Continuous gait phase estimation method based on angle between thighs
Through the gait phase estimation method based on the angle between the thighs, data is obtained by using the inertial measurement unit and gait period division and normalization are solved, and the problems of thigh angle asymmetry and phase shift are realized, and high-precision gait phase estimation in various environments is achieved.
Patent Information
- Application Number
- CN202510374663.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-27
- Publication Date
- 2025-07-11
AI Technical Summary
The existing continuous gait phase estimation method based on thigh angle faces the problems of gait asymmetry and phase shift of the human body, resulting in deviations from the real phase from the estimation results, making it difficult to achieve accurate and stable gait phase estimation in various environments.
By using the inertial measurement unit to obtain the angle between the thighs, divide the gait period into four sub-stages, generate a cosine curve orthogonal to the angle between the thighs, and perform normalization processing, calculate the gait phase using the inverse tangent function, and construct a phase map to achieve accurate gait phase estimation.
Real-time accurate gait phase monitoring in various scenarios is achieved, the accuracy of gait phase estimation and anti-interference ability are improved, and the generalization ability across terrain is enhanced.
Smart Images
Figure CN120284247A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of motion control, and particularly relates to a continuous gait phase estimation method based on the angle between the thighs. Background Art
[0002] Wearable exoskeletons and powered prostheses are core technologies in the field of wearable robotics, aiming to enhance human motor abilities or replace missing limb functions through bionics and mechatronics means. In recent years, with the urgent needs for the rehabilitation of disabled people, the assistance in an aging society, and the enhancement of human body functions in the military / industrial fields, significant progress has been made in the research on assistive devices such as wearable exoskeletons and powered prostheses. However, there are still limitations in the control methods. The synchronization degree between the assistive device and the wearer's kinematics has become the core key to whether the wearer's movement is coordinated and stable. Therefore, it is crucial to perform gait phase estimation to obtain the current gait progress of the wearer.
[0003] With the continuous in - depth research on gait phase estimation, various continuous gait phase estimation methods have emerged one after another. One of the methods is the phase variable method. This method processes the motion signal into a phase variable, parameterizes the entire gait cycle according to kinematic changes, and then realizes the accurate detection of the gait cycle stage. The phase variable method does not need to consider problems such as gait transition discontinuity and convergence, so it is widely used. Since the thigh angle is easy to obtain and can monotonically represent the entire gait cycle, it is considered a good candidate for the phase variable. However, its performance is limited by the inherent biomechanical characteristics of human lower limb movement. Specifically, there are two obvious problems in the thigh angle curve during the movement process: (1) The areas of the positive and negative regions are not equal. Especially when the road surface slope changes, the gait in the sagittal plane of the human body shows significant asymmetry due to the need for center - of - gravity adjustment, and the phenomenon of unequal areas of the positive and negative regions is even more obvious; (2) There is a phase shift between the curve and the standard cosine function. These two problems will cause a serious deviation between the gait phase estimation result and the true phase. Therefore, there is an urgent need for a gait phase estimation method with high accuracy and strong environmental generalization ability. Summary of the Invention
[0004] In view of this, the purpose of the present invention is to provide a continuous gait phase estimation method based on the angle between the thighs, which can accurately estimate the real - time continuous gait phase in various scenarios.
[0005] To achieve the above - mentioned technical purpose, the technical solution of the present invention is as follows:
[0006] A continuous gait phase estimation method based on the angle between the thighs, comprising the following steps:
[0007] Step 1: Use an inertial measurement unit to collect the bilateral thigh angles of the user and calculate the angle between the thighs;
[0008] Step 2: Introduce multi-gait phase characteristics to split the complete gait cycle and estimate each phase of the current gait cycle;
[0009] Step 3: Generate a cosine curve orthogonal to the angle between the thighs based on the estimated values of each phase of the current gait cycle and normalize the angle between the thighs;
[0010] Step 4: Construct a phase diagram based on the normalized angle between the thighs and its orthogonal cosine curve, and use the arctangent function to obtain the mapping relationship between the gait phase and the angle, completing the continuous gait phase estimation.
[0011] A further improvement of the technical solution of the present invention is that the specific implementation method of Step 1 is as follows:
[0012] Since human movement mainly occurs in the sagittal plane, this study uses inertial measurement units fixed on both thighs to obtain the angles between both thighs and the gravity direction in the sagittal plane respectively, and defines that the thigh angle is positive when the thigh is in front of the body, and the angle between the thighs is expressed as the difference between the thigh angle of the right leg and the thigh angle of the left leg.
[0013] A further improvement of the technical solution of the present invention is that the specific implementation method of Step 2 is as follows:
[0014] Define the complete gait cycle as starting from the rising zero point of the angle curve between the thighs and ending before the next same event occurs, and divide the complete gait cycle into four sub-phases, which are: from the rising zero point to the peak point; from the peak point to the falling zero point; from the falling zero point to the valley point; from the valley point to the rising zero point of the next cycle.
[0015] Estimate the duration of the first sub-phase based on historical gait cycle data, and estimate the durations of the second, third, and fourth sub-phases based on historical gait cycle data and the data of the first sub-phase of the current gait cycle.
[0016] A further improvement of the technical solution of the present invention is that the specific implementation method of Step 3 is as follows:
[0017] For each sub-phase, generate orthogonal curves respectively, and the mathematical description is as follows:
[0018]
[0019] where t 1_est , t 2_est , t 3_est and t 4_est respectively represent the estimated lengths of the four sub-phases, and Respectively represent the generated data within four sub - phases.
[0020] To avoid changing the position of the zero point of the thigh - to - thigh angle curve, the positive and negative parts of the thigh - to - thigh angle in the current gait cycle are normalized respectively using the peak and valley values of the historical gait cycle.
[0021] A further improvement of the technical solution of the present invention lies in that the specific implementation method of step four is as follows:
[0022] The normalized angle and its orthogonal curve are respectively plotted on the vertical axis and the horizontal axis of the Cartesian coordinate system and transformed to the phase plane to obtain a phase diagram, and the gait phase is calculated using the arctangent function and the position of the thigh - to - thigh angle in the phase diagram.
[0023] Due to the adoption of the above - mentioned technical solution, the beneficial effects of the present invention are:
[0024] The present invention only performs mathematical calculation processing on the angle, with a small amount of calculation, and can accurately monitor the gait phase in real time;
[0025] The present invention estimates the gait phase based on the thigh - to - thigh angle. Relying on the inherent symmetry of the two - leg movement, the thigh - to - thigh angle has good cosine characteristics in various scenarios, improving the cross - terrain generalization ability of gait phase estimation;
[0026] The present invention generates an orthogonal curve based on the thigh - to - thigh angle, avoiding the use of integral values or differential values, and improving the anti - interference ability of gait phase estimation. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 is the overall algorithm of the technical solution of the present invention;
[0028] Figure 2 is the schematic diagram of the definition of the thigh - to - thigh angle of the present invention;
[0029] Figure 3 is the schematic diagram of the sub - phase division of the complete gait cycle of the present invention;
[0030] Figure 4 is the process of constructing a phase diagram using the normalized angle curve and its orthogonal curve of the present invention;
[0031] Figure 5 is the gait phase estimation result of the present invention in various scenarios. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0032] The following further describes the embodiments of the present invention with reference to the drawings.
[0033] The present invention proposes a gait phase estimation method based on the thigh - to - thigh angle. The overall algorithm is as Figure 1 shown, including the following steps:
[0034] Step 1: Use inertial measurement units fixed on both thighs to obtain the angles between both thighs and the direction of gravity in the sagittal plane respectively. Define the thigh angle as positive when the thigh is in front of the body, and the angle between thighs is expressed as the difference between the right thigh angle and the left thigh angle. As shown in Figure 2 , the inertial measurement unit includes a triaxial accelerometer, a triaxial gyroscope, and a triaxial magnetometer. Figure 2 As shown, the inertial measurement unit includes a triaxial accelerometer, a triaxial gyroscope, and a triaxial magnetometer.
[0035] Step 2: Define a complete gait cycle starting from the rising zero point of the angle between thighs and ending before the next same event occurs. A complete gait cycle corresponds to one period of a standard sine curve. Divide the complete gait cycle into four sub - stages, namely: from the rising zero point to the peak point; from the peak point to the falling zero point; from the falling zero point to the trough point; from the trough point to the rising zero point of the next cycle. As shown in Figure 3 . Figure 3 As shown.
[0036] Since the length changes of each sub - stage in different gait cycles have strong correlations, and the ratios of the lengths of different sub - stages to the corresponding sub - stages in adjacent gait cycles are consistent, the present invention proposes to estimate the length of the first sub - stage based on historical gait cycle data, and estimate the lengths of the second, third, and fourth sub - stages based on historical gait cycle data and the data of the first sub - stage in the current gait cycle.
[0037]
[0038] Among them, \(t_{i,j}\), \(t_{i,1}\), \(t_{i,2}\), \(t_{i,3}\), \(t_{i,4}\) respectively represent the estimated lengths of the four sub - stages, \(l_{i,j}\) is the length of the \(i\) - th sub - stage in the \(j\) - th recent historical gait cycle, \(t_{i,1}\) is estimated based on the historical \(N\) gait cycles at the start moment of the current gait cycle, \(t_{i,2}\), \(t_{i,3}\), \(t_{i,4}\) are estimated at the end moment of the first sub - stage in the current gait cycle. 1_est t 2_est t 3_est and t 4_est respectively represent the estimated lengths of the four sub - stages, is the length of the \(i\) - th sub - stage in the \(j\) - th recent historical gait cycle, t 1_est is estimated based on the historical \(N\) gait cycles at the start moment of the current gait cycle, t 2_est ,t 3_est ,t 4_est are estimated at the end moment of the first sub - stage in the current gait cycle.
[0039] Step 3: Since constructing a phase diagram requires two mutually orthogonal curves, the present invention proposes to generate a cosine curve orthogonal to the angle between thighs according to the estimated values of each stage in the current gait cycle.
[0040]
[0041] Among them, and Respectively represent the generated data within four sub - stages. Since the lengths of each stage of this curve are estimated from the real thigh - between - angle curve, the orthogonality with the original curve can be maximally maintained.
[0042] In practical applications, t 1_est and t1 are not necessarily equal, and t2, t3, t4 are estimated at the end of the first sub - stage (the length of the first sub - stage is t1). Therefore, to ensure the smooth construction of the phase diagram, the following processing is adopted: If t 1_est ≥t1, only the data with a length of t1 is retained. If t 1_est <t1, the difference between t 1_est and t1 is filled with the last data within the length of t 1_est . Similarly, if t 4_est cannot fully meet the requirement that the constructed curve length is consistent with the real - time angle curve length, the same processing method is adopted, retaining part of the data or filling the data.
[0043] During the process of normalizing the thigh - between - angle, to avoid changing the zero - point position (the zero - point position is an important basis for gait cycle division and sub - stage division), the normalization process is carried out by separately processing the positive and negative parts.
[0044]
[0045] where θ inter (t Peak ) and θ inter (t Trough ) respectively represent the peak and valley values of the previous gait cycle.
[0046] Step 4: Plot the normalized angle and its orthogonal curve on the vertical and horizontal axes of the Cartesian coordinate system respectively and transform them to the phase plane to obtain the phase diagram. Calculate the position of the thigh - between - angle in the phase diagram using the arctangent function to obtain the gait phase, as Figure 4 shown. To make the range of the gait phase be within 0% - 100%, it is finally expressed as:
[0047]
[0048] where θ inter_nor and θ est respectively represent the normalized thigh - between - angle and the constructed curve orthogonal to it, represents the final gait - phase result.
[0049] Use the inertial measurement unit, micro - controller and PC to verify the performance and stability of the proposed gait - phase estimation method in various motion scenarios such as going up and down stairs, going up and down slopes, etc. The results are as Figure 5As shown. For the results of each motion scenario, on the left is the phase diagram constructed using the normalized angle and its orthogonal curve, and on the right is the gait phase result obtained using this phase diagram. It can be seen that the phase diagrams in each scenario are very close to a standard circle (ideally, it should be a standard circle). Therefore, the linearity of the obtained gait phase is very high, and the estimated phase (green solid line) is very close to the true phase (black dashed line).
Claims
1. A continuous gait phase estimation method based on the angle between the thighs, characterized in that: The inter-thigh angle refers to the angular difference between the bilateral thighs in the sagittal plane. The continuous gait phase estimation method includes the following steps: Step 1: Use an inertial measurement unit to collect the angles of the user's bilateral thighs and calculate the inter-thigh angle; Step 2: Introduce multi-gait stage characteristics to split the complete gait cycle and estimate each stage of the current gait cycle; Step 3: Generate a cosine curve orthogonal to the inter-thigh angle based on the estimated values of each stage of the current gait cycle and normalize the inter-thigh angle; Step 4: Construct a phase diagram based on the normalized inter-thigh angle and its orthogonal cosine curve, and use the arctangent function to obtain the mapping relationship between the gait phase and the angle, thereby completing the continuous gait phase estimation.
2. The continuous gait phase estimation method based on the angle between thighs according to claim 1, wherein: In Step 1, the inertial measurement unit includes a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer. The inertial measurement unit is used to obtain the angles between the bilateral thighs and the gravity direction in the sagittal plane respectively. When the thigh is in front of the body, the thigh angle is positive, and the inter-thigh angle is expressed as the difference between the right thigh angle and the left thigh angle.
3. A continuous gait phase estimation method based on the angle between the thighs according to claim 1, characterized in that: In Step 2, the complete gait cycle is defined as starting from the rising zero point of the inter-thigh angle curve and ending before the next same event occurs. The complete gait cycle is divided into four sub-stages, namely: from the rising zero point to the peak point, from the peak point to the falling zero point, from the falling zero point to the trough point, and from the trough point to the rising zero point of the next cycle.
4. A continuous gait phase estimation method based on the angle between the thighs according to claim 3, characterized in that: In Step 2, the length of the first sub-stage is estimated based on historical gait cycle data, and the lengths of the second, third, and fourth sub-stages are estimated based on historical gait cycle data and the data of the first sub-stage of the current gait cycle. where, t 1_est , t 2_est , t 3_est and t 4_est respectively represent the estimated lengths of four sub - phases, is the length of the i - th sub - phase in the j - th recent historical gait cycle, and t 1_est is estimated based on the historical N gait cycles at the start of the current gait cycle, 2_est , t 3_est , t 4_est is estimated at the end of the first sub - phase within the current gait cycle.
5. A continuous gait phase estimation method based on the angle between the thighs according to claim 4, characterized in that: In Step 3, for each sub-stage, orthogonal curve generation is performed respectively. The mathematical description is as follows: Among them, and respectively represent the generated data within four sub - stages.
6. The continuous gait phase estimation method based on the angle between thighs according to claim 5, characterized in that: In Step 3, the positive and negative parts of the inter-thigh angle are normalized separately to avoid changing the position of the zero point of the inter-thigh angle curve. where θ inter (t Peak ) and θ inter (t Trough ) represent the peak and valley values of the previous gait cycle, respectively.
7. A continuous gait phase estimation method based on the angle between thighs according to claim 1, characterized in that: In Step 4, the normalized inter-thigh angle and its orthogonal curve are respectively plotted on the vertical axis and the horizontal axis of the Cartesian coordinate system and transformed to the phase plane to obtain the phase diagram. The gait phase is calculated by using the arctangent function to calculate the position of the inter-thigh angle in the phase diagram. To make the range of the gait phase be within 0%-100%, it is finally expressed as: where θ inter_nor and θ est represent the normalized inter-thigh angle and the constructed curve orthogonal to it respectively, representing the final gait phase result.