Methods for real-time adjustment of gait training parameters

Abstract

A method for real-time adjustment of gait training parameters includes the following steps: (a) collecting muscle-relaxed gait data of a first user in a muscle-relaxed state and active-force gait data of a first user in an active-force state, and establishing a standard motion model by the ratio of the first user's active-force gait data to the first user's muscle-relaxed gait data; (b) obtaining a motion model of a second user, including the second user's muscle-relaxed gait data in a muscle-relaxed state, and predicting a personalized training model by combining the second user's muscle-relaxed gait data with the standard motion model; (c) determining whether the second user's actual training meets the standard of the personalized training model, thereby adjusting the personalized training model and providing an auxiliary training model.

Description

Technical Field

[0001] This invention relates to gait training technology, and more particularly to a method for adjusting gait training parameters in real time. Background Technology

[0002] In general, gait training equipment is used to assist users in gait training.

[0003] For example, a dynamic orthotic device provided by US Patent No. 8147436, as shown in Figure 7 of the patent, mainly collects walking data of six normal people who actively exert force as an ideal model of gait trajectory, and plans a tunnel-shaped allowable error space around the gait trajectory, thereby providing users with a training effect close to the ideal model in gait training.

[0004] However, as described in the aforementioned US Patent No. 8147436, the technique of applying an ideal model to all users for training does not take into account individual differences among users when planning the motion model. Therefore, the ideal model planned in this patent is difficult to suit different users.

[0005] In addition, an adaptive active training system, as shown in Figure 6 of the Chinese patent CN 113244084A, mainly includes a sensing module, a control module and a motion module. It records the physiological signals of each segment when the user's muscles are relaxed and driven by the exoskeleton, calculates the physiological signal threshold of each segment, and adjusts the training difficulty in real time according to the user's physiological state signals during training.

[0006] However, as mentioned in the aforementioned Chinese patent CN 113244084A, the gait data of the user itself is used as a reference only, without referencing the gait data of normal people or other users. Therefore, the training model planned in this patent cannot achieve the optimized gait training effect. Summary of the Invention

[0007] The main objective of this invention is to provide a method for adjusting gait training parameters in real time, which can plan a personalized motion model based on the state of different users, and recommend a suitable training difficulty based on the user's force output data during training, so as to achieve the effect of adjusting the training difficulty in real time according to the actual performance during training.

[0008] To achieve the above objectives, the present invention provides a method for real-time adjustment of gait training parameters, applicable to a gait training device. The gait training device includes a sensing unit, a training unit, and a control unit. The control unit is electrically connected to the sensing unit and the training unit and controls the operation of the training unit. The method for real-time adjustment of gait training parameters includes the following steps:

[0009] Step (a) The sensing unit collects muscle-relaxed gait data of at least one first user in a gait training state and active-force gait data of at least one first user in an active-force state during gait training. The control unit establishes a standard motion model by the ratio of the active-force gait data of the first user to the muscle-relaxed gait data of the first user.

[0010] Step (b) The control unit obtains a motion model of a second user, including second user muscle relaxation gait data measured during gait training in a muscle relaxation state, and predicts at least one humanized training model by combining the second user muscle relaxation gait data with the standard motion model.

[0011] Step (c) The control unit determines whether the actual training state of the second user meets the standard of the at least one humanized training model, and then adjusts the at least one humanized training model and provides an auxiliary training model.

[0012] Therefore, the present invention provides a method for real-time adjustment of gait training parameters, which can plan a personalized motion model for the second user based on the second user's state, and recommend a suitable auxiliary training model based on the second user's force output data during training, thereby achieving the effect of adjusting the training difficulty in real time according to the actual performance during training. Attached Figure Description

[0013] Figure 1 This is a flowchart of a preferred embodiment of the present invention.

[0014] Figure 2 is a schematic diagram of the usage state of a preferred embodiment of the present invention in conjunction with a gait training device.

[0015] Figure 2a is a schematic diagram of a preferred embodiment of the present invention, showing the gait cycle.

[0016] Figure 2b is a graph of a preferred embodiment of the present invention, showing the muscle-relaxed gait data of the first user measured during gait training in a muscle-relaxed state.

[0017] Figure 2c is a graph of a preferred embodiment of the present invention, showing the active force gait data of the first user measured during gait training in an active force state.

[0018] Figure 2d is a graph of a preferred embodiment of the present invention, showing the ratio of the first user's active force gait data to the first user's muscle relaxation gait data.

[0019] Figure 2e is a graph of a preferred embodiment of the present invention, showing the ratio and average value of data measured by multiple first users in active exertion and muscle relaxation states.

[0020] Figure 2f is a graph of a preferred embodiment of the present invention, showing the center of gravity transfer interval and the hip flexion interval.

[0021] Figure 2g is a graph of a preferred embodiment of the present invention, showing the weight transfer interval and the knee extension interval.

[0022] Figure 2h is a schematic diagram of a preferred embodiment of the present invention, illustrating the implementation status of the upper sensing component and the lower sensing component of the knee pressure sensor.

[0023] Figure 3 is a graph of a preferred embodiment of the present invention, showing the personalized training model.

[0024] Figure 3a is a graph of a preferred embodiment of the present invention, showing multiple personalized training models.

[0025] Figure 3b is a graph of a preferred embodiment of the present invention, showing the muscle relaxation state data of the second user.

[0026] Figure 3c is a graph of a preferred embodiment of the present invention, showing the maximum and minimum predicted active output values ​​in the personalized training model for predicting the second user.

[0027] Figure 3d is a graph of a preferred embodiment of the present invention, showing the maximum and minimum actual active output of the second user in the actual training state.

[0028] Figure 3e is a graph of a preferred embodiment of the present invention, showing the auxiliary training model obtained after adjusting the personalized training model.

[0029] Explanation of reference numerals in the attached figures

[0030] 10: Methods for real-time adjustment of gait training parameters

[0031] 100: Gait training equipment

[0032] 101: Left Foot Force Sensor

[0033] 102: Right foot force sensor

[0034] 103: Left knee pressure sensor

[0035] 104: Right knee pressure sensor

[0036] 105: Upper Sensing Component

[0037] 106: Lower Sensing Component

[0038] 11: Pedal

[0039] F1: Center of gravity shift interval

[0040] F2: Hip flexion zone

[0041] F3: Knee extension zone

[0042] L1: 100% Difficulty Curve

[0043] L12: 10% Difficulty Curve Detailed Implementation

[0044] To illustrate the technical features of the present invention in detail, a preferred embodiment is described below with reference to the accompanying drawings, wherein, as Figure 1 As shown in Figure 2, the method 10 for real-time adjustment of gait training parameters of the present invention is mainly used in conjunction with a gait training device 100. The gait training device 100 mainly includes a sensing unit, a training unit, and a control unit. The sensing unit includes two foot force sensors and two knee pressure sensors. The foot force sensors are a left foot force sensor 101 and a right foot force sensor 102, and the knee pressure sensors are a left knee pressure sensor 103 and a right knee pressure sensor 104. The training unit includes two pedals 11 and other components for driving the user's lower limbs for training. The left foot force sensor 101 is disposed on one of the pedals 11, and the right foot force sensor 102 is disposed on the other pedal 11. The control unit is electrically connected to the sensing unit and the training unit and controls the operation of the training unit. The control unit has analysis and calculation capabilities and can be, but is not limited to, a central processing unit (CPU) or other signal processing components with analysis and calculation capabilities. When a first user or a second user uses the gait training device 100, the gait training device 100 provides the control, calculation, and actions required for the real-time adjustment of gait training parameters method 10, which mainly includes steps (a), (b), and (c). In this preferred embodiment, the left foot pressure sensor 101 and the right foot pressure sensor 102 are load cells, and the left knee pressure sensor 103 and the right knee pressure sensor 104 are thin-film pressure sensors. It is worth mentioning that the user can choose appropriate sensors according to actual needs, and is not limited thereto.

[0045] In this preferred embodiment, as shown in Figure 2a, the gait training includes at least one gait cycle, which corresponds to the gait trajectory of one foot. The gait trajectory simulates the process of a human walking from the right heel striking the ground to the left toe lifting off the ground, from the left heel striking the ground to the right toe lifting off the ground, and finally back to the right heel striking the ground. The horizontal axis of Figures 2b, 2c, 2d, 2e, 2f, 2g, 3, 3a, 3b, 3c, 3d, and 3e corresponds to the gait cycle in Figure 2a. The data in the figure is divided into 100 equal parts. The position where the user's heel touches the ground corresponds to the starting point of the gait cycle (i.e., the data point marked 0 on the horizontal axis), and the position before the heel of the same foot touches the ground again corresponds to the 99th data point on the horizontal axis.

[0046] like Figure 1 As shown in Figures 2b, 2c, 2d, and 2e, in step (a), the sensing unit collects muscle-relaxed gait data of the first user during gait training in a muscle-relaxed state (Figure 2b), and active-force gait data of the first user during gait training in an active-force state (Figure 2c). The control unit establishes a standard motion model (Figure 2d) by using the ratio of the active-force gait data to the muscle-relaxed gait data. Here, "muscle-relaxed state" means that the user does not need to exert force during gait training; the training unit is driven by the control unit of the gait training device 100, which drives the user's feet to swing. "Active-force state" means that the user's feet must actively exert force during the operation of the training unit.

[0047] In this preferred embodiment, to improve the stability of the data, multiple samples of the first user are used as an example. The standard motion model is established by the ratio of the average of multiple active force gait data of the first user to the average of multiple muscle relaxation gait data of the first user (as shown in Figure 2e). In other preferred embodiments, if the active force gait data and the muscle relaxation gait data of one first user are sufficient to be representative, the number of the first user can also be one. Therefore, the number of the first user is not limited to this preferred embodiment.

[0048] In this preferred embodiment, as shown in Figures 2a, 2f and 2g, the gait cycle is mainly divided into a weight transfer interval F1, a hip flexion interval F2 and a knee extension interval F3. The weight transfer interval F1 is divided into 0-40 equal parts in the gait cycle, the hip flexion interval F2 is divided into 45-70 equal parts in the gait cycle, and the knee extension interval F3 is divided into 80-99 equal parts in the gait cycle.

[0049] Taking the right foot of the first user as an example (the judgment method for the left foot is the same and will not be repeated here), as shown in Figure 2f, when the first user is in the center of gravity transfer interval F1, the value sensed by the right foot force sensor 102 is greater than a model threshold (the predicted value of the first user's 100% active force output). When the first user is in the hip flexion interval F2, the value sensed by the right foot force sensor 102 is less than the model threshold, as shown in Figure 2g. When the first user is in the knee extension interval F3, the value sensed by the right knee pressure sensor 104 is less than the model threshold.

[0050] In this preferred embodiment, as shown in FIG2h, the left knee pressure sensor 103 and the right knee pressure sensor 104 each have an upper sensing component 105 and a lower sensing component 106 (since the upper and lower sensing components 105 and 106 of the left and right knee pressure sensors 103 and 104 are the same components and have the same configuration relationship, only one diagram is used to illustrate the left knee pressure sensor 103 and the right knee pressure sensor 104). It is assumed that the pressure value measured by the upper sensing component 105 is... The pressure value measured by the lower sensing component 106 is The shortest distance between the center point of the upper sensing component 105 and the lower end face of the lower sensing component 106 is (In this embodiment, it is 100mm), the shortest distance between the center point of the lower sensing component 106 and the lower end face of the lower sensing component 106 is (In this embodiment, it is 10mm), then the pressure center position of the left knee pressure sensor 103 (or the right knee pressure sensor 104) = .

[0051] like Figure 1 As shown in Figure 3, in step (b), the control unit obtains a motion model of a second user. This second user motion model includes muscle-relaxed gait data measured during gait training of the second user in a muscle-relaxed state. By combining the muscle-relaxed gait data of the second user with the standard motion model, a personalized training model is predicted (as shown in Figure 3). In this preferred embodiment, as shown in Figure 3a, the personalized training model predicts difficulty curves of 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% (only the 100% difficulty curve L1 and the 10% difficulty curve L12 are shown in Figure 3a). This establishes personalized training models of different difficulties. These difficulty curves can be adjusted as needed, therefore, these difficulty curves are not limited to this preferred embodiment.

[0052] Specifically, taking the right foot as an example, since the weight transfer interval F1 corresponds to the downward stepping phase of the right foot, a higher value measured by the right foot force sensor 102 indicates a higher difficulty level for the personalized training model, and a lower value indicates a lower difficulty level. Therefore, within approximately the gait cycle of 0-40, the uppermost curve represents the 100% difficulty curve L1, and the lowermost curve represents the 10% difficulty curve L12. In the flexion zone F2 and the extension zone F3, which is the phase when the right foot is lifted, the lower the value measured by the right foot force sensor 102, the higher the difficulty of the personalized training model. The higher the value measured by the right foot force sensor 102, the lower the difficulty of the personalized training model. Therefore, in the gait cycle of approximately 45-100, the lowermost curve is the 100% difficulty curve L1, and the uppermost curve is the 10% difficulty curve L12.

[0053] In this preferred embodiment, as shown in Figures 3b, 3c, 3d, and 3e, the method for predicting the most suitable personalized training model for the second user from personalized training models of different difficulties involves the following steps: First, the second user is in a muscle-relaxed state. Taking the right foot as an example, the values ​​obtained from the right foot force sensor 102 are used to obtain the maximum and minimum values ​​of muscle relaxation in the gait cycle when the second user is in a muscle-relaxed state (as shown in Figure 3b). Then, the standard motion model is used to predict the maximum and minimum values ​​of active force exerted by the second user in the gait cycle when actively exerting force (as shown in Figure 3c). Finally, the method is used to predict the maximum and minimum values ​​of active force exerted by the second user in the active force exertion state. The values ​​obtained from the right foot force sensor 102 under the current state are used to obtain the maximum and minimum actual active force values ​​in the gait cycle when the second user actively exerts force (as shown in Figure 3d). The maximum actual active force value, the predicted maximum active force value, and the maximum value of the muscle relaxation state are then substituted into a weight transfer interval calculation formula, and the minimum actual active force value, the predicted minimum active force value, and the minimum value of the muscle relaxation state are substituted into a hip flexion interval calculation formula to obtain the force level in the weight transfer interval and the hip flexion interval. A suitable personalized training model is then recommended based on the lower force level (as shown in Figure 3e). The specific calculation method is as follows:

[0054] The formula for calculating the centroid shift interval is as follows:

[0055] .

[0056] In this preferred embodiment, as shown in Figures 3c and 3d, the output level of the second user in the center of gravity transfer interval F1 is = .

[0057] Formula for calculating hip flexion intervals:

[0058] .

[0059] In this preferred embodiment, as shown in Figures 3c and 3d, the force exerted by the second user in the hip flexion zone F2 is = .

[0060] Since the force output of the hip flexion interval F2 is less than that of the center of gravity transfer interval F1, the appropriate personalized training model is recommended based on the force output of the hip flexion interval F2. For example, the model represented by the 80% difficulty curve in Figure 3a is used as the personalized training model.

[0061] like Figure 1 As shown in 3c, 3d, and 3e, in step (c), the control unit determines whether the user's actual training state meets the standard of the personalized training model, and then adjusts the personalized training model and provides an auxiliary training model.

[0062] In this preferred embodiment, the control unit determines whether the actual training state of the second user meets the standards of the personalized training model in a manner that includes a continuous judgment method within an interval and a single-point trigger judgment method. The continuous judgment method within an interval continuously judges whether the actual training state of the second user meets the standards of the personalized training model within an interval of the gait cycle (e.g., the center of gravity transfer interval F1, the hip flexion interval F2, and the knee extension interval F3). When the second user meets the standards from the start of training, the training unit of the gait training device 100 maintains the originally set speed. When the user does not meet the standards, the control unit controls the training unit to reduce its operating speed (in this... In a preferred embodiment, the running speed is reduced by 12% each time, and the minimum reduction is 25% of the original set speed (but not limited to this). When the second user meets the standard after the training unit slows down, the control unit controls the running speed of the training unit to increase by 38% each time, up to a maximum of 100% of the original set speed. The single-point trigger judgment method is that if any data in one interval of the gait cycle (e.g., the center of gravity transfer interval F1, the hip flexion interval F2, and the knee extension interval F3) meets the standard of the personalized training model, it is considered that the standard of the personalized training model has been met, so as to avoid the second user needing to exert continuous force to adjust the personalized training model.

[0063] In this preferred embodiment, one method for the control unit to determine whether the second user has reached hip flexion in the hip flexion range F2 is, taking the right foot as an example, combining the pressure center position measured by the right knee pressure sensor 104 when the second user is in an active force exertion state. The average pressure center location measured by the second user in a relaxed state. The right foot force sensor 102 senses the force exerted by the second user in an active force output state. The right foot force sensor 102 senses the force of the second user in a relaxed state. This personalized training model The system uses parameters such as difficulty R% for judgment; when it is determined that the second user has reached hip flexion in the hip flexion interval F2, the following conditions must be met: Among them, the range of pressure center variation of the second user in a relaxed state The algorithm is as follows: the average pressure center position is half the difference between the maximum and minimum values ​​of the pressure center position recorded by the second user in the hip flexion interval F2 during the relaxed state. It is the average value of the pressure center position recorded in F2 of the hip flexion interval.

[0064] In this preferred embodiment, one method for the control unit to determine whether the second user has achieved knee extension in the knee extension range F3 is, taking the right foot as an example, to combine the pressure value measured by the right knee pressure sensor 104 when the second user is in an active force exertion state. The pressure value measured by the right knee pressure sensor 104 when the second user's muscles are relaxed. The judgment is made based on parameters such as the difficulty level R%. When it is determined that the second user has achieved knee extension in the knee extension range F3, the following conditions must be met: .

[0065] In this preferred embodiment, the determination of whether the second user's actual training state meets the standard of the personalized training model is based on continuous judgment within the interval. Specifically, it is determined whether the second user's measurement data in the hip flexion interval F2 reaches 80% of the predicted value of the personalized training model. If it does not reach 80% of the predicted value, it is considered non-compliant. If it only reaches 50% of the predicted value, it is considered as having participated in gait training. Furthermore, in the personalized training model, if the second user meets the standard of the personalized training model in four out of five gait cycles, the control unit will provide an auxiliary training model by increasing the difficulty of the personalized training model. Conversely, if the second user does not meet the standard of the personalized training model in four out of five gait cycles, the control unit will provide an auxiliary training model by decreasing the difficulty of the personalized training model.

[0066] In this preferred embodiment, as shown in Figure 3e, 70% difficulty of the personalized training model is used as an example of the auxiliary training model. In other preferred embodiments, the personalized training model is judged and adjusted by continuous judgment within the interval, and difficulty levels of 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 80%, 90%, and 100% can be defined as the auxiliary training model; the personalized training model is judged and adjusted by single-point trigger judgment, and difficulty levels of 20%, 40%, 60%, 80%, and 100% can be defined as the auxiliary training model.

[0067] Therefore, the present invention provides a method for real-time adjustment of gait training parameters, which can plan a personalized motion model for the second user based on the second user's state, and recommend a suitable auxiliary training model based on the second user's force output data during training, thereby achieving the effect of adjusting the training difficulty in real time according to the actual performance during training.

[0068] The above preferred embodiments are provided to aid in understanding the principles and methods of the present invention, and the present invention is not limited to the above preferred embodiments. Any combination and modification within the spirit and principles of the present invention should be within the protection scope of the present invention.

Claims

1. A method for real-time adjustment of gait training parameters, applicable to a gait training device, the gait training device comprising a sensing unit, a training unit, and a control unit, the control unit being electrically connected to the sensing unit and the training unit, and controlling the operation of the training unit, the method for real-time adjustment of gait training parameters comprising the steps of: The sensing unit collects muscle-relaxed gait data of at least one first user during gait training in a muscle-relaxed state, and active-force gait data of at least one first user during gait training in an active-force state. The control unit establishes a standard motion model by the ratio of the active-force gait data of the first user to the muscle-relaxed gait data of the first user. The control unit acquires a motion model of a second user, including muscle-relaxed gait data measured during gait training of the second user in a muscle-relaxed state. By combining this muscle-relaxed gait data with the standard motion model, it predicts at least one humanized training model; and The control unit determines whether the actual training state of the second user meets the standard of the at least one humanized training model, and then adjusts the at least one humanized training model and provides an auxiliary training model that is suitable for the training difficulty of the second user. in: When the second user's actual training status does not meet the standard of the personalized training model, the control unit provides the auxiliary training model by reducing the difficulty of the personalized training model; when the second user's actual training status already meets the standard of the personalized training model, the control unit provides the auxiliary training model by increasing the difficulty of the personalized training model.

2. The method for real-time adjustment of gait training parameters according to claim 1, wherein: This gait training includes at least one gait cycle, which is mainly divided into a weight transfer interval, a hip flexion interval, and a knee extension interval.

3. The method for real-time adjustment of gait training parameters according to claim 1, wherein: The at least one humanized training model can be multiple.

4. The method for real-time adjustment of gait training parameters according to claim 2, wherein: The control unit predicts the calculation method of the personalized training model as follows: It obtains a maximum and a minimum value of muscle relaxation in the gait cycle when the second user is in a relaxed state, based on the second user's muscle relaxation gait data. Then, it predicts a maximum and a minimum value of active force exertion in the gait cycle when the second user is actively exerting force, based on the standard motion model. Finally, it obtains a maximum and a minimum value of actual active force exertion in the gait cycle when the second user is actively exerting force, based on the values ​​obtained by the sensing unit. The maximum and minimum values ​​of actual active force exertion, the predicted maximum and the maximum value of muscle relaxation are then substituted into a weight transfer interval calculation formula, and the minimum and the predicted minimum values ​​of muscle relaxation are substituted into a hip flexion interval calculation formula to obtain the force exertion level in the weight transfer interval and the hip flexion interval. The lower force exertion level is then used as the personalized training model for the second user.

5. The method for real-time adjustment of gait training parameters according to claim 2, wherein: The sensing unit includes two knee pressure sensors and two foot force sensors. In this method of real-time adjustment of gait training parameters, the control unit determines whether the second user has reached hip flexion in the hip flexion interval F2, which must meet the following conditions: ,in, This refers to the pressure center position measured by one of the knee pressure sensors when the second user is actively exerting force. This refers to the average pressure center position measured by the knee pressure sensor when the second user is in a relaxed state. This refers to the force value measured by one of the foot force sensors when the second user is in a relaxed state. For this personalized training model, To set the difficulty level.

6. The method for real-time adjustment of gait training parameters according to claim 2, wherein: The sensing unit includes two knee pressure sensors and two foot force sensors. In this method of real-time adjustment of gait training parameters, the control unit determines whether the second user has achieved knee extension in the knee extension range F3, which must meet the following conditions: ,in, This refers to the pressure value measured by one of the knee pressure sensors when the second user is actively exerting force. The pressure value measured by the knee pressure sensor when the second user's muscles are relaxed. To set the difficulty level.

7. The method for real-time adjustment of gait training parameters according to claim 1, wherein: The methods for determining whether the actual training state of the second user meets the standard of the personalized training model include continuous judgment within an interval and single-point trigger judgment.

Images