Upper and lower limb gait synchronization method and system for exoskeleton

Abstract

The invention discloses an upper and lower limb gait synchronization method and system for an exoskeleton, and belongs to the field of exoskeleton control, and the method comprises the steps: analyzing the data of an exoskeleton joint sensor, recognizing a gait cycle and a key event, and obtaining a gait recognition result and a key event recognition result; according to human kinematics characteristics, a mathematical model of shoulder joint angle change is established, and a shoulder joint motion reference curve matched with a gait recognition result is generated; the gait recognition result, the key event recognition result and the shoulder joint motion reference curve are combined, a control instruction of an exoskeleton shoulder joint motor is generated, and motion synchronization of the upper limb and the lower limb is achieved. The technical problem that in existing exoskeleton walking rehabilitation training, upper limb movement is neglected, and consequently the gait recovery effect is poor is solved, through sensor data analysis, upper limb shoulder joint movement and lower limb gait are synchronously controlled, exoskeleton upper limb and lower limb movement is coordinated, and the technical effect of improving the rehabilitation training effect is achieved.

Description

Technical Field

[0001] The present invention relates to the field of exoskeleton control, and in particular to an upper and lower limb gait synchronization method and system for an exoskeleton. Background Art

[0002] In the field of rehabilitation, lower limb exoskeleton devices have been widely used to help patients with lower limb motor dysfunction to perform walking training and rehabilitation. Traditional lower limb exoskeletons mainly focus on the assistance and control of lower limb movements. By driving the hip, knee, ankle and other joints, the patient's lower limbs can produce movements that simulate normal gait. However, the existing technology often ignores the important role of the upper limbs, especially the shoulder movements, in gait recovery. When people walk normally, the upper limbs and lower limbs are coordinated. The natural swing of the upper limbs not only helps to maintain body balance, but also promotes the symmetry and coordination of the lower limb gait. Therefore, in walking rehabilitation training, only focusing on the lower limbs and ignoring the movement of the upper limbs will cause the recovered gait to lack natural coordination, resulting in poor gait recovery effect. Summary of the invention

[0003] The present application provides a method and system for synchronizing upper and lower limb gait for an exoskeleton, aiming to solve the technical problem of neglecting upper limb movement in existing exoskeleton walking rehabilitation training, resulting in poor gait recovery effect.

[0004] In view of the above problems, the present application provides a method and system for synchronizing upper and lower limb gait of an exoskeleton.

[0005] The first aspect disclosed in the present application provides a method for synchronizing the gait of upper and lower limbs for an exoskeleton, the method comprising: analyzing exoskeleton joint sensor data to identify gait cycles and key events, and obtaining gait recognition results and key event recognition results; establishing a mathematical model of shoulder joint angle changes based on human kinematic characteristics, and generating a shoulder joint motion reference curve that matches the gait recognition result; combining the gait recognition result, the key event recognition result, and the shoulder joint motion reference curve to generate control instructions for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

[0006] Another aspect disclosed in the present application provides an upper and lower limb gait synchronization system for an exoskeleton, the system comprising: a gait recognition module, for analyzing exoskeleton joint sensor data, identifying gait cycles and key events, and obtaining gait recognition results and key event recognition results; a curve generation module, for establishing a mathematical model of shoulder joint angle changes based on human kinematic characteristics, and generating a shoulder joint motion reference curve that matches the gait recognition result; and a synchronization control module, for combining the gait recognition result, the key event recognition result and the shoulder joint motion reference curve, and generating control instructions for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] By analyzing the exoskeleton joint sensor data, identifying the gait cycle and key events, and obtaining the gait recognition results and key event recognition results, the current gait cycle and key events in the gait, such as the heel-landing and toe-off moments, are obtained, providing a basis for the subsequent generation of the shoulder joint motion reference curve and motion synchronization control; according to the human kinematic characteristics, a mathematical model of the shoulder joint angle change is established to generate a shoulder joint motion reference curve that matches the gait recognition results, providing a theoretical basis and data support for achieving coordinated movement of the upper and lower limbs; the gait recognition results, the key event recognition results and the shoulder joint motion reference curve are combined to generate control instructions for the exoskeleton shoulder joint motor, and a technical solution is achieved to achieve synchronization of upper and lower limb movements, which solves the technical problem of ignoring upper limb movement in existing exoskeleton walking rehabilitation training, resulting in poor gait recovery effects. Through sensor data analysis, the upper limb shoulder joint movement and the lower limb gait are synchronously controlled to coordinate the exoskeleton upper and lower limb movements, achieving the technical effect of improving the rehabilitation training effect.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a method for synchronizing upper and lower limb gait of an exoskeleton is provided for an embodiment of the present application;

[0011] Figure 2 is a schematic diagram of an asymmetric structure simple pendulum model in an embodiment of the present invention;

[0012] Figure 3 A structural schematic diagram of an upper and lower limb gait synchronization system for an exoskeleton is provided for an embodiment of the present application.

[0013] Description of reference numerals: gait recognition module 11 , curve generation module 12 , synchronization control module 13 . DETAILED DESCRIPTION

[0014] The overall idea of ​​the technical solution provided by this application is as follows:

[0015] The embodiment of the present application provides a method and system for gait synchronization of upper and lower limbs for an exoskeleton. First, the exoskeleton joint sensor data is used to analyze it, and the gait cycle and key events are identified, providing a reliable basis for the subsequent shoulder joint motion control. On this basis, according to the kinematic characteristics of the human body, a mathematical model of shoulder joint angle change is established, and a shoulder joint motion reference curve matching the gait is generated, which lays the foundation for achieving natural and coordinated synchronous movement of upper and lower limbs. Afterwards, the gait recognition results, the key event recognition results and the shoulder joint motion reference curve are fused to generate an exoskeleton shoulder joint motor control instruction, so as to achieve synchronization of the upper and lower limb movements of the exoskeleton, so as to achieve a coordinated and natural gait of the upper and lower limbs, and effectively improve the rehabilitation training effect.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0017] Embodiment 1, as Figure 1 As shown, an embodiment of the present application provides a method for synchronizing upper and lower limb gait for an exoskeleton, the method comprising:

[0018] S1: By analyzing the exoskeleton joint sensor data, the gait cycle and key events are identified to obtain the gait recognition results and key event recognition results.

[0019] Specifically, the exoskeleton device is equipped with multiple joint sensors for real-time collection of motion data of the wearer's lower limb joints. The joint sensors include but are not limited to angle sensors, angular velocity sensors, acceleration sensors, etc., which are respectively set at the hip joint, knee joint and ankle joint of the exoskeleton to measure the angle, angular velocity and acceleration of the corresponding joints.

[0020] After acquiring the exoskeleton joint sensor data, the data is preprocessed, including denoising, filtering, normalization and other operations to improve data quality. Then, the preprocessed data is analyzed using pattern recognition, machine learning and other algorithms to identify the wearer's current gait cycle stage, such as the foothold phase, swing phase, etc., to obtain gait recognition results. At the same time, by analyzing the characteristic changes in the joint sensor data, such as sudden changes or peaks in angles and angular velocities, key events in the gait cycle, such as heel landing and toe lift-off, are further identified to obtain key event recognition results.

[0021] The acquisition of gait recognition results and key event recognition results provides important time and phase basis for subsequent shoulder joint movement control, and provides a foundation for achieving synchronous movement of shoulder joint and lower limb gait.

[0022] S2: According to the kinematic characteristics of the human body, a mathematical model of shoulder joint angle changes is established to generate a shoulder joint motion reference curve that matches the gait recognition result.

[0023] Specifically, based on the knowledge of human anatomy and kinematics, it is found that there is a certain correspondence between the change of shoulder joint angle and the gait cycle under natural gait. By analyzing the law of shoulder joint angle change when healthy people walk, a corresponding mathematical model is established to describe this correspondence. In the modeling process, the biological characteristics of shoulder joint movement are fully considered. Among them, the shoulder joint presents a periodic change law similar to the sine function during the gait cycle, and there are certain differences in speed and amplitude between the forward and backward swings, showing asymmetry. Therefore, the sine function can be used to fit and describe the shoulder joint angle change curve. The specific form and parameters of the model are obtained by analyzing a large amount of healthy people's gait data and machine learning, in order to maximize the approximation of the real physiological characteristics. After establishing the mathematical model of shoulder joint angle change, combined with the gait recognition results, a reference curve of shoulder joint movement matching the current gait is generated, which provides a clear goal and path for the subsequent shoulder joint movement control.

[0024] S3: Combining the gait recognition result, the key event recognition result and the shoulder joint motion reference curve to generate control instructions for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

[0025] Specifically, first, the gait recognition result is aligned with the shoulder joint motion reference curve in time. By normalizing the gait cycle and matching the key events in the key event recognition result, such as heel strike and toe lift, with specific time points on the shoulder joint motion reference curve, a time mapping relationship between the two is established. Then, according to the key event recognition result, the target angle value at the corresponding moment is extracted on the shoulder joint motion reference curve. For example, when the heel strike event is recognized, the corresponding time point is found on the shoulder joint motion reference curve, and the shoulder joint target angle at that point is read. Next, the current angle of the exoskeleton shoulder joint motor is measured in real time, and it is compared with the target angle to calculate the deviation between the two. According to the size and direction of the deviation value, the corresponding control instruction is generated to adjust the motor speed and direction so that it reaches the target angle as soon as possible. By continuously executing the above process, that is, measuring the current angle, calculating the target angle, generating control instructions, and continuously updating the target value according to the gait recognition results and key event recognition results, the dynamic synchronization of the exoskeleton shoulder joint movement and the lower limb gait is achieved.

[0026] Furthermore, the embodiment of the present application also includes:

[0027] Sensors installed at the hip, knee, and ankle joints of the exoskeleton collect angle data of each joint in real time;

[0028] Filtering the angle data to remove noise interference;

[0029] Compare the filtered angle data with the preset standard gait template to identify the current gait cycle and obtain the gait recognition result;

[0030] By analyzing the characteristic points of the angle data, key events in the gait recognition result are detected to obtain key event recognition results, wherein the key event recognition results include heel strike and toe lift.

[0031] In a preferred embodiment, when obtaining gait recognition results and key event recognition results, first, multiple joint sensors are set at the hip joint, knee joint and ankle joint of the exoskeleton to collect the angle data of each joint in real time. Wherein, the joint sensor can be an angle sensor such as a potentiometer, an encoder, a Hall sensor, etc., which can convert the rotation angle of the joint into an electrical signal and transmit it to the data acquisition module by a wired or wireless method. Because in the actual application environment, the angle data collected by the sensor is often mixed with various noises, such as electromagnetic interference, mechanical vibration, etc., which can affect the accuracy of data analysis. Therefore, the original angle data is filtered, such as low-pass filtering, median filtering, Kalman filtering, etc., according to the characteristics of the noise and the real-time requirements of the system. By filtering, high-frequency noise and mutation interference are removed to obtain smooth and stable angle data.

[0032] After obtaining the filtered angle data, it is compared with the preset standard gait template to identify the current gait cycle. The preset standard gait template is an angle change curve summarized based on a large amount of healthy people's walking data, which reflects the angle change law of the hip, knee and ankle joints in a typical gait cycle. By calculating the similarity between the filtered angle data and the standard gait template, the current gait stage, such as the foothold period, the swing period, etc., is judged, thereby obtaining the gait recognition result. In addition to identifying the gait cycle, the filtered angle data is further analyzed to detect the feature points therein to identify key events. Specifically, the focus is on the inflection points, peaks and valleys in the angle data curve, which often correspond to key events in the gait. For example, when the curve has a steep inflection point, it usually indicates the occurrence of the heel landing event; and when the curve has a steep inflection point, it indicates the occurrence of the toe off the ground event. By setting the threshold and judgment rules, these feature points are automatically detected, and then the key event recognition results are obtained. Among them, the key event recognition results include the two events of heel landing and toe off the ground.

[0033] Through comprehensive analysis of exoskeleton joint sensor data, the recognition results of gait cycle and key events are obtained, providing a reliable basis for the subsequent generation of shoulder joint motion reference curves and synchronous control.

[0034] Furthermore, the mathematical model of the shoulder joint angle change is a sine function model, including:

[0035] According to the walking data of healthy people, the sine function expression of the shoulder joint angle change was fitted;

[0036] According to the gait recognition result, the function parameters of the sine function expression are determined to generate a shoulder joint angle reference curve.

[0037] In a preferred embodiment, by analyzing the walking data of multiple healthy subjects, it is found that the change in the angle of the shoulder joint shows obvious periodicity. In a complete gait cycle, the shoulder joint angle approximates the change law of the sine function. Therefore, the sine function is selected to fit and express the shoulder joint angle change curve.

[0038] In order to determine the parameters in the sine function model, the walking data of healthy subjects were collected and analyzed. For example, a high-precision inertial measurement unit was used to record the changes in the shoulder joint angle of the subjects at different walking speeds to obtain a large amount of real data. Then, these data were preprocessed, including denoising, smoothing, segmentation, etc., to improve the data quality. Then, mathematical tools such as the least squares method were used to fit the preprocessed data with a sine function to obtain the sine function expression.

[0039] In practical applications, according to the gait recognition results, the corresponding function parameters are selected to generate a shoulder joint angle reference curve that matches the current gait. Specifically, the gait cycle is normalized, and then the percentage position of the key event (such as heel landing) in the cycle is determined according to the gait recognition results. According to the percentage position, the theoretical value of the shoulder joint angle at the corresponding moment is obtained from the sine function expression as the control point of the shoulder joint angle change curve. Then, through interpolation and other methods, a continuous shoulder joint angle reference curve is generated to reflect the change law of the shoulder joint angle synchronized with the gait, which serves as the goal of exoskeleton motion control.

[0040] By using a sinusoidal function model and adjusting parameters according to the gait recognition results, a shoulder joint motion reference curve that matches the gait and conforms to physiological laws is generated, providing an ideal target for the motion control of the exoskeleton and helping to achieve natural and coordinated gait synchronization between the upper and lower limbs.

[0041] Furthermore, the sine function expression is:

[0042]

[0043] Among them, A shoulder is the maximum swing angle of the shoulder joint, ω shoulder is the swing frequency of the shoulder joint, is the initial phase of the shoulder joint.

[0044] In a preferred embodiment, the sinusoidal function expression of the shoulder joint angle change is given as:

[0045]

[0046] Among them, A shoulder A represents the maximum swing angle of the shoulder joint, that is, the swing amplitude of the shoulder joint in a gait cycle, which reflects the strength of the upper limb swing and is closely related to the speed and stability of walking. Generally speaking, the faster the walking speed, the greater the swing angle of the shoulder joint. shoulder The larger the value of ; on the contrary, the slower the walking speed, the shoulder The smaller the value of ω shoulder The oscillation frequency of the shoulder joint, that is, the speed of the shoulder joint swinging in a gait cycle, is closely related to the duration of the gait cycle. The shorter the gait cycle, the higher the ω shoulder The larger the value of ; on the contrary, the longer the gait cycle, ω shoulder The smaller the value of . It represents the initial phase of the shoulder joint, that is, the starting angle of the shoulder joint swing at the beginning of a gait cycle, reflecting the phase relationship between the upper limb swing and the lower limb gait.

[0047] By adjusting the above three function parameters, we can get shoulder joint angle change curves of different shapes and styles. In practical applications, according to the gait recognition results, combined with the wearer's physiological characteristics and rehabilitation needs, the function parameters in the sine function expression are dynamically adjusted to generate a personalized shoulder joint angle reference curve, thereby guiding the exoskeleton's motion control, helping the wearer to restore a natural, comfortable, and efficient gait pattern, and improving the rehabilitation effect.

[0048] Furthermore, the mathematical model of the shoulder joint angle change is an asymmetric pendulum model, including:

[0049] Establish a simple pendulum model of asymmetric structure;

[0050] Based on the asymmetric structure single pendulum model, determining model parameters;

[0051] According to the asymmetric structure pendulum model and the model parameters, an asymmetric pendulum mathematical model is established to derive the single-arm swing period;

[0052] The single-arm swing cycle is matched with the gait recognition result to generate the shoulder joint angle reference curve.

[0053] In another preferred embodiment, during a person's walking process, the swing of the upper limbs exhibits a certain asymmetry in the forward and backward directions, that is, the amplitude and speed of the forward swing are often greater than those of the backward swing. Based on this asymmetry, a shoulder joint angle reference curve that is more in line with physiological characteristics is established.

[0054] In order to accurately describe the asymmetry of upper limb swing, firstly, an asymmetric single pendulum model is established. Figure 2 As shown in the figure, the asymmetric structure single pendulum model includes a pendulum ball and two fixed fulcrums. The two fulcrums are spaced a certain distance apart in the vertical direction, denoted as O1 and O2 respectively. During the swinging process of the pendulum ball, a motion similar to a single pendulum is formed. Different from the traditional single pendulum model, the asymmetric structure single pendulum model introduces two different pendulum lengths and swing angles. When the pendulum ball swings forward, its pendulum length is L1 and its swing angle is θ1; when the pendulum ball swings backward, its pendulum length is L2 and its swing angle is θ2. After establishing the asymmetric structure single pendulum model, the model in the model is determined. By setting the values ​​of L1, L2, θ1 and θ2, the asymmetry of forward and backward swings can be simulated, which is more in line with the actual situation of human upper limb swing. In order to obtain reasonable values ​​of these parameters, gait data of multiple healthy subjects were collected and analyzed. By measuring the trajectory and speed of the upper limb swing of the subjects and combining anthropometric data, the statistical distribution of parameters such as L1, L2, θ1 and θ2 was obtained, and the model parameters were adjusted according to specific users to obtain model parameters.

[0055] Subsequently, based on the determined asymmetric pendulum model and model parameters, an asymmetric pendulum mathematical model is further established to describe the motion law under the asymmetric structure, and the single-arm swing cycle is derived to reflect the time required for a complete reciprocating motion. Afterwards, the derived single-arm swing cycle is matched with the gait recognition result to generate a shoulder joint angle reference curve corresponding to the gait. Since the asymmetric pendulum model fully considers the asymmetric characteristics of the upper limb swing, the generated shoulder joint angle reference curve also shows obvious asymmetry. Compared with the reference curve based on the sine function, it can more realistically describe the real physiological motion process and provide a more accurate and natural target trajectory for the motion control of the exoskeleton.

[0056] Furthermore, the single-arm swing period is:

[0057]

[0058] Where T is the swing period of a single arm, L1 is the length of the first arm when it swings forward in the asymmetric pendulum model, g is the acceleration of gravity, and θ 1max is the maximum angle of the first arm swinging forward in the asymmetric structure single pendulum model, R is the vertical distance between the left and right swing centers in the asymmetric structure single pendulum model, L1-R is the pendulum length when the second arm swings backward in the asymmetric structure single pendulum model, θ2max It is the maximum backward swing angle of the second arm in the asymmetric simple pendulum model.

[0059] In a preferred embodiment, the mathematical expression of the single-arm swing period under the asymmetric pendulum model is given as:

[0060]

[0061] In the asymmetric pendulum model, the motion process of the asymmetric pendulum is divided into two parts, left and right, with the center line as the boundary. At any time, the swing angles on the left and right sides are θ1 and θ2 respectively. When the swing angle is large, the motion of the asymmetric pendulum is no longer a simple harmonic vibration, but a nonlinear vibration. Under nonlinear vibration conditions, due to the existence of two different pendulum lengths and swing angles, the single-arm swing period T consists of two parts: the time T1 for the first arm to swing forward and the time T2 for the second arm to swing backward. According to the motion characteristics of the pendulum, the expressions of T1 and T2 are obtained, respectively:

[0062]

[0063]

[0064] Among them, for the first arm swinging forward, its pendulum length is L1 and the maximum swing angle is θ 1max For the second arm to swing backward, its swing length is L2 and the maximum swing angle is θ 2max θ1 and θ2 are the swing angles on the left and right sides.

[0065] According to the above expressions of T1 and T2, the approximate motion times of the asymmetric pendulum on both sides are:

[0066]

[0067] From the asymmetric structure pendulum model, we know that L2=L1―R, so the above formula can be transformed into:

[0068]

[0069] This expression reflects the time required for a single arm to swing a complete cycle under an asymmetric structure. By deriving the single arm swing cycle under the asymmetric pendulum model and combining specific model parameters, the time characteristics of the single arm swing motion are quantitatively described. Combining this cycle information with the gait recognition results, a shoulder joint angle reference curve that matches the gait can be generated, providing an important time benchmark for the coordinated control of the upper and lower limbs of the exoskeleton.

[0070] Furthermore, the embodiment of the present application also includes:

[0071] Mapping the time points of the key events in the key event recognition results to the shoulder joint angle reference curve to obtain a target angle value;

[0072] Acquire the current angle value of the exoskeleton shoulder joint motor in real time, and calculate the angle deviation between the current angle value and the target angle value;

[0073] Based on the angle deviation, a control instruction for the exoskeleton shoulder joint motor is generated.

[0074] In a feasible implementation, first, extract the key events in the key event recognition results, such as heel landing, toe off the ground, etc., and determine the corresponding time points, which are matched with the specific time points on the shoulder joint angle reference curve. Through this mapping, the target angle value corresponding to each key event is found on the shoulder joint motion reference curve, so as to establish the corresponding relationship between the gait key event and the shoulder joint target angle. Secondly, obtain the current angle value of the exoskeleton shoulder joint motor in real time, and calculate the angle deviation between the current angle value and the target angle value. By measuring the actual angle of the exoskeleton shoulder joint motor in real time and comparing it with the obtained target angle value, the angle deviation between the two is calculated, which reflects the difference between the actual shoulder joint motion and the ideal motion. Afterwards, according to the calculated angle deviation, the control instructions of the exoskeleton shoulder joint motor are generated, including specific parameters such as motor speed and steering, aiming to adjust the motion state of the shoulder joint so that it gradually approaches the target angle, thereby keeping synchronization with the lower limb gait.

[0075] By considering the gait recognition results and key event recognition results, combined with the shoulder joint motion reference curve, the exoskeleton upper limb movement and lower limb gait are coordinated, achieving more precise and personalized shoulder joint control, adapting to the user's actual gait changes, and facilitating rehabilitation training.

[0076] In summary, the upper and lower limb gait synchronization method for an exoskeleton provided in the embodiments of the present application has the following technical effects:

[0077] By analyzing the exoskeleton joint sensor data, identifying the gait cycle and key events, and obtaining the gait recognition results and key event recognition results, the necessary time and phase basis is provided for the subsequent generation of the shoulder joint motion reference curve and motion synchronization control, which is the basis for achieving coordinated movement of the upper and lower limbs. According to the kinematic characteristics of the human body, a mathematical model of the shoulder joint angle change is established, and a shoulder joint motion reference curve matching the gait recognition result is generated, which reflects the target angle that the shoulder joint should reach at different times, and provides a basis and support for achieving natural and coordinated synchronous movement of the upper and lower limbs. The gait recognition results, the key event recognition results and the shoulder joint motion reference curve are combined to generate control instructions for the exoskeleton shoulder joint motor, so that the upper limb movement is coordinated with the lower limb gait in terms of time and amplitude, and the synchronization of the upper and lower limb movements of the exoskeleton is achieved, thereby improving the rehabilitation training effect.

[0078] Embodiment 2 is based on the same inventive concept as the method for synchronizing upper and lower limb gait for an exoskeleton in the aforementioned embodiment. Figure 3 As shown, an embodiment of the present application provides an upper and lower limb gait synchronization system for an exoskeleton, the system comprising:

[0079] A gait recognition module 11 is used to analyze the exoskeleton joint sensor data, identify the gait cycle and key events, and obtain gait recognition results and key event recognition results;

[0080] A curve generation module 12 is used to establish a mathematical model of shoulder joint angle changes according to human kinematic characteristics, and generate a shoulder joint motion reference curve that matches the gait recognition result;

[0081] The synchronization control module 13 is used to combine the gait recognition result, the key event recognition result and the shoulder joint motion reference curve to generate a control instruction for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

[0082] Furthermore, the gait recognition module 11 includes the following execution steps:

[0083] Sensors installed at the hip, knee, and ankle joints of the exoskeleton collect angle data of each joint in real time;

[0084] Filtering the angle data to remove noise interference;

[0085] Compare the filtered angle data with the preset standard gait template to identify the current gait cycle and obtain the gait recognition result;

[0086] By analyzing the characteristic points of the angle data, key events in the gait recognition result are detected to obtain key event recognition results, wherein the key event recognition results include heel strike and toe lift.

[0087] Furthermore, the curve generating module 12 includes the following execution steps:

[0088] According to the walking data of healthy people, the sine function expression of the shoulder joint angle change was fitted;

[0089] According to the gait recognition result, the function parameters of the sine function expression are determined to generate a shoulder joint angle reference curve.

[0090] Furthermore, the curve generating module 12 also includes the following contents:

[0091] The sine function expression is:

[0092]

[0093] Among them, A shoulder is the maximum swing angle of the shoulder joint, ω shoulder is the swing frequency of the shoulder joint, is the initial phase of the shoulder joint.

[0094] Furthermore, the curve generating module 12 includes the following execution steps:

[0095] Establish a simple pendulum model of asymmetric structure;

[0096] Based on the asymmetric structure single pendulum model, determining model parameters;

[0097] According to the asymmetric structure pendulum model and the model parameters, an asymmetric pendulum mathematical model is established to derive the single-arm swing period;

[0098] The single-arm swing cycle is matched with the gait recognition result to generate the shoulder joint angle reference curve.

[0099] Furthermore, the curve generation module 12 includes the following contents:

[0100] The single-arm swing period is:

[0101]

[0102] Where T is the swing period of a single arm, L1 is the length of the first arm when it swings forward in the asymmetric pendulum model, g is the acceleration of gravity, and θ 1max is the maximum angle of the first arm swinging forward in the asymmetric structure single pendulum model, R is the vertical distance between the left and right swing centers in the asymmetric structure single pendulum model, L1-R is the pendulum length when the second arm swings backward in the asymmetric structure single pendulum model, θ 2max It is the maximum backward swing angle of the second arm in the asymmetric simple pendulum model.

[0103] Furthermore, the synchronization control module 13 includes the following execution steps:

[0104] Mapping the time points of the key events in the key event recognition results to the shoulder joint angle reference curve to obtain a target angle value;

[0105] Acquire the current angle value of the exoskeleton shoulder joint motor in real time, and calculate the angle deviation between the current angle value and the target angle value;

[0106] Based on the angle deviation, a control instruction for the exoskeleton shoulder joint motor is generated.

[0107] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0108] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to the selection of multiple elements individually or in whole. Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these changes and variations.

Claims

1. A method for synchronizing upper and lower limb gait for an exoskeleton, characterized in that: include: By analyzing the exoskeleton joint sensor data, the gait cycle and key events are identified, and the gait recognition results and key event recognition results are obtained; According to the kinematic characteristics of the human body, a mathematical model of shoulder joint angle change is established to generate a shoulder joint motion reference curve matching the gait recognition result; The gait recognition result, the key event recognition result and the shoulder joint motion reference curve are combined to generate control instructions for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

2. A method for synchronizing upper and lower limb gait for an exoskeleton according to claim 1, characterized in that: By analyzing the exoskeleton joint sensor data, the gait cycle and key events are identified, and the gait recognition results and key event recognition results are obtained, including: Sensors installed at the hip, knee, and ankle joints of the exoskeleton collect angle data of each joint in real time; Filtering the angle data to remove noise interference; Compare the filtered angle data with the preset standard gait template to identify the current gait cycle and obtain the gait recognition result; By analyzing the characteristic points of the angle data, key events in the gait recognition result are detected to obtain key event recognition results, wherein the key event recognition results include heel strike and toe lift.

3. The method for synchronizing upper and lower limb gait for an exoskeleton according to claim 1, characterized in that: The mathematical model of shoulder joint angle change is a sinusoidal function model, including: According to the walking data of healthy people, the sine function expression of the shoulder joint angle change was fitted; According to the gait recognition result, the function parameters of the sine function expression are determined to generate a shoulder joint angle reference curve.

4. A method for synchronizing upper and lower limb gait for an exoskeleton according to claim 3, characterized in that: The sine function expression is: Among them, A shoulder is the maximum swing angle of the shoulder joint, ω shoulder is the swing frequency of the shoulder joint, is the initial phase of the shoulder joint.

5. The method for synchronizing upper and lower limb gait for an exoskeleton according to claim 1, characterized in that: The mathematical model of shoulder joint angle change is an asymmetric pendulum model, including: Establish a simple pendulum model of asymmetric structure; Based on the asymmetric structure single pendulum model, determining model parameters; According to the asymmetric structure pendulum model and the model parameters, an asymmetric pendulum mathematical model is established to derive the single-arm swing period; The single-arm swing cycle is matched with the gait recognition result to generate the shoulder joint angle reference curve.

6. A method for synchronizing upper and lower limb gait for an exoskeleton according to claim 5, characterized in that: The single-arm swing period is: Where T is the swing period of a single arm, L1 is the length of the first arm when it swings forward in the asymmetric pendulum model, g is the acceleration of gravity, and θ 1max is the maximum angle of the first arm swinging forward in the asymmetric structure single pendulum model, R is the vertical distance between the left and right swing centers in the asymmetric structure single pendulum model, L1-R is the pendulum length when the second arm swings backward in the asymmetric structure single pendulum model, θ 2max It is the maximum backward swing angle of the second arm in the asymmetric simple pendulum model.

7. The method for synchronizing upper and lower limb gait for an exoskeleton according to claim 1, characterized in that: Combining the gait recognition result, the key event recognition result and the shoulder joint motion reference curve to generate a control instruction for an exoskeleton shoulder joint motor, including: Mapping the time points of the key events in the key event recognition results to the shoulder joint angle reference curve to obtain a target angle value; Acquire the current angle value of the exoskeleton shoulder joint motor in real time, and calculate the angle deviation between the current angle value and the target angle value; Based on the angle deviation, a control instruction for the exoskeleton shoulder joint motor is generated.

8. A method for synchronizing upper and lower limb gait for an exoskeleton, characterized in that: A method for synchronizing upper and lower limb gait for an exoskeleton for implementing any one of claims 1 to 7, comprising: A gait recognition module, which is used to analyze the exoskeleton joint sensor data, identify the gait cycle and key events, and obtain gait recognition results and key event recognition results; A curve generation module, wherein the curve generation module is used to establish a mathematical model of shoulder joint angle changes according to human kinematic characteristics, and generate a shoulder joint motion reference curve that matches the gait recognition result; A synchronization control module is used to combine the gait recognition result, the key event recognition result and the shoulder joint motion reference curve to generate control instructions for the exoskeleton shoulder joint motor to achieve synchronization of upper and lower limb movements.

Images