Control method and system of intelligent exoskeleton device for hip joint physical training

Abstract

The invention discloses a control method and system of an intelligent exoskeleton device for hip joint physical training, an inertial measurement unit is fixed in a thigh side support structure, and two sole pressure sensors for acquiring heel pressure and tiptoe pressure are arranged at the sole; the main control module is used for receiving data of the inertial measurement unit, the plantar pressure sensor and the power output device, and the main control module is in communication connection with the upper computer; a user wears the intelligent exoskeleton device, and multiple training modes are preset in an upper computer interface; a user clicks a corresponding function key through a mode selection area of the upper computer and sends a training mode instruction to the lower computer controller; after the controller receives the corresponding instruction, the internal training state structural body is set to be in a corresponding mode; initializing the sensor; resistance parameters and loading of the resistance parameters are carried out; starting an action recognition algorithm; output control of motor torque of the power output device is carried out; and recording and evaluating training indexes.

Description

Technical Field

[0001] This invention relates to the field of intelligent sports training technology, and specifically to a control method and system for an intelligent exoskeleton device used for hip joint sports training. Background Technology

[0002] The hip joint is the main power joint in human walking, running, going up and down stairs, etc. Insufficient or asymmetrical hip flexion, extension and extension movements will directly affect gait stability and athletic performance.

[0003] Existing hip training methods have the following problems: 1. Action recognition relies on manual judgment and cannot provide real-time feedback on user actions; 2. Assisted training relies on fixed resistance / speed equipment (such as rehabilitation power bikes), which cannot adapt to the human body's actual gait; 3. The lack of high-frequency sensor data fusion technology makes it difficult to perform high-precision hip joint motion analysis; 4. It lacks intelligent algorithms to dynamically adjust the assist intensity and cannot adjust it in real time according to changes in the user's fatigue level. Summary of the Invention

[0004] Therefore, the present invention provides a control method and system for an intelligent exoskeleton device for hip joint sports training to solve the above-mentioned problems in the prior art. To achieve the above objective, the present invention provides the following technical solution: According to a first aspect of the present invention, a control method and system for an intelligent exoskeleton device for hip joint sports training, the intelligent exoskeleton device includes a hip area main load-bearing structure, a thigh side support structure, and a hip joint rotation mechanism connecting the two. A power output device is integrated within the hip joint rotation mechanism. An inertial measurement unit is fixed in the thigh side support structure. The inertial measurement unit is used to acquire real-time hip joint angle and angular velocity. Two plantar pressure sensors are provided on the sole of the foot to acquire heel pressure and toe pressure. A main control module is used to receive data from the inertial measurement unit, the plantar pressure sensors, and the power output device. The main control module is communicatively connected to a host computer. The control method includes the following steps: Step S100: The user wears the smart exoskeleton device and presets multiple training modes in the host computer interface; Step S200: The user selects the region through the mode selection on the host computer, clicks the corresponding function button, and sends the training mode command to the lower-level controller. Step S300: After receiving the corresponding instruction, the controller sets the internal training state structure to the corresponding mode. Step S400: Initialize the sensor; Step S500: Apply resistance parameters and load resistance parameters; Step S600: Enable the action recognition algorithm; Step S700: Perform output control of the motor torque of the power output device; Step S800: Record and evaluate the training indicators.

[0005] Furthermore, the training modes set in step S100 include, but are not limited to, hip flexion training mode, hip extension training mode, hip extension training mode, high knee training mode, squat training mode, and freestyle leg kick training mode.

[0006] Further, step S400 specifically involves zero-position calibration of the thigh inertial measurement unit and static calibration of the plantar pressure sensor to obtain the hip joint flexion-extension angle, static heel pressure, and static toe pressure as initial reference values.

[0007] Further, step S500 specifically involves reading the basic resistance level, assistance coefficient, target hip joint range of motion, target frequency, and target interval training parameters for this step's energy from a preset parameter table according to the selected mode.

[0008] Further, step S600 specifically involves calling the motion recognition algorithm module corresponding to the mode to perform real-time analysis of the inertial measurement unit angle, angular velocity, and plantar pressure signals to determine the current phase.

[0009] Further, step S700 specifically involves outputting directional resistance or assist torque according to the mode type when a valid training phase is identified; and reducing the motor torque to zero or setting it to a safe value when a non-training phase or abnormal state is not identified.

[0010] Furthermore, step S800 specifically involves calculating the energy, motion performance index, motion quality index, and left-right symmetry index for each complete movement at the end of the movement, and displaying them in real time on the host computer.

[0011] Furthermore, the energy in step S800 The calculation formula is: In this formula, This represents the energy produced in a single action. Indicates the starting point of a complete action. Indicates the end time of a complete action. Indicates instantaneous power. Represents the integral over time; Instantaneous power The calculation formula is: In this formula, This represents the real-time torque that the motor outputs or acts in the opposite direction on the user's hip joint at time t. This represents the angular velocity of the user's hip joint at time t; The formula for calculating the Performance Index (MPI) is as follows: In this formula, This represents the peak angular velocity of the hip joint during one movement cycle. This represents the peak power of the hip joint, and ROM represents the range of motion of the hip joint during the middle of this training session. Sym The index represents left-right symmetry, and the index represents smoothness. ~ Indicates the weight of each indicator; The formula for calculating the range of motion (ROM) of the hip joint is: In this formula, This indicates the maximum hip angle during this movement. This indicates the minimum hip angle for this movement.

[0012] Furthermore, it also includes step S900: when the automatic training mode or intelligent adjustment option is selected, the system automatically adjusts the resistance level, resistance magnitude and training rhythm based on the changes in energy, hip joint range of motion and athletic performance index of the most recent steps, so as to keep the training intensity within the target range.

[0013] According to a second aspect of the present invention, a control system for an intelligent exoskeleton device for hip joint sports training is provided, the system being used to execute a control method for an intelligent exoskeleton device for hip joint sports training as described in any of the first aspects of the present invention.

[0014] The present invention has the following advantages: The control method and system of the intelligent exoskeleton device for hip joint sports training of the present invention is a hip joint intelligent exoskeleton system for sports training. It monitors the dynamics of the hip joint in real time through a multi-sensor fusion algorithm, and realizes the quantification, intelligence and personalization of sports training through functions such as assistance, resistance, movement quality assessment and automatic training program generation. Attached Figure Description Figure 1 A flowchart illustrating a control method for an intelligent exoskeleton device for hip joint sports training, provided for some embodiments of the present invention.

[0015] Figure 2 This is a structural diagram of an intelligent exoskeleton device in a system for hip joint sports training, provided for some embodiments of the present invention.

[0016] Figure 3This is a schematic diagram of the foot sensor installation structure of an intelligent exoskeleton device for hip joint sports training provided in some embodiments of the present invention.

[0017] 1. Hip area main load-bearing structure; 2. Thigh side support structure; 3. Hip joint rotation mechanism; 4. Inertial measurement unit; 5. Waist strap; 6. Leg strap; 7. Toe and foot pressure sensor; 8. Heel and foot pressure sensor; 9. Sub-control box; 10. Main control box. Detailed Implementation

[0018] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. 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.

[0019] like Figure 1 As shown, this invention provides a control method for an intelligent exoskeleton device for hip joint sports training. Based on a hip joint exoskeleton hardware system, a sensor system, and a main control algorithm system, a closed-loop control process is constructed, encompassing motion recognition, torque control, training evaluation, and automatic adjustment. In the hip joint sports training system of this invention, the host computer interface presets multiple training modes, including but not limited to hip flexion training mode, hip extension training mode, hip extension training mode, high knee training mode, squat training mode, and freestyle leg kick training mode. Users can send training mode commands to the lower-level computer by clicking the corresponding function button in the mode selection area of ​​the host computer. Upon receiving the command, the controller sets its internal training state structure to the corresponding mode.

[0020] After the mode is set, the system will execute the following steps in sequence: 1. Sensor Initialization: Perform zero-point calibration on the thigh IMU and static calibration on the foot pressure sensor to obtain... = 0°、 , As an initial reference value; 2. Resistance / Assist Parameter Loading: Based on the selected mode, the base resistance level is read from the preset parameter table. Assistance coefficient Training parameters include target ROM range, target frequency, or target energy range for this step; 3. Motion recognition algorithm enabled: The motion recognition algorithm module corresponding to this mode is invoked to perform real-time analysis of IMU angle, angular velocity and plantar pressure signals to determine the current phase (such as hip flexion phase, hip extension phase, hip extension phase, leg kick phase, etc.). 4. Motor torque output control: When a valid training phase is identified, directional resistance or assist torque is output according to the mode type; in non-training phases or abnormal conditions, the motor torque is reduced to zero or set to a safe value. 5. Training Indicator Recording and Evaluation: Calculate the energy for each step at the end of each complete movement. The system displays indicators such as the MPI (Motor Performance Index) and left-right symmetry in real time on the host computer. 6. Automatic parameter adjustment (AUTO function): When the "Automatic Training Mode" or "Intelligent Adjustment" option is selected, the system can adjust parameters based on the most recent training data. Changes in ROM and MPI automatically adjust the resistance level, assist level, and training pace to keep the training intensity within the target range.

[0021] Through the above-described setup method, this invention not only enables free switching between multiple training modes, but also allows for adaptive parameter adjustment based on the user's actual performance in each mode, making the training process more intelligent and personalized.

[0022] The intelligent hip exoskeleton training system of this invention comprises a wearable exoskeleton module, a sensor acquisition module, a drive execution module, a communication module, and a safety control module. The system employs a lightweight carbon fiber structure, a modular dual-side drive design, a replaceable strap system, and an intelligent sensor fusion architecture to meet the high-dynamic environment requirements of sports and rehabilitation training.

[0023] like Figure 2 As shown, the wearable exoskeleton structure consists of several parts: a hip main load-bearing structure 1, a thigh side support structure 2, a hip joint rotation mechanism 3, and a main control box 10. The hip main load-bearing structure 1 is made of high-temperature hot-pressed integrally molded carbon fiber and is fixed to the waist via a waist strap 5. It possesses the following properties: high strength / high modulus, employing a multi-layer woven carbon fiber and resin matrix composite structure, capable of withstanding the torque generated by the hip joint during high-speed, high-load movements; lightweight, with a typical weight of... 300g, significantly lower than metal hip frames (typically) (1kg); wearable and conforms to the contours of the human pelvis, secured to the waist with flexible straps, ensuring stability during running, high knees, and swimming leg kick training; usable on one or both sides, with pre-installed mounting points for two actuators on the frame, supporting unilateral hip joint training, bilateral symmetrical training, and training with different intensities for each actuator (sports rehabilitation scenarios). The thigh side support structure 2 is made of carbon fiber and is fixed to the outer thigh via leg straps 6, featuring: an arc-shaped wrapping structure that conforms to the human muscle and does not affect running movements; a length approximately half the thigh length, effectively transmitting hip joint torque; flexible straps to adapt to different user body types; a thigh IMU inertial measurement unit 4 fixed in the middle of the thigh side support structure 2 to avoid vibration; and a carbon fiber structure that forms a rigid torque transmission path to ensure stable training results. The main control box 10 can be directly attached to the waist or placed in the installation bag and then attached to the waist belt. The main control box 10 is equipped with a control chip for controlling the exoskeleton, a motor battery, and a voice chip. The main control box 10 communicates with the host computer via Bluetooth. The hip joint rotation mechanism has the following features: the rotation axis is basically coincident with the human anatomical axis, reducing rotational deviation and improving the naturalness of training; it adopts a small planetary gear or harmonic reducer, and the output torque can reach 15–30 Nm; it has a quick plug-and-play structure, and users can complete the disassembly and assembly within 10 seconds; it adopts a "side-mounted drive", with the driver installed on the outside of the thigh, rather than the back or waist, which can reduce cross-interference and improve dynamic response; it supports dual-side modular combination, with the left and right drivers operating independently, and can also achieve synchronous and mirror training through the host computer.

[0024] In this invention, the thigh IMU (Inertial Measurement Unit) module is fixed to the middle of the carbon fiber support plate on the outer side of the thigh and is used to collect triaxial angular velocity, triaxial acceleration, and attitude angles (after attitude calculation). Its function is to obtain real-time hip joint angles and angular velocities through the thigh IMU. To unify the training algorithms for the left and right legs, the system adopts direction normalization processing.

[0025] Angle normalization formula: In this formula, The original hip flexion-extension angle was measured by a thigh IMU. Directional factors for left and right legs: right leg = +1; left leg = -1. To unify the coordinates, hip flexion is positive and hip extension is negative. To standardize the angles, they can be directly used for algorithm training. Since the left and right legs appear to have opposite directions—for example, when the left leg is raised forward, the IMU might measure a negative angle—multiplying by lr and -1 makes both legs conform to a unified standard. Thigh forward (hip flexion), thigh backward (hip extension) greatly simplifies the bilateral algorithm, allowing both legs to operate under the same control logic. Calculate the hip joint flexion-extension angle. Calculate the angular velocity of the hip joint. It provides a "leg lift acceleration" feature value for determining the onset of hip flexion: It provides periodic components for recognizing swimming leg kick rhythm; IMU frequency: 50–200 Hz.

[0026] like Figure 3 As shown, the insole has two built-in pressure sensors: a toe-foot pressure sensor 7 located at the forefoot and a heel-foot pressure sensor 8 located at the heel. Both are electrically connected to a control box 9 via signal and wire lines. The control box 9 communicates with a host computer via Bluetooth. The control box 9 also contains a battery to power the foot pressure sensors: heel pressure... and toe pressure It is used for gait event recognition, determining the support / swing phase, identifying hip propulsion behavior (toe rise, heel fall), squat weight transfer, and determining "non-weight-bearing mode" in swimming training; calibration method: standing still for one second. and As a threshold benchmark, plantar pressure is used to determine the type of movement, specifically, for toe-landing: In this formula, Indicates real-time toe pressure. Indicates the pressure on the toes at rest. Indicates the threshold for judgment (e.g., the static pressure when a person is standing); heel strike judgment: In this formula, Indicates real-time heel pressure. Indicates static heel pressure. This indicates the threshold value used to determine when the heel strikes the ground.

[0027] Introduction to the dual-modular structure: The left module is exactly the same as the right module. You only need to select the left (left side) / right (right side) role through the software; you can choose any training method: unilateral training (such as rehabilitation), bilateral synchronous training, bilateral mirror training, and left-right difference training (asymmetrical training); the exoskeleton does not require rigid connection, and the left and right legs are completely independent, unlike most one-piece structures on the market (with a back beam). Advantages: more comfortable to wear, does not restrict trunk movement, does not interfere with running / swimming training, and is compatible with different body types.

[0028] The brushless motor and reduction gear (harmonic / planetary) are connected to the main controller via CAN communication. The motor can perform: maximum current (torque) control, target position control, zero torque mode, and automatic assist / resistance adjustment. The MCU, or main control module, is responsible for the following tasks: receiving data from the IMU, pressure, and motor current; executing motion recognition algorithms; implementing torque control strategies; communicating with the host computer; communicating with the motor driver (CAN / TWAI); and outputting metrics such as step energy and motion performance index (MPI).

[0029] The host computer is responsible for: displaying angle, speed, power, and energy curves; displaying movement phase in real time; adjusting resistance levels; selecting training modes (hip flexion / hip extension / hip extension / swimming leg kick, etc.); displaying step energy and performance index (MPI); and exporting training reports.

[0030] The following explains the meaning of the variables in this invention: This represents the first threshold for angular velocity, indicating the minimum angular velocity used to detect whether a user actively lifts their leg; it is typically taken as 10–20° / s. > This indicates that the user is genuinely beginning to flex their hip, rather than experiencing noise fluctuations. This represents the second threshold for angular velocity, used to identify the thigh's return (descent) process, and is generally taken as -5 to -15° / s; This indicates the acceleration threshold; if the IMU forward acceleration... If so, it can be determined that the user has a clear tendency to swing forward; This indicates the threshold for changes in forefoot pressure. This indicates reduced load on the toes, a typical characteristic of leg raises.

[0031] In this invention, instantaneous torque is used. With angular velocity Calculate instantaneous power The energy is integrated over the complete hip joint movement cycle to obtain the energy per step. ; and according to Adaptive adjustment of resistance or assist output is crucial in strength training, running training, sports performance analysis, and swimming leg training. More importantly, it's essential to understand how much energy each hip joint outputs, the amount of active energy, the amount of motor-assisted energy, whether the left and right legs are symmetrical, and whether the output is stable with each step. Traditional methods rely on ground reaction platforms or motion capture systems, which are unsuitable for wearable devices. A method is needed to quantify the true training intensity and energy contribution of each step. This invention proposes a Step Energy model based on "instantaneous power integral of the hip joint." This model directly utilizes: IMU (angular velocity), motor torque (converted from current), plantar pressure (gait events), and time integral to construct a realistic energy measurement formula consistent with physics, achieving a complete quantification process of "energy per step."

[0032] This step's energy is the energy integral between each complete motion cycle, specifically... The calculation formula is: In this formula, This represents the energy produced in a single action. Indicates the starting point of a complete action. Indicates the end time of a complete action. Indicates instantaneous power. Represents the integral over time; where instantaneous power The calculation formula is: In this formula, This represents the real-time torque that the motor outputs or acts in the opposite direction on the user's hip joint at time t. It represents the angular velocity of the user's hip joint at time t; it is a comprehensive representation of force × velocity × angle, similar to an "energy ledger for each step".

[0033] The classification of active and passive energy is determined by the signs of torque τ and angular velocity ω. This is a key innovation of the Step Energy system: it can distinguish between "work done by the user" and "work done with the help of the motor", which can be used to determine whether the user is slacking off, to determine the true force output, and to assess "active participation" in sports training.

[0034] This system uses an energy closed-loop regulation system to adjust training resistance / assistance. Specific rules: When… < This indicates that the resistance is too high, and therefore it automatically reduces resistance; when > This indicates that the resistance is too small, and thus automatically increases the resistance; when A continuous decline indicates fatigue, which will then automatically increase assist; when If the difference between the left and right sides is large, automatic symmetry training will be performed.

[0035] The host computer (HMI) can display in real time: the current instantaneous power curve (Pt curve), single-step energy bar chart, left-right energy comparison, movement cycle energy envelope, swimming leg kick energy radar chart, and historical training energy trends; and can generate: training reports, rehabilitation reports, swimming-specific training reports, and symmetry correction suggestions.

[0036] Traditional training methods cannot display training indices. To fully reflect movement quality, strength, stability, bilateral symmetry, and fatigue level, this invention proposes the Motion Performance Index (MPI) for adjusting training difficulty. The MPI is defined as a comprehensive training ability indicator composed of five physical signals. To achieve quantitative evaluation of hip joint training effects, this invention proposes a comprehensive scoring model based on speed, power, activity level, symmetry, and movement smoothness. The Motion Performance Index (MPI) is defined as follows: In this formula, This represents the peak angular velocity of the hip joint during one movement cycle. This represents the peak power of the hip joint, and ROM represents the range of motion of the hip joint during the middle of this training session. Sym The index represents left-right symmetry, and the index represents smoothness. ~ This indicates the weight of each indicator.

[0037] MPI's Five Core Indicators: Logistics Definitions and Formulas, Peak Hip Angular Velocity =max|ω(t)|, which means that the larger the value, the stronger the explosive power. It is used to evaluate running, high knees, and the upward swing motion in swimming; Peak hip power: Meaning: Reflects the combined output of "strength × speed," determining running propulsion, hip extension ability, and squat recovery ability; the formula for calculating the hip joint range of motion (ROM) is: In this formula, This indicates the maximum hip angle during this movement. This indicates the minimum hip joint angle during the movement, representing hip joint flexibility, and is a key parameter in both sports training and rehabilitation training.

[0038] For the sake of left-right symmetry, see the following equations: ROM symmetry: ; Power symmetry: ; Gait symmetry (energy): ; Final synthesis: ,in: , , These are weighting coefficients. Significance: For athletes: it can identify "unilateral control"; for sports rehabilitation patients: it can assess the recovery progress of the affected side; for swimming leg kicks: it can identify rhythm errors between the left and right legs.

[0039] Smoothness (jerk stability) is defined as the derivative of acceleration. It reflects the drastic degree of acceleration change.

[0040] Smoothness index: ; Jerks can cause discomfort while riding (such as the bumpy feeling when riding in an elevator or a car accelerating rapidly), and can increase wear and vibration in mechanical systems. Square integrals can produce smoother, more comfortable, and gentler trajectories. In the field of robotics, reducing jerk helps improve motion accuracy and reduce residual vibration; a higher smoothness indicates smoother movements, while a lower smoothness indicates jittery, unstable, and poor technique, which is especially applicable to judging swimming leg kick techniques.

[0041] The significant role of MPI in training: In intelligent difficulty adjustment (AUTO mode), if the MPI is too low, assistance is needed to increase it and resistance to decrease; if the MPI is too high, resistance needs to be increased. Long-term tracking of MPI can help determine strength development trends. In swimming leg training, MPI reflects the energy of the upper and lower swings, left-right symmetry, leg rhythm, smoothness, whether the leg technique is correct, whether the lower leg is involved in incorrect force generation, whether there are pauses mid-strike, and whether there is excessive knee flexion.

[0042] In this invention, the system further provides multiple hip joint training modes. By performing motion recognition and phase analysis on thigh IMU posture, hip angular velocity, and plantar pressure signals, and combining them with real-time torque control strategies, different hip-specific training can be achieved. Among them, hip flexion training and hip extension training are the most basic modes and are the mother movements of all other training modes (such as hip extension, high knees, squats, running mode, swimming leg kicks), and have a core position in the training system.

[0043] I. Hip flexion training mode, hip flexion movement recognition rules: The characteristics of hip flexion are the raising of the thigh forward and upward from a neutral position, accompanied by an increase in angle, a positive angular velocity, and a decrease in plantar pressure. This invention divides hip flexion recognition into three stages: 1. Action initiation recognition stage. Angular velocity threshold condition (proactive determination). ,in: This represents the hip flexion initiation velocity threshold, typically 10–20° / s. It indicates the user is actively lifting their leg, rather than experiencing noise or jitter; the forward acceleration of the thigh IMU is increased. > In this formula, Indicates acceleration. This represents the forward acceleration at a certain moment. This indicates the forward movement threshold, meaning the forward acceleration of the thigh IMU is greater than the threshold, the thigh is swinging forward, and the hip flexion trend is clear; forefoot pressure decreases. < - In this formula, a decrease in pressure on the forefoot indicates the start of a leg lift, effectively distinguishing between a "leg lift" and a "swing in place"; the angle and direction are also confirmed. > In this formula, = 5°, indicating that the thigh has truly left the neutral position. The training control system utilizes the normalized hip angle. and Determining the occurrence of hip joint movement: Conditions for hip flexion: >0 >0, meaning: The thigh is raised forward; the thigh is accelerating upward. Initial threshold determination: > > ,in = 5°, leaving neutral =1° / s: The minimum effective speed determines the entry into the lifting phase after the user actively participates.

[0044] 2. Hip flexion and ascent phase (peak position). As the user continues to lift their leg, the angle continues to increase: rise, >0 The system enters constant resistance or boost mode; when the peak value is reached, , ,in, It is the expected target hip flexion angle (which can be 25°, 35°, or 45°, depending on the training mode). This indicates that the motion has reached its peak, at which point the return phase begins; 3. Hip flexion return detection. As the thigh begins to descend... And return to the neutral zone, < The action is counted and completed.

[0045] II. In the hip extension training mode, hip extension is used to train the gluteus maximus, hamstrings, and running backswing strength. It improves sprinting ability, stride length, and explosiveness. Advantages of hip extension recognition: It is extremely accurate in recognizing the running propulsion phase (backswing), can be used for glute strength training, and also has adaptability value for swimming leg kicks (backswing phase feature matching); the hip extension movement is the reverse of the hip flexion movement, but has obvious biomechanical characteristics: the angle moves into the negative zone, the angular velocity is negative, and the heel pressure increases.

[0046] 1. Hip extension initiation conditions. <- ,in: It is the hip extension velocity threshold (typically 8-20° / s) and satisfies: , Angle threshold, typically: This indicates that the thigh is swinging backward; 2. Increased pressure on the heel. + ,in, Real-time heel pressure, The static heel pressure, and the change in this pressure, are used to confirm that hip extension is a "driven" backswing, rather than a random swing.

[0047] 3. Determining peak hip extension. When: ,and Reaching peak, common peak hip extension angle The range is: -15° to -25°. 4. Determine the hip extension return stroke. When the angle approaches the neutral zone again, -5°, action count +1.

[0048] III. Hip Extension Training Mode: This mode trains hip-dominant movements, enhancing the coordinated power of the gluteus maximus, hamstrings, and core muscles to improve stride length, acceleration, and sprinting ability. The system uses an IMU and plantar pressure sensors to identify hip extension phase characteristics in real time, providing assistance or resistance in the direction of hip extension based on the movement phase. Traditional systems can hardly identify hip extension movements. This invention achieves effective detection through IMU and plantar pressure sensors. Significance of Hip Extension Identification: The hip extension movement is a transition from "moderate hip flexion to hip extension." Unlike hip flexion and extension, it often occurs at the moment of hip extension during: the end of the running propulsion phase / squat / deadlift top / step climbing / swimming leg kick. It can identify the moment of hip extension during the running propulsion phase, provide strength training (resistance type) for athletes, improve stride length and stride efficiency for ordinary people, and assess propulsion ability in conjunction with the energy of the current step. Hip extension occurs at the end of the support phase and has the following characteristics: the angle is close to the peak (but not necessarily very large), the angular velocity is close to 0 (i.e., about to exert force), and the plantar pressure on the toes increases significantly (center of gravity shifts forward).

[0049] The biomechanical characteristics of hip thrust are composed of the following three factors simultaneously: 1. The hip angle does not change much, but the acceleration is obvious. The hip thrust is not lifting the thigh, but rather the pelvis is "thrusting" forward and upward. The typical manifestation is: That is, the hip angle is close to the neutral position; 2. The angular velocity changes rapidly in the rearward direction (hip extension direction): This refers to the instantaneous extension of the joint.

[0050] 3. A sharp increase in pressure on the soles and toes (shifting the center of gravity forward): ,in, Represents real-time toe pressure; Represents static toe pressure; This represents the threshold for judgment, and is accompanied by: Where, transd is the forward shift ratio coefficient, which is a typical physical signal for hip delivery: force from the toes → forward shift of the center of gravity → rapid force from the hip to push forward.

[0051] The system uses the cross-judgment of the above three features to identify the hip extension movement. During the hip extension phase: after entering the hip extension phase, the system outputs a training torque in the hip extension direction. Resistance training (strength training): to improve gluteus maximus strength, the system outputs reverse resistance.

[0052] IV. In the squat training mode, the squat is a compound chain movement of the human lower limbs, with its core power coming from hip joint flexion and extension. This invention, based on the fusion of hip IMU and plantar pressure sensors, achieves real-time recognition, phase-by-phase control, resistance / assisted training, and full-cycle ROM and energy analysis of the squat movement. The biomechanical characteristics of the squatting movement mainly consist of three phases: 1. Squatting phase: Hip joint flexion-extension angle Hip joint angular velocity That is, the hip flexion angle increases / the thigh leans forward / the plantar pressure gradually shifts backward, and the heel pressure increases. > This is a typical characteristic of squats; 2. Base stability zone: When the hip angle reaches maximum flexion, 0. The plantar pressure is evenly distributed: The pause in movement is short, but it is the most dangerous and requires the most protection. 3. Rising phase: Hip extension: The pressure on the sole of the foot shifts forward. This is the key stage for the output of explosive power from the lower limbs.

[0053] Squat motion recognition algorithm: Squat recognition uses a trimodal judgment based on IMU + plantar fusion: 1. Changes in hip angle: Typical value: , ; 2. Foot pressure center of gravity shifts backward / forward: Squatting increases heel pressure. When landing on your heels, the pressure on your toes increases as you stand up. Judging by touching the ground with your toes; 3. Range of motion of the hip To meet the squat range: If it is less than this range, it may be a normal hip flexion movement; If all three of the above actions are met simultaneously, it is determined to be a squatting action. This recognition method is highly innovative because the squatting action comes from changes in the hip joint and lower limb kinetic chain, and can only be reliably recognized by using an IMU and the sole of the foot.

[0054] V. In the high knee training mode, the high knee training mode is used to improve hip flexor strength, hip flexion speed, stride frequency rhythm, and forward swing efficiency of running gait. It is the most commonly used movement in running-specific training. This invention achieves accurate recognition and training control of the high knee movement through IMU + foot sensor fusion.

[0055] The biomechanical characteristics of high knee raises differ from those of ordinary hip flexion / running swing, and have the following key features: 1. The maximum hip flexion angle (ROM) is relatively large, typically reaching: 40 Even professional runners can exceed 50°; 2. Extremely high hip flexion speed (rapid swing phase): Extremely high angular velocity means rapid leg lifting; 3. No weight-bearing is required on the soles of the feet (swinging motion); pressure on the toe and heel is close to zero. This is the core characteristic that distinguishes high knees from the propulsive movement in running; 4. Obvious rhythm: continuous leg raises → rapid alternation → high rhythm, which the system can identify through time thresholds.

[0056] High knee raise recognition criteria: High knee raise recognition uses a combination of hip angle and angular velocity for judgment, and does not rely on the sole of the foot. 1. Wide hip flexion: ,in: =35°-45°, this is the minimum standard for high knee raises; 2. High-speed leg raise: in: =80°-120° The higher the speed, the more "running-like" the movement becomes; 3. The sole of the foot is in a swinging direction: This indicates that the object is not bearing weight and is swinging. All three conditions must be met simultaneously for the system to enter high knee mode.

[0057] VI. In the freestyle leg kick training mode, the user leans forward (e.g., lying on a training bed, poolside, or stable table), with legs hanging naturally, and wears the hip joint exoskeleton of this invention. The system detects the left and right hip joint angles and angular velocities to achieve the following motion control.

[0058] 1. Alternating control logic for left and right legs: When the right leg enters the hip extension phase (kicking downwards), the system applies assistance or resistance to the right leg; at the same time, when the left leg enters the hip flexion phase (upwards), the system applies corresponding reverse resistance or assistance; when phase reversal is detected, the system automatically switches between left and right movements.

[0059] 2. Leg kick recognition: The system reads the angle signals of the left and right hip joints. angular velocity With step energy expansion parameters, This leads to the following motion recognition model: when the angle of one hip joint changes rapidly from hip flexion to hip extension and the angular velocity... When the threshold is reached, it is determined to be a "downward kicking phase"; when the angle on that side retracts from hip extension to hip flexion and... When the threshold is negative, it is determined to be "upward recovery phase".

[0060] The system identifies the phase difference between the left and right legs in real time. and according to The approximately 180° freestyle stroke characteristic causes the left and right exoskeleton motors to generate the following controls: Hip extension phase: provides assistance or resistance to aid in the downward kick; Hip flexion phase: provides assistance or resistance in the opposite direction, forming a complete training cycle; Automatic rhythm mode: based on... ,frequency , The output is automatically adjusted to make the training movements closer to the technical characteristics of professional swimmers.

[0061] The system can record each kick in real time. Energy consumption, frequency, and left-right symmetry are used for training assessment or rehabilitation monitoring.

[0062] In one specific embodiment, taking the freestyle leg kick as an example, the application of the present invention in swimming is illustrated.

[0063] 1. Mode selection: "Freestyle kicking", and set the target kicking frequency (e.g., 1.0~1.6Hz), target kicking energy range, and whether to enable AI frequency prediction in the host computer. Simultaneously, enable dual hip joint mirror control in the slave computer, aligning the left and right legs to the equivalent coordinate system of the swimming posture.

[0064] 2. Leg kick recognition: In this mode, the system uses hip angular velocity to recognize the two phases of the downward kick and the upward leg retraction; when >+ When this is determined to be a "downward leg swing phase," it indicates that the thigh is exhibiting a significant downward acceleration and swing relative to the body; when <- At this point, it is determined to be the "upward leg retraction phase," indicating that the thigh swings back from below. Among these, The angular velocity threshold for kicking is used to distinguish between effective kicks and minor vibrations. The system uses one complete cycle of the phase from "downward kick → upward retraction" as one kick cycle, and calculates the time interval based on this. Calculate the frequency of leg kicks: .

[0065] 3. Assist and Resistance Control Phase: During the technique learning phase, "Assist Training" can be selected: In the downward leg kick phase, the motor outputs assist torque in the same direction as hip extension, helping the user form the correct leg kick trajectory and rhythm. In the upward leg retraction phase, the torque is reduced or set to slight damping to ensure the continuity of the movement. During the strength and endurance training phase, "Resistance Training" can be selected: When kicking downward, resistance is output in the opposite direction to hip extension, strengthening the training of the hip extensor muscles and core stabilizing muscles. The resistance magnitude can be related to the leg kick angular velocity, making "high resistance for fast kicks and low resistance for slow kicks," which is closer to the resistance characteristics in real water.

[0066] 4. Kick Energy Extension Parameter and AI Frequency Prediction: For each kick cycle, the system also calculates the kick energy for that kick. (The calculation method is similar to Estep), and energy histograms and frequency curves are plotted on the host computer. Simultaneously, a simplified AI prediction model is used to smoothly predict the frequency of the next leg strike. In this formula, This refers to the frequency of the most recent leg kick. This represents the average kicking frequency during the current training phase. The weighting coefficient is between 0 and 1. This prediction frequency can be used for rhythm training, bi-leg coordination training, and fatigue trend monitoring.

[0067] 5. Training Evaluation and Automatic Adjustment: The host computer displays in real-time leg kick frequency, kick energy, left-right symmetry, and swimming-specific MPI. The system can adjust accordingly. With the target energy range < < The relationship automatically adjusts the resistance level: if If the resistance is significantly below the lower limit, the training intensity is deemed insufficient; therefore, the resistance should be increased appropriately or the target frequency increased. If this occurs repeatedly... If the resistance exceeds the limit, it may indicate excessive fatigue. It is recommended to reduce the resistance or extend the interval.

[0068] In this invention, the core idea of ​​resistance training (strength training) is: the motor output is opposite to the direction of the human body's force exertion, providing "reverse damping," and the resistance control formula is as follows: In this formula, This represents the speed drag coefficient (the higher the speed, the greater the drag). Represents real-time angular velocity. This represents the basic resistance, simulating a resistance band, and corresponds to the resistance levels. The greater the angular velocity, the greater the resistance—simulating resistance band training; when the angular velocity approaches zero, the resistance decreases—protecting the joints; the resistance can be automatically adjusted—used for strength training.

[0069] Specifically, the direction of resistance depends on the movement: when the hip is flexed Then the motor outputs in the reverse direction (backward); when the hip extends... If the motor outputs in the opposite direction (backwards), the motor will output in the opposite direction. When swimming forward, the output will be opposite to the direction of the forward swing of the body. When swimming backward, the output will be opposite to the direction of the backward swing of the body.

[0070] In this invention, assisted training (rhythm training) is used to improve knee lift and hip flexion in running, address insufficient strength in sports rehabilitation, correct leg kick technique in swimming, and assist in climbing stairs. The core objective of this assistance is to compensate for the user's insufficient strength and help complete the movement. To improve training rhythm and help establish a high-frequency leg lift rhythm, the system outputs a consistent "forward assist": In this formula, This indicates the speed boost factor (the faster the speed, the greater the boost). This indicates the basic assist bias, suitable for running technique training and high-knee rhythm training. Peak resistance identification and retracement control: Reaching peak resistance conditions: ≤0 means that the upward swing speed of the thigh changes from positive to zero, indicating that the hip flexion has reached its highest point and the return phase has begun, at which point the system outputs damping.

[0071] This invention unifies and abstracts all training controls into: ,in, This indicates the target torque command. Indicates the real-time hip joint angle. This represents the real-time hip joint angular velocity. This indicates the current training phase (0 / 1 / 2). Indicates the resistance level (0–30A). Indicates the power assist level (0–30A). The training mode is indicated (hip flexion, hip extension, hip extension, swimming leg kick, automatic mode, etc.). Regardless of the training mode, the control process of this invention follows the following three-stage training structure.

[0072] Three-stage training control structure: 1. Before the start-up preparation / waiting phase, and before the motion recognition "starting conditions" are met: calibrate the zero position, maintain low or zero torque on the motor, monitor angle / speed, and use plantar pressure to determine whether the body is in the support / swing phase. Objective: To ensure natural and free movement without interfering with the body's force application.

[0073] 2. During the target direction force exertion phase (main training phase), the system detected: hip flexion: , ;Hip extension: , Hip extension: forward shift of center of gravity + small angular velocity; Squat: hip angle gradually increases, pressure increases; Swimming: no foot support, periodic back-and-forth swing, this stage outputs the main training torque.

[0074] 3. Recovery phase: After reaching the peak of the movement, the peak hip flexion is: ≤0; Peak hip extension: ≥0; Squat stand-up: >0; Swimming backswing → frontswing: Speed ​​and direction change. During this phase, control stability and avoid mechanical rebound. At the same time, record ROM, Step Energy, and number of training sessions.

[0075] In this invention, AUTO adaptive training closed-loop control is employed: This system provides an intelligent closed-loop training mode based on Step Energy and Performance Index (MPI). In this mode, the resistance is not manually adjusted by the user; instead, the system automatically adjusts it according to the user's performance, achieving truly personalized training. The core idea of ​​AUTO mode is that the system automatically adjusts resistance, assistance, and pace according to the user's ability, ensuring that the training is neither too easy nor too difficult, continuously maintaining the most effective training range.

[0076] This invention features an automatic training mode (AUTO mode) that automatically adjusts the motor's assist / resistance level based on the user's actual training performance without requiring frequent manual adjustments. This ensures the training intensity remains within a suitable range and can adapt in real-time based on indicators such as fatigue level, range of motion, and energy per step. The AUTO mode is a universal hip joint AUTO mode applicable to various exercise scenarios, including hip flexion training, hip extension training, hip extension training, squatting training, and swimming leg kick training.

[0077] State variables and parameter definitions: In AUTO mode, each leg maintains a set of training state variables, denoted as: The current resistance level is an integer, typically ranging from 0 to 30, corresponding to the motor current limiting value. The current power assist level (if using independent power assist parameters) can also be set from 0 to 30. The scope of activities for this round of training is defined as follows: , , These represent the maximum and minimum hip joint angles during this movement cycle; Step energy represents the total work done in one complete action cycle. Motion Performance Index (MPI) is a comprehensive score that considers factors such as speed, power, ROM, and symmetry. Symmetry indices, such as ROM symmetry or energy symmetry; Fatigue indicators can be derived from multiple consecutive steps. The downward trend or Step Energy is calculated based on a sustained downward trend. The system has pre-set target range parameters. , : Expected activity range; , : Expected energy range for this step; Target athletic performance index; , Resistance level upper and lower limits (e.g., 5 to 25).

[0078] In AUTO mode, each complete movement cycle (e.g., a complete hip flexion and return, or a complete squat and stand-up) is considered a "training unit." The system calculates the energy for this cycle; for details on the calculation method, please refer to the aforementioned formula for calculating energy for this step. Energy for this step It reflects the overall power level and effort exerted in this training exercise and is one of the important bases for adjusting the parameters of AUTO mode.

[0079] In a preferred embodiment, the AUTO mode initially employs ROM-based automatic adjustment logic. Let the range of motion for the current training exercise be... The automatic adjustment rule is as follows: If: This indicates that the resistance is too high or the user's strength is too weak, and the current range of motion is insufficient. The system will automatically reduce the resistance level. = - ,like > This indicates that the resistance is too low, and the user is lifting their leg or extending their hip too easily. The system will automatically increase the resistance level. = + ,in: , To adjust the resistance step size, 1 to 3 levels are generally used; to ensure safety, the final resistance level is limited. = ( ( , ), Explanation: When the user cannot lift or squat down, the ROM will be significantly smaller, and the system will automatically reduce resistance to avoid overload; when the user can easily complete large movements, the ROM will be larger, and the system will automatically increase resistance to make the training more intense.

[0080] The automatic adjustment rule driven by this step: In another preferred embodiment, the system automatically adjusts based on its own energy. Let the desired energy range be […]. , ],like This indicates that the energy output in this step is insufficient, which may be due to excessive resistance causing a reduction in movement, or it may be due to a decrease in the user's strength. At this time, the system can choose two methods: (1) In the case of insufficient strength, reduce resistance: = - (2) Emphasizing technical training scenarios, keeping the resistance constant, and only providing the prompt "increase the amplitude". If: This indicates that the current training load is high and can be considered high-intensity training; the system can: maintain the resistance during short sprint phases; and slightly reduce the resistance during long-duration training to prevent over-fatigue. = - Among them, the resistance adjustment step size based on energy feedback can generally be set to level 1. Therefore, through energy range control in this step, the system can keep the training at a reasonable intensity that is "both loaded and not overly fatigued".

[0081] In a comprehensive implementation of AUTO mode, the system considers both ROM and current-step energy metrics, adjusting them using rule priority or weighted logic. For example, the following combined strategy can be used: 1) First check... :like Significantly smaller than Prioritize reducing resistance; if Significantly greater than 1) Prioritize increasing resistance; 2) When ROM is within the normal range, check the energy of this step. Is it stable at [ , If the energy deviates significantly from the target range, fine-tuning is performed. A comprehensive deviation function is defined: ,in: = ; = ,in , These are the weighting coefficients. They determine... Weighting with energy. When the overall deviation As the resistance continues to increase, the system automatically adjusts the resistance level to reduce it. Thus achieving "with Automatic closed-loop control that is "primarily based on energy and secondarily based on power".

[0082] The AUTO mode can adjust the training phase based on the MPI score, for example: when the MPI is low (e.g. < The system automatically enters the "basic strength training phase," where resistance is relatively low, prioritizing ROM; when When in the middle range ( The system increases resistance, entering the "strength + explosive power training phase"; when At higher levels ( The system focuses on optimizing symmetry and smoothness, appropriately reducing resistance and increasing frequency, and enters the "advanced technology training stage".

[0083] In some optional embodiments of the present invention, the swimming leg kick AUTO mode is an important special case of the present invention in hip joint training, mainly targeting freestyle leg kick training. Key indicators: leg kick frequency and leg kick energy. Two core indicators are introduced in freestyle leg kick training: 1. Leg kick frequency ( ): ;in: : The time of one complete leg kick cycle (unit: seconds), including "hip flexion and upward swing + hip extension and downward swing"; constant 60 is the number of seconds per minute, used to convert "per cycle movement" into "number of leg kicks per minute"; 2. Kick Energy: Calculated for each kick cycle: , in: This indicates the instantaneous torque of the hip joint drive motor in the leg-thrusting direction; This indicates the angular velocity of the thigh around the hip joint; , Indicates the start and end times of one leg-kicking cycle. Meaning: It reflects the "pace"; KickEnergy reflects the "energy output level of a single kick".

[0084] In swimming leg kick training, the system presets a target range: target leg kick frequency range: For example: 80-140 times / minute (amateur) or 150-200 times / minute (professional); Target leg kick energy range: During training, users can choose between "Pace Priority" mode → "Strength Priority" mode → "Comprehensive Training" mode → a combination of both.

[0085] Automatic resistance / assistance adjustment logic (for swimming).

[0086] Taking resistance training as an example, let the current leg kick frequency be KickRate, and the energy be... The current resistance level is In AUTO mode, the system can employ the following logic: 1. Rhythm Maintenance Logic: If: ,and It has approached or exceeded This indicates that the user may be too fatigued and unable to maintain the rhythm. In this case, the resistance can be appropriately reduced. ,like: ,and The resistance is too low, indicating that the user is kicking too fast but lacks power. The resistance can be increased appropriately to make the kicking feel more realistic. ; 2. Leg kick energy maintenance logic: If: If the rhythm is normal, the system can prompt the user to "increase effort" or slightly reduce resistance; if: Furthermore, over multiple cycles, the system can reduce resistance to prevent excessive training load. Specifically: The resistance fine-tuning step size based on swimming indicators is generally 1 level.

[0087] Assisted training in swimming AUTO mode.

[0088] When used for sports rehabilitation or for beginners practicing leg kicks, the assisted training mode can be activated. In this mode, the system provides phase-synchronized directional assistance to the user based on information such as leg kick phase and kick rate. The AUTO mode synchronizes the assistance amplitude with the user's rhythm and, to some extent, guides them to reach the target frequency.

[0089] For example, a boost gain coefficient can be defined: ,in: The frequency of kicks to the target This is the gain coefficient for frequency error. When the user's leg kick is too slow, the assistance is slightly increased to guide them to a faster pace; when the user's pace has reached the target, the assistance is gradually reduced, allowing them to rely more on their own muscles to complete the leg kick, which is beneficial for training effectiveness and progress evaluation.

[0090] In the description of this invention, it should be understood that the terms "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0091] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0092] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0093] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0094] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

[0095] In the description of this specification, the references to terms such as "Embodiment 1," "Embodiment 2," "Example," "Specific Example," or "Some Examples," etc., indicate that the specific method, apparatus, or feature described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, methods, apparatus, or features described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0096] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control method for an intelligent exoskeleton device used in hip joint sports training, characterized in that, The intelligent exoskeleton device includes a hip area main load-bearing structure, a thigh side support structure, and a hip joint rotation mechanism connecting the two. The hip joint rotation mechanism integrates a power output device. An inertial measurement unit is fixed in the thigh side support structure. The inertial measurement unit is used to acquire real-time hip joint angle and angular velocity. Two plantar pressure sensors are installed on the sole of the foot to acquire heel pressure and toe pressure. The main control module is used to receive data from the inertial measurement unit, plantar pressure sensors, and power output device. The main control module is connected to the host computer for communication. The control method includes the following steps: Step S100: The user wears the smart exoskeleton device and presets multiple training modes in the host computer interface; Step S200: The user selects the region through the mode selection on the host computer, clicks the corresponding function button, and sends the training mode command to the lower-level controller. Step S300: After receiving the corresponding instruction, the controller sets the internal training state structure to the corresponding mode. Step S400: Initialize the sensor; Step S500: Apply resistance parameters and load resistance parameters; Step S600: Enable the action recognition algorithm; Step S700: Perform output control of the motor torque of the power output device; Step S800: Record and evaluate the training indicators.

2. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 1, characterized in that, The training modes set in step S100 include, but are not limited to, hip flexion training mode, hip extension training mode, hip extension training mode, high knee training mode, squat training mode, and freestyle leg kick training mode.

3. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 2, characterized in that, Step S400 specifically involves zero-position calibration of the thigh inertial measurement unit and static calibration of the plantar pressure sensor to obtain the hip joint flexion-extension angle, static heel pressure, and static toe pressure as initial reference values.

4. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 3, characterized in that, Step S500 specifically involves reading the basic resistance level, assistance coefficient, target hip joint range of motion, target frequency, and target interval training parameters for this step's energy from the preset parameter table according to the selected mode.

5. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 4, characterized in that, Step S600 specifically involves calling the motion recognition algorithm module corresponding to the mode to perform real-time analysis of the inertial measurement unit's angle, angular velocity, and plantar pressure signals to determine the current phase.

6. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 5, characterized in that, Specifically, step S700 involves outputting directional resistance or assist torque according to the mode type when a valid training phase is identified; and reducing the motor torque to zero or setting it to a safe value when a non-training phase or abnormal state is detected.

7. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 6, characterized in that, Step S800 specifically involves calculating the energy, motion performance index, motion quality index, and left-right symmetry index for each complete movement at the end of the movement, and displaying them in real time on the host computer.

8. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 7, characterized in that, Energy in step S800 The calculation formula is: In this formula, This represents the energy produced in a single action. Indicates the starting point of a complete action. Indicates the end time of a complete action. Indicates instantaneous power. Represents the integral over time; Instantaneous power The calculation formula is: In this formula, This represents the real-time torque that the motor outputs or acts in the opposite direction on the user's hip joint at time t. This represents the angular velocity of the user's hip joint at time t; The formula for calculating the Performance Index (MPI) is as follows: In this formula, This represents the peak angular velocity of the hip joint during one movement cycle. This represents the peak power of the hip joint, and ROM represents the range of motion of the hip joint during the middle of this training session. Sym The index represents left-right symmetry, and the index represents smoothness. ~ Indicates the weight of each indicator; The formula for calculating the range of motion (ROM) of the hip joint is: In this formula, This indicates the maximum hip angle during this movement. This indicates the minimum hip angle for this movement.

9. The control method for the intelligent exoskeleton device for hip joint sports training according to claim 8, characterized in that, It also includes step S900, where when the automatic training mode or intelligent adjustment option is selected, the system automatically adjusts the resistance level, resistance magnitude, and training rhythm based on the changes in energy, hip joint range of motion, and athletic performance index of the most recent steps, so as to keep the training intensity within the target range.

10. A control system for an intelligent exoskeleton device for hip joint sports training, characterized in that, The system is used to perform a control method for an intelligent exoskeleton device for hip joint sports training as described in any one of claims 1 to 9.

Images