, and a personalized training mode unit; The gait phase recognition unit identifies the support phase and swing phase in the gait cycle in real time based on data from inertial sensors and joint angle sensors. The collaborative control algorithm unit is used to optimize the collaborative output of treadmill speed and lower limb assist torque by using reinforcement learning methods, so that the gait training process is natural, smooth and energy-saving. The personalized training mode unit is used to support automatic switching between passive training, active assistance, and resistance training modes according to the patient's rehabilitation stage.

[0011] As a further aspect of the present invention, the active assistance mode may include several rehabilitation stages, and the treadmill platform corresponding to each rehabilitation stage has a corresponding assistance coefficient K.

[0012] As a further embodiment of the present invention, the central control unit (40) is also used to monitor the patient's postural stability, abnormal gait or risk of falling in real time, and to trigger an emergency braking or support intervention mechanism when an abnormality is detected.

[0013] As a further aspect of the present invention, a cloud data management platform is also included. The central control unit is communicatively connected to the cloud data management platform and is used to store the exercise parameters, physiological indicators and rehabilitation progress of each training session so that doctors can access the data.

[0014] Secondly, in another embodiment of the present invention, a control method for an intelligent lower limb walking aid is provided, the control method comprising: real-time acquisition of patient gait parameters, lower limb joint motion data and treadmill operating status; Gait phase recognition is performed using patient gait parameters and lower limb joint motion data to obtain gait phase recognition results, and the gait rhythm of the next cycle is predicted based on the gait phase recognition results; Based on the prediction results, the treadmill speed and the driving torque of the lower limb assist device are dynamically adjusted to match the patient's natural gait kinematically.

[0015] The technical solution provided by this invention has the following beneficial effects: This invention enables patients to independently perform lower limb rehabilitation exercises. By coordinating a treadmill platform, a trunk assist device, and a central control unit, it ensures exercise safety while simultaneously monitoring the patient's walking status in real time and providing personalized support based on their needs. This significantly improves patient safety and comfort, as well as the efficiency and effectiveness of rehabilitation training, allowing patients to perform lower limb function recovery training in a more autonomous and convenient environment. This invention also eliminates the discontinuous gait sensation of missing a step, sudden stop, and restart, making the training process continuous and natural, greatly reducing psychological fear, and transforming passive, forced walking into active rhythmic following. The patient's muscle activation pattern is closer to normal walking, resulting in better neuroplasticity remodeling. The upper limbs do not require continuous exertion, the trunk remains naturally upright, and the foot pressure distribution and stride length / width data are more accurate. By eliminating reliance on handrails, the patient's center of gravity control is achieved by the lower limbs and core, returning the training goal to walking itself, resulting in a rehabilitation efficiency of over 20%.

[0016] These or other aspects of the invention will become more apparent from the following description of embodiments. It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a structural diagram of an intelligent lower limb walking aid according to an embodiment of the present invention.

[0019] Figure 2 This is a flowchart illustrating the control method of an intelligent lower limb walking aid according to an embodiment of the present invention.

[0020] Figure 3 This is a flowchart illustrating step S20 of the control method for an intelligent lower limb walking aid according to an embodiment of the present invention.

[0021] In the diagram: 10-Lower limb assist device, 20-Treadmill platform, 30-Body trunk assist device, 40-Central control unit, 50-Cloud data management platform, 101-Waist belt installation, 102-Hip joint drive block, 103-Knee joint drive block, 301-Support, 302-Telescopic seat, 303-Restraint assembly, 304-Automatic extension bar. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0024] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0025] Specifically, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0026] Please see Figure 1 ,Figure 1 This is a structural diagram of an intelligent lower limb walking aid provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the intelligent lower limb walking aid includes a lower limb assist device 10, a treadmill platform 20, a body trunk assist device 30, and a central control unit 40, which together constitute a complete rehabilitation training system.

[0027] Specifically, the lower limb assist device 10 is a device specifically designed to be worn on the lower limbs of patients. Its main function is to provide joint driving force, thereby effectively assisting patients in walking and helping them gradually restore normal gait during rehabilitation. The lower limb assist device 10 is also used to collect patient joint data and electromechanical data and transmit the collected data to the central control unit 40.

[0028] The treadmill platform 20 is a controllable motion surface that can adjust the speed and incline according to training needs, providing patients with a stable and safe gait training environment, enabling them to carry out effective rehabilitation training under simulated real walking conditions.

[0029] The torso support device 30 is an auxiliary device installed on the treadmill platform 20, whose main function is to support the patient's torso. This support significantly reduces the burden on the patient's upper limbs and effectively maintains balance, preventing falls and other accidents during training. This allows the patient to focus more on lower limb gait training, improving training effectiveness.

[0030] In addition, the lower limb assist device 10, the treadmill platform 20, and the torso assist device 30 are all electrically connected to the central control unit 40.

[0031] The central control unit 40 is the core component of the system, possessing real-time data acquisition and processing capabilities. Specifically, the central control unit 40 can collect key data such as the patient's gait parameters, the treadmill's operating speed, and the driving status of the lower limb assist device in real time. Based on this data, the central control unit 40 utilizes a multimodal sensor fusion algorithm to dynamically adjust the treadmill speed and the assist output of the lower limb assist device. Through this dynamic adjustment, the system can achieve adaptive synchronization between the treadmill speed and the patient's gait rhythm, ensuring that the patient maintains optimal gait matching during training. This not only significantly improves the effectiveness of rehabilitation training but also effectively avoids secondary injuries caused by gait incoordination, providing strong technical support for the patient's comprehensive rehabilitation. Gait parameters can be collected based on the lower limb assist device 10.

[0032] In traditional treadmill training, the exoskeleton and the running platform operate independently—the running platform has a fixed speed, and the exoskeleton is driven according to a preset trajectory. If the patient's stride frequency cannot keep up, they will be pulled down or misstep. This invention eliminates the discontinuous gait sensation of misstepping-sudden stop-restart, making the training process continuous and natural, significantly reducing psychological fear, and transforming passive forced walking into active rhythmic following. The patient's muscle activation pattern is closer to normal walking, resulting in better neuroplasticity remodeling. The upper limbs do not need to exert continuous force, the trunk remains naturally upright, and the data on plantar pressure distribution and stride length / width are more accurate. After eliminating reliance on handrails, the patient's center of gravity control is completed by the lower limbs and core, and the training goal returns to walking itself, with a rehabilitation efficiency of over 20%.

[0033] In this embodiment of the invention, the lower limb assistive device 10 mainly consists of the following key components: a mounting belt 101, a hip joint drive block 102, and a knee joint drive block 103. Specifically, the mounting belt 101 is designed to be securely fitted around the patient's waist, ensuring that the device does not shift during use. The hip joint drive block 102 is mounted on the mounting belt 101, and its main function is to drive the patient's hip joint, helping the patient achieve normal hip joint movement. Simultaneously, the knee joint drive block 103 is mounted on the hip joint drive block 102, and its main function is to drive the patient's knee joint, assisting the patient in completing knee flexion and extension movements. In use, the hip joint drive block 102 assists the patient in bending their thigh, while the knee joint drive block 103, moving with the hip joint drive block 102, assists the patient in bending their lower leg, thus assisting the patient in lower limb exercises.

[0034] To further enhance the stability and fit of the device, a first restraint strap is specially provided on the hip joint drive block 102. The main purpose of this restraint strap is to securely fit the hip joint drive block 102 onto the patient's thigh, ensuring a close fit between the drive block and the patient's limb. Similarly, a second restraint strap is also provided on the knee joint drive block 103. The function of this restraint strap is to securely fit the knee joint drive block 103 onto the patient's lower leg, ensuring that the drive block does not slip or shift during movement.

[0035] It's worth noting that these two restraint straps are made of a composite material combining memory foam and breathable mesh. Memory foam has excellent elasticity and memory function, automatically adjusting to the patient's body shape and movement to provide an optimal fit; while the breathable mesh ensures breathability during prolonged use, preventing skin discomfort or allergies caused by prolonged pressure. This composite material choice not only improves the comfort and safety of the device but also greatly enhances the patient's user experience.

[0036] In this embodiment of the invention, flexible pressure sensor arrays are embedded on the inner sides of both the first and second restraint straps to monitor the limb contact pressure distribution in real time and prevent local compression from causing poor blood circulation. The flexible pressure sensor arrays are electrically connected to the central control unit 40, and the central control unit 40 issues an alarm signal when the pressure exceeds a certain threshold.

[0037] In this embodiment of the invention, both the knee joint drive block 103 and the hip joint drive block 102 are equipped with various high-precision sensors, such as electromyography sensors, force sensors, angle sensors, and inertial measurement units. These sensors synchronously and efficiently collect three-dimensional motion data of the hip and knee joints at a sampling frequency of 200 times per second. To ensure the accuracy and reliability of the data, the raw data collected by the aforementioned sensors is transmitted to the central control unit 40, which uses an advanced Kalman filter algorithm to eliminate noise interference. The denoised motion trajectory data is then compared and analyzed in real time with a pre-set standard gait model, thereby achieving accurate monitoring and evaluation of the user's gait.

[0038] In this embodiment of the invention, the end of the restraint strap adopts a magnetic quick-release structure, allowing the patient to complete the wearing and adjustment with one hand without the need for assistance from others.

[0039] In this embodiment of the invention, the drive joints in the lower limb assistive device 10 can adopt a modular design. Both the hip joint drive block 102 and the knee joint drive block 103 include a drive unit and an execution component mounted on the drive unit. The execution component is secured to the patient's lower limb with a restraint strap; the execution component can be rod-shaped. The execution component is mounted on the output end of the drive unit via a quick-release interface. The quick-release interface can be a magnetic quick-release connector supporting hot-swapping. The drive unit controls the swing amplitude of the execution component to meet the precise control requirements of joint mobility at different rehabilitation stages. Each transmission component is equipped with strain gauges to monitor load changes in real time; an overload protection mechanism is triggered when the torque exceeds a safety threshold. Electromyography (EMG) sensors, force sensors, angle sensors, and inertial measurement units can be integrated onto the execution component.

[0040] In this embodiment of the invention, the body trunk assistive device 30 mainly consists of three core parts: a support 301, a telescopic seat 302, and a restraint assembly 303. The support 301 is designed to be fixedly installed on the outer shell of the treadmill platform 20 to ensure the stability and safety of the entire device. The telescopic seat 302 is cleverly installed on the support 301, and its adjustable feature allows it to flexibly adapt to different height requirements. The restraint assembly 303 is located at the extended end of the telescopic seat 302 and is specifically designed to provide effective auxiliary support for the patient's upper limbs.

[0041] Specifically, the restraint component 303 provides stable support for the patient's upper limbs, significantly reducing fatigue experienced during rehabilitation training. This design not only enhances patient comfort but also contributes to improved rehabilitation effectiveness. Furthermore, the telescopic seat 302 takes into account individual patient differences. Its adjustable telescopic function easily meets the varying height requirements of different patients, allowing each patient to perform rehabilitation training at the most suitable height, further enhancing the device's applicability and user-friendly design.

[0042] In this embodiment of the invention, the restraint assembly 303 may include a shoulder support pad and at least one fixing strap. Together, they can effectively provide auxiliary support for the patient's upper limb.

[0043] In this embodiment of the invention, the body trunk assist device 30 is constructed from lightweight, high-strength flexible materials (such as silicone or composite materials) to form pneumatic artificial muscles, thereby achieving good flexibility and durability.

[0044] In this embodiment of the invention, the central control unit 40 includes: a gait phase recognition unit, a collaborative control algorithm unit, and a personalized training mode unit.

[0045] In this embodiment of the invention, the gait phase recognition unit identifies the support phase and swing phase in the gait cycle in real time based on data from inertial sensors and joint angle sensors. Through a multi-sensor data fusion algorithm, the system can determine the patient's current gait stage with an accuracy rate of over 95%.

[0046] In this embodiment of the invention, the collaborative control algorithm unit employs model predictive control (MPC) or reinforcement learning methods to optimize the coordinated output of treadmill speed and lower limb assist torque, making the gait training process natural, smooth, and energy-efficient. The collaborative control algorithm module dynamically adjusts the torque output curves of each joint drive unit based on gait phase changes, ensuring that the assist torque precisely matches the patient's actual movement intention. When a gait abnormality is detected, the collaborative control algorithm unit can immediately generate a compensation torque command, precisely adjusting the output torque of each joint drive motor.

[0047] In this embodiment of the invention, the algorithm learns the patient's movement patterns in real time, gradually reducing the mechanical assistance force to promote the active participation of the neuromuscular system. The system's built-in force feedback sensor accurately captures the patient's exertion intention, dynamically adjusting the assist torque within 0.1 seconds. Furthermore, through electromyography (EMG) sensors, the activation status of eight major muscle groups in the lower limbs can be monitored in real time. Machine learning algorithms analyze muscle synergistic contraction patterns to provide a quantitative assessment of neuromuscular function recovery. During training, the system automatically fine-tunes the assist torque distribution based on changes in EMG signals, focusing on strengthening the neural recruitment ability of weak muscle groups.

[0048] In this embodiment of the invention, the personalized training mode unit is used to support automatic switching between passive training, active assistance, and resistance training modes according to the patient's rehabilitation stage.

[0049] In this embodiment of the invention, the active assistance mode may include several rehabilitation stages, and the treadmill platform 20 corresponding to each rehabilitation stage has a corresponding assistance coefficient K.

[0050] In some embodiments, the rehabilitation process is divided into at least four stages: During the inflammatory phase (0–2 weeks post-surgery), the assist strength coefficient K = 0.15–0.25; During the callus stage (2–6 weeks), K = 0.30–0.45; During the callus formation stage (6–12 weeks), K = 0.50–0.70; Renovation period (>12 weeks), K=0.75–1.00.

[0051] The assist strength coefficient K is defined as the ratio of the peak torque output by the motor to the predicted maximum isometric muscle strength (MVC) of the patient, and K is ≤1 at any stage to avoid generating additional shear stress on the fracture site exceeding 20% ​​of the treadmill platform.

[0052] The MVC prediction value is obtained through preoperative cloud records or isometric contraction measurements on postoperative day N, and is automatically updated weekly using the following formula: MVC(t)=MVC0·e^(−0.015·t)+α·ΣF_load(t), Where MVC(t) is the maximum voluntary muscle contraction force at time t, MVC0 is the maximum voluntary muscle contraction force at the initial time (t=0), e^(−0.015·t) is the natural decay term of MVC when there is no external load, t is the number of days after surgery, α=0.8 is the gain constant, and ΣF_load(t) is the average voluntary force of patients recorded by the system in the past week.

[0053] Among them, the assist curves for the inflammatory phase and the cartilage callus phase adopt a ramp-plateau composite waveform: the ramp segment lasts for 200ms, with a peak force not exceeding 0.4·K·MVC; the plateau segment is maintained for 300ms with a constant force value, which is used to avoid impact load on the fracture ends; in this way, while ensuring walking function, the maximum shear force at the fracture ends is controlled to be below 0.8 times the body weight.

[0054] The present invention also includes a cloud data management platform 50, wherein the central control unit 40 is communicatively connected to the cloud data management platform 50 for storing the exercise parameters, physiological indicators and rehabilitation progress of each training session, so that doctors can access the data and issue training tasks to the central control unit 40 through the cloud data management platform 50.

[0055] The central control unit 40 of the present invention is also used to monitor the patient's postural stability, abnormal gait or risk of falling in real time, and to trigger an emergency braking or support intervention mechanism when an abnormality is detected.

[0056] In this embodiment of the invention, the telescopic seat 302 is also provided with an automatic extension rod 304 for limiting and supporting the patient on both sides. There are two automatic extension rods 304, and the outer side of the automatic extension rod 304 is covered with a soft rubber sleeve.

[0057] In this embodiment of the invention, the central control unit 40 is further configured to establish a gait feature database using the inertial measurement unit and pressure sensors built into the treadmill platform 20, combined with a pre-built deep learning model, to determine whether the patient is at risk of falling. If so, it controls the body trunk assist device 30 to extend the automatic extension bar 304, while simultaneously controlling the treadmill platform 20 and the lower limb assist device 10 to reduce their speed. Specifically, when the central control unit 40 detects that the center of gravity shift exceeds a safety threshold, the system immediately controls the automatic extension bar 304 to extend, while simultaneously reducing the treadmill speed to a safe range. This dual protection design significantly improves the safety of rehabilitation training.

[0058] This invention improves gait phase recognition accuracy to ±2.3° through multi-sensor fusion technology and employs an adaptive Kalman filter algorithm to eliminate motion artifact interference. An electromyography (EMG) signal acquisition module monitors the activity status of key muscle groups such as the rectus femoris and vastus lateralis in real time at a sampling rate of 1000Hz, and constructs a biomechanical model by combining this data with three-dimensional force platform data. This model can accurately calculate the lower limb joint torque distribution and dynamically optimize the assist curve through an iterative learning control algorithm.

[0059] This invention aims to provide patients with a flexible rehabilitation exercise program. When a patient is physically weak, they can choose to wear a specially designed lower limb assistive device 10, which effectively provides support and assistance to help them perform targeted rehabilitation exercises on the treadmill platform 20. To further ensure patient safety during exercise, the patient can also wear a trunk assistive device 30, which provides additional stability and protection to prevent accidents. In this way, even when a patient is physically weak, they can still perform effective rehabilitation training in a safe environment.

[0060] When the patient's health is relatively good, they can choose to wear only the lower limb assistive device 10, freeing them from the limitations of the treadmill and allowing them to engage in independent activities outdoors. This flexible application not only meets the rehabilitation needs of patients in various scenarios but also effectively enhances their self-confidence and quality of life, truly achieving the rehabilitation goal of multi-scenario application. Whether indoors or outdoors, whether weak or healthy, this invention provides patients with comprehensive rehabilitation support.

[0061] As attached Figure 2 As shown, the present invention also discloses a control method for an intelligent lower limb walking aid, the method comprising the following steps: S10-S30.

[0062] S10. Real-time collection of patient gait parameters, lower limb joint motion data, and treadmill operating status; In this embodiment of the invention, gait parameters include temporal-spatial parameters, kinematic parameters, and dynamic parameters. Gait parameters are collected based on sensors installed in the body trunk assist device 30 and the central control unit 40.

[0063] In this embodiment of the invention, the central control unit 40 is also used to process gait parameters and lower limb joint motion data using an adaptive Kalman filter algorithm to eliminate motion artifact interference.

[0064] S20. Use patient gait parameters and lower limb joint motion data to perform gait phase recognition, obtain gait phase recognition results, and predict the gait rhythm of the next cycle based on the gait phase recognition results.

[0065] Gait phase recognition is performed using patient gait parameters and lower limb joint motion data, including: Feature extraction is performed on the gait parameters and lower limb joint motion data to obtain feature data. The feature data is then fused using a multimodal sensor fusion algorithm to obtain data to be identified. Gait phase recognition results are obtained based on the data to be identified.

[0066] In this embodiment of the invention, S20, predicting the gait rhythm of the next cycle based on the gait phase recognition result, includes: S201. Based on the gait phase recognition results, extract gait rhythm features and establish a rhythm model.

[0067] In this embodiment of the invention, the rhythm model can be either a uniform velocity model or a state-space model. The state-space model uses the "true gait phase" and the "phase change rate (i.e., rhythm)" as the system's state variables. This state is continuously updated and optimally estimated using sensor observations (identified phases). The advantage of this model is that it can not only estimate the current state but also naturally predict future states. The state-space model is updated in real-time using a Kalman filter algorithm, effectively handling sensor noise and improving prediction accuracy. This model is particularly suitable for rehabilitation patients with unstable gait, dynamically adjusting prediction parameters according to their actual movement state to achieve personalized gait rhythm prediction. The prediction results serve as an important input to the collaborative control algorithm, ensuring that the walker's output remains highly synchronized with the patient's movement intentions. During gait rhythm prediction, the system continuously monitors prediction errors and dynamically compensates for them.

[0068] S202. Based on the rhythm characteristics, a rhythm model is established to predict the gait rhythm of the next cycle.

[0069] S30. Based on the prediction results, dynamically adjust the treadmill speed and the driving torque of the lower limb assist device to match the patient's natural gait kinematically. Specifically, calculate the control parameters based on the prediction results to obtain the treadmill speed adjustment and the driving torque of the lower limb assist device for dynamic adjustment.

[0070] This invention utilizes advanced data acquisition technology to accurately analyze and predict the next gait phase a patient is about to enter. Based on these detailed data analysis results, the system can provide targeted, precise, and effective assistance for the patient's lower limb movements. This assistance not only significantly improves the patient's exercise efficiency and stability but also effectively reduces potential risks during exercise. Furthermore, the treadmill equipped with this invention features intelligent adaptation, flexibly adjusting its speed according to the patient's specific needs and exercise status, ensuring the patient remains safe and comfortable throughout exercise, thereby comprehensively improving the effectiveness and safety of rehabilitation training.

[0071] It should be understood that, as used herein, the singular form "a" is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations of one or more of the associatedly listed items. The embodiment numbers disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0072] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples. Within the framework of the invention, technical features of the above embodiments or different embodiments can be combined, and many other variations of different aspects of the invention exist, which are not provided in detail for the sake of brevity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. An intelligent lower extremity walking aid, characterized by, The method comprises: a lower limb assisting device (10) worn on the lower limbs of a patient to provide joint driving force to assist walking; A treadmill platform (20) for providing a controllable moving surface to support the patient to perform gait training; A body trunk assisting device (30) installed on the treadmill platform (20) for supporting the trunk part of the patient to reduce the burden of the upper limbs and maintain body balance; And a central control unit (40), the lower limb assisting device (10), the treadmill platform (20) and the body trunk assisting device (30) are all electrically connected with the central control unit (40), the central control unit (40) is used for collecting the gait parameters of the patient, the speed of the treadmill and the driving state of the lower limb assisting device in real time, and based on a multi-modal sensor fusion algorithm, the speed of the treadmill and the assisting output of the lower limb assisting device are dynamically adjusted to realize the adaptive synchronization of the speed of the treadmill and the gait rhythm of the patient.

2. The intelligent lower leg walker according to claim 1, characterized in that The lower limb assisting device (10) comprises a mounting waist belt (101), a hip joint driving block (102) and a knee joint driving block (103), the mounting waist belt (101) is used for being sleeved on the waist of the patient, the hip joint driving block (102) is installed on the mounting waist belt (101), the hip joint driving block (102) is used for driving the hip joint of the patient, the knee joint driving block (103) is installed on the mounting hip joint driving block (102), and the knee joint driving block (103) is used for driving the knee joint of the patient; a first binding belt is further arranged on the hip joint driving block (102) and used for sleeving the hip joint driving block (102) on the thigh of the patient, and a second binding belt is arranged on the knee joint driving block (103) and used for sleeving the knee joint driving block (103) on the calf of the patient.

3. The intelligent lower leg walker according to claim 2, characterized in that The mounting waist belt (101) and the hip joint driving block (102) are all provided with electromyographic sensors, force sensors, angle sensors and inertial measurement units.

4. The intelligent lower leg walker according to claim 1, wherein The body trunk assisting device (30) comprises a support (301), a telescopic seat (302) and a binding assembly (303), the support (301) is fixedly installed on the treadmill platform (20), the telescopic seat (302) is installed on the support (301), and the binding assembly (303) is arranged at the output end of the telescopic seat (302) and used for assisting and supporting the upper limbs of the patient.

5. The intelligent lower leg walker according to claim 4, characterized in that The binding assembly (303) comprises a shoulder support pad and not less than one fixing belt.

6. The intelligent lower leg walker according to claim 1, wherein The central control unit (40) comprises a gait phase recognition unit, a cooperative control algorithm unit and a personalized training mode unit; The gait phase recognition unit recognizes the support phase and swing phase in the gait cycle in real time based on inertial sensor and joint angle sensor data; The cooperative control algorithm unit is used for optimizing the cooperative output of the speed of the treadmill and the lower limb assisting torque by using a reinforcement learning method, so that the gait training process is natural, smooth and energy-saving; The personalized training mode unit is used for supporting automatic switching of passive training, active assistance and resistance training modes according to the rehabilitation stage of the patient.

7. The intelligent lower leg walker according to claim 1, wherein The active assistance mode can include several rehabilitation stages, and the several rehabilitation stages correspond to the treadmill platform (20) with corresponding assistance coefficients K.

8. The intelligent lower leg walker according to claim 7, characterized in that The central control unit (40) is also used for monitoring the patient posture stability, abnormal gait or falling risk in real time, and triggering an emergency braking or support intervention mechanism when an abnormality is detected.

9. The intelligent lower leg walker according to claim 1, wherein A cloud data management platform (50) is also included, which is in communication connection with the central control unit (40) and is used for storing the motion parameters, physiological indexes and rehabilitation progress of each training, so that the doctor can call the data.

10. A control method of an intelligent lower extremity walking aid, characterized by, It comprises: Real-time acquisition of patient gait parameters, lower limb joint motion data and treadmill running state; Using the patient gait parameters and lower limb joint motion data to identify the gait phase to obtain the gait phase identification result, and predicting the next cycle gait rhythm based on the gait phase identification result; According to the prediction result, dynamically adjust the speed of the treadmill and the driving torque of the lower limb assistance device, so that they are kinematically matched with the natural gait of the patient.

Images