Gait detection method, system and equipment based on inertial sensor and storage medium
Through the gait detection method based on inertial sensors, the problem that traditional wearable devices cannot comprehensively detect motion data is solved, and accurate detection and adaptability of gait cycles and events are achieved.
Patent Information
- Application Number
- CN202411985204.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-31
- Publication Date
- 2025-05-09
AI Technical Summary
Traditional wearable devices cannot comprehensively detect user's motion data, such as motion state detection, gait cycle and gait events, resulting in low detection accuracy and poor adaptability.
The gait detection method based on inertial sensor is used to read acceleration and angular velocity data in real time, preprocess and adaptive adjustments are performed to identify gait periods and gait events.
Accurate detection of gait cycles and events is achieved, adapting to different sports scenes and individual differences, and improving detection accuracy and adaptability.
Smart Images
Figure CN119949813A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of motion detection technology, and in particular to a gait detection method, system, device and storage medium based on an inertial sensor. Background Art
[0002] Traditional wearable devices can only detect the number of steps, step frequency and exercise time, and cannot perform more comprehensive detection of the user's exercise data, such as motion state detection, gait cycle and gait events, etc. It is difficult to adapt well to changing sports scenes, resulting in technical problems such as low detection accuracy and poor adaptability. Summary of the invention
[0003] In order to overcome the above shortcomings, the purpose of this application is to provide a gait detection method, system, device and storage medium based on inertial sensors, so as to effectively solve the above technical problems.
[0004] In order to achieve the above objectives, this application adopts the following technical solutions:
[0005] In a first aspect, the present application provides a gait detection method based on an inertial sensor, comprising:
[0006] Read the acceleration and angular velocity data obtained by the inertial sensor in real time, and pre-process the read acceleration and angular velocity data.
[0007] Preset the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular acceleration energy threshold,
[0008] In the acquired angular velocity data, continuous sampling points are selected in the angular velocity data in the first direction, and a plurality of angular velocity peak values greater than the maximum angular velocity threshold value or angular velocity valley values less than the minimum angular velocity threshold value are simultaneously selected from the continuous sampling points as pre-selected segmentation points,
[0009] Selecting two consecutive pre-selected segmentation points from the pre-selected segmentation points, when the interval time value between the two consecutive pre-selected segmentation points is between the period minimum time threshold and the period maximum time threshold, and the maximum angular velocity energy of each sampling point between the two consecutive pre-selected segmentation points is greater than the angular velocity maximum energy threshold, performing motion recognition on the two consecutive pre-selected segmentation points that meet the requirements based on the motion recognition module,
[0010] When the two consecutive pre-selected segmentation points are identified as being in a moving state, the two consecutive pre-selected segmentation points are used as gait cycle segmentation points,
[0011] Based on the current motion state, the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold are adaptively adjusted for the next gait cycle segmentation point analysis.
[0012] Based on continuous sampling points on the gait cycle between two gait cycle segmentation points, a plurality of gait waveforms are formed, and gait events are detected by comprehensive analysis of the plurality of gait waveforms, wherein the gait events include an initial ground contact event, a foot-off event, and a mid-stance event.
[0013] Furthermore, the sensor data acquired by the sensor includes three-axis acceleration data and three-axis angular velocity data on the spatial axes, and the angular velocity energy threshold is the sum of the squares of the three-axis angular velocity data.
[0014] Furthermore, preprocessing the read sensor data includes filtering the acceleration and angular velocity data by low-pass filtering.
[0015] Furthermore, after determining that the period segmentation points are period segmentation points, the first step of detecting whether the two gait period segmentation points are in a corresponding motion state includes the following:
[0016] When the detection is the first step of the current motion state, the two gait cycle segmentation points are the first step of the current motion state, and the two gait cycle segmentation points are determined to be the determined gait cycle segmentation points.
[0017] When it is detected that it is not the first step of the current motion state, the previous gait cycle between the two gait cycle segmentation points is selected for further comparison and verification.
[0018] Furthermore, when the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold are adaptively adjusted in the current motion state, it specifically includes the following:
[0019] The maximum angular velocity threshold value adjustment or the minimum angular velocity threshold value adjustment is to multiply the maximum angular velocity value or the minimum angular velocity value in the current gait cycle by the first coefficient or the second coefficient,
[0020] The maximum cycle time threshold is adjusted to be multiplied by a third coefficient based on the cycle time value of the current gait cycle.
[0021] The minimum cycle time threshold is adjusted to be multiplied by a fourth coefficient based on the cycle time value of the current gait cycle.
[0022] The angular velocity maximum energy threshold is adjusted by multiplying the angular velocity maximum energy value of the current gait cycle by a fifth coefficient.
[0023] Furthermore, when the maximum angular velocity threshold under the current motion state is adjusted, the calculation formula is:
[0024] P max =(1-C0)*P1
[0025] Where P max is the maximum angular velocity threshold, (1-C0) is the first coefficient, P1 is the maximum angular velocity value in the current gait cycle,
[0026] When you choose to adjust the minimum angular velocity threshold in the current motion state, the calculation formula is:
[0027] P min =(1-C1)*P2
[0028] Where P min is the minimum angular velocity threshold, (1-C1) is the second coefficient, and P2 is the minimum angular velocity value in the current gait cycle;
[0029] The calculation formula for adjusting the maximum time threshold of the cycle is:
[0030] T max =(1-C2)*T
[0031] Where, T max is the maximum time threshold of the cycle, (1-C2) is the third coefficient, T is the cycle time value of the current gait cycle,
[0032] The calculation formula for adjusting the minimum period time threshold is:
[0033] T min =(1+C3)*T
[0034] Where, T min is the minimum time threshold of the cycle, C3 is the fourth coefficient, T is the cycle time value of the current gait cycle,
[0035] The T max and T min Need to meet:
[0036] T1>T max >T min >T0
[0037] Wherein, T1 is the preset maximum time threshold of the cycle, and T0 is the preset minimum time threshold of the cycle;
[0038] The calculation formula for adjusting the angular velocity maximum energy threshold is:
[0039] Are max =(1-C4)*Are
[0040] Where, Are max is the maximum energy threshold of angular velocity, C4 is the fifth coefficient, and Are is the maximum energy value of angular velocity in the current gait cycle.
[0041] Furthermore, the multiple gait waveforms include an angular velocity waveform rotating around a foot in a first direction, an acceleration amplitude waveform, and an angular velocity energy waveform. Detecting gait events by comprehensively analyzing the gait waveforms includes the following steps:
[0042] Preset angular velocity reference point threshold,
[0043] The analysis area of the angular velocity waveform is determined by two consecutive gait segmentation points. The first peak or trough position that meets the angular velocity reference point threshold is selected as the first reference point in the determined angular velocity waveform analysis area. The acceleration amplitude waveform corresponding to several sampling points in the adjacent area on both sides of the first reference point is analyzed. The largest acceleration amplitude peak value among the several sampling points is selected.
[0044] When the maximum acceleration amplitude peak value among the sampling points is greater than the calculated acceleration amplitude peak value threshold value, the sampling point time corresponding to the maximum acceleration amplitude peak value is determined as the initial touchdown event,
[0045] When the acceleration amplitude waveforms of several sampling points in the vicinity of the first reference point do not have peaks, or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the first reference point or several sampling points in advance are selected as the initial touchdown event.
[0046] The angular velocity energy waveform of the area within the set time range after the reference point is selected, and the sampling point with the minimum angular velocity energy value is located in the selected area, and the located sampling point is used as the mid-stance event.
[0047] Select the angular velocity waveform of the area within the set time range after the located sampling point, select the second peak or trough position that meets the angular velocity reference point threshold as the second reference point, analyze the acceleration amplitude waveforms of several sampling points in the adjacent area on both sides of the second reference point, and select the largest or smallest acceleration amplitude peak value among the several sampling points.
[0048] When the maximum acceleration amplitude peak value is greater than the acceleration amplitude peak value threshold value, the sampling point time corresponding to the maximum acceleration amplitude peak value is determined as a foot-off event,
[0049] When the acceleration amplitude waveforms of several sampling points in the adjacent areas on both sides of the second reference point do not have peaks or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the second reference point or several delayed sampling points are selected as the foot-off event.
[0050] In a second aspect, the present application also provides a gait detection system based on an inertial sensor, and the detection system is used to execute any one of the gait detection methods described above.
[0051] In a third aspect, the present application further provides a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement any of the gait detection methods described.
[0052] In a fourth aspect, the present application also provides a wearable motion detection device, which includes the gait detection system.
[0053] Beneficial Effects
[0054] The present application provides a gait detection method, system, device and storage medium based on an inertial sensor. Based on a wearable motion detection device worn on the foot, the acceleration and gyroscope data are used to perform cycle segmentation in combination with machine learning motion type recognition and the similarity of adjacent gait cycles, and each gait cycle during walking or running is adaptively detected, and further a method of comprehensively using data waveforms for analysis to detect gait events, including gait cycle time, initial contact event IC (initial contact), foot off event FO (foot off), midstance MS (midstance), etc. For different runners, different habits and running postures, different road surfaces and changes in speed during exercise, the actual data waveforms recorded by the sensor vary greatly, but adjacent cycles are still similar, and the relevant thresholds and parameters are automatically adjusted according to the changes in the actual data, which can better adapt to different situations. At the same time, the complementary effects between multiple types of data waveforms can more reliably detect related gait events. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] The accompanying drawings are used to provide an understanding of the technical solution of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solution of the present disclosure and do not constitute a limitation on the technical solution of the present disclosure. The shapes and sizes of the components in the accompanying drawings do not reflect the actual proportions and are only intended to illustrate the content of the present application.
[0056] Figure 1 A flow chart of a gait detection method provided in one embodiment of the present application.
[0057] Figure 2 This is an angular velocity waveform diagram provided by an embodiment of the present application.
[0058] Figure 3 This is a flow chart of the algorithm of the gait detection system provided in one embodiment of the present application. DETAILED DESCRIPTION
[0059] The above scheme is further described below in conjunction with specific examples. It should be understood that these examples are used to illustrate the present application and are not limited to the scope of the present application. The implementation conditions adopted in the examples can be further adjusted as the conditions of the specific manufacturer, and the unspecified implementation conditions are usually the conditions in conventional experiments.
[0060] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connecting" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. In this article, "electrical connection" includes the situation where the constituent elements are connected together through an element with some electrical function. "Elements with some electrical function" are not particularly limited as long as they can transfer electrical signals between the connected constituent elements. "Elements with some electrical function" can be, for example, electrodes or wiring, or switching elements such as transistors, or other functional elements such as resistors, inductors or capacitors. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.
[0061] In this application, the directions or positional relationships indicated by the terms "upper", "lower", "inner", "middle", etc. are based on the directions or positional relationships shown in the drawings. These terms are mainly used to better describe the present application and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific direction, or to be constructed and operated in a specific direction.
[0062] Example
[0063] An embodiment of the present application provides a gait detection method based on an inertial sensor, such as Figure 1 As shown,
[0064] S1. Read the acceleration and angular velocity data obtained by the inertial sensor in real time, and pre-process the read acceleration and angular velocity data.
[0065] S2, preset a maximum angular velocity threshold or a minimum angular velocity threshold, a maximum cycle time threshold and a minimum cycle time threshold, and a maximum angular velocity energy threshold,
[0066] S3, in the acquired angular velocity data, select continuous sampling points in the angular velocity data in the first direction (along the lateral direction of the foot), and simultaneously select multiple angular velocity peak values greater than the maximum angular velocity threshold value, or simultaneously select angular velocity valley values less than the minimum angular velocity threshold value as pre-selected segmentation points in the continuous sampling points,
[0067] S4, selecting two consecutive pre-selected segmentation points from the pre-selected segmentation points, when the interval time value between the two consecutive pre-selected segmentation points is between the minimum time threshold of the cycle and the maximum time threshold of the cycle, and the maximum angular velocity energy in each sampling point between the two consecutive pre-selected segmentation points is greater than the maximum angular velocity energy threshold, performing motion recognition on the two consecutive pre-selected segmentation points that meet the requirements based on the motion recognition module, and when the motion recognition of the two consecutive pre-selected segmentation points is in a motion state, the two consecutive pre-selected segmentation points are used as gait cycle segmentation points,
[0068] S5, based on the current motion state, adaptively adjust the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold for use in the next gait cycle segmentation point analysis,
[0069] S6. Based on continuous sampling points on the gait cycle between two gait cycle segmentation points, a plurality of gait waveforms are formed, and gait events are detected by comprehensive analysis of the plurality of gait waveforms. The gait events include initial touchdown events, foot-off events, and mid-stance events.
[0070] In some embodiments, the sensor data acquired by the sensor includes three-axis acceleration data and three-axis angular velocity data on the spatial axes, and the angular velocity energy threshold is the sum of the squares of the three-axis angular velocity data.
[0071] In some embodiments, preprocessing the read sensor data includes filtering the acceleration and angular velocity data by low-pass filtering.
[0072] In some embodiments, the first step of detecting whether the two gait cycle segmentation points are in a corresponding motion state after determining the cycle segmentation points includes the following:
[0073] When the detection is the first step of the current motion state, the two gait cycle segmentation points are the first step of the current motion state, and the two gait cycle segmentation points are determined as the determined gait cycle segmentation points.
[0074] When it is detected that it is not the first step of the current motion state, the previous gait cycle between the two gait cycle segmentation points is selected for further comparison and verification. After determining the motion state within the gait cycle, the gait cycle is detected to see if it is the first step of the corresponding motion state. When the motion state corresponding to the gait cycle is not the first step, the previous gait cycle of the gait cycle is selected for comparison and verification. After identifying that it is walking or running, if it is the first step, it is considered that a new step is detected; if it is not the first step, it is further compared with the sensor data of the previous step for verification, because the adjacent cycle waveforms of walking or running are similar, which can further improve reliability and prevent false detection. The shorter of the two consecutive cycles can be obtained by interpolation to obtain two cycles of the same length, and then the sensor data of the two consecutive cycles are cross-correlated. If the obtained cross-correlation coefficient is greater than a preset threshold, it is considered that a new step is detected, otherwise it is considered to be other actions, and the current cycle segmentation point is skipped to continue searching forward.
[0075] In some embodiments, when the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold are adaptively adjusted in the current motion state, it specifically includes the following:
[0076] The maximum angular velocity threshold value adjustment or the minimum angular velocity threshold value adjustment is to multiply the maximum angular velocity value or the minimum angular velocity value in the current gait cycle by the first coefficient or the second coefficient, and the calculation formula is: max =(1-C0)*P1, where P max is the maximum angular velocity threshold, (1-C0) is the first coefficient, P1 is the maximum angular velocity value in the current gait cycle, when the minimum angular velocity threshold under the current motion state is adjusted, the calculation formula is: P min =(1-C1)*P2, where P min is the minimum angular velocity threshold, (1-C1) is the second coefficient, and P2 is the minimum angular velocity value in the current gait cycle.
[0077] The maximum cycle time threshold is adjusted by multiplying the cycle time value of the current gait cycle by the third coefficient. The calculation formula for adjusting the maximum cycle time threshold is: T max =(1-C2)*T, where T max is the maximum time threshold of the cycle, (1-C2) is the third coefficient, T is the cycle time value of the current gait cycle, and the calculation formula for adjusting the minimum time threshold of the cycle is: T min =(1+C3)*T, where T min is the minimum time threshold of the cycle, C3 is the fourth coefficient, T is the cycle time value of the current gait cycle, T max and T minNeed to meet: T1>T max >T min >T0, where T1 is the preset maximum time threshold of the cycle, and T0 is the preset minimum time threshold of the cycle;
[0078] The maximum energy threshold of angular velocity is adjusted by multiplying the maximum energy value of angular velocity in the current gait cycle by the fifth coefficient. The calculation formula for adjusting the maximum energy threshold of angular velocity is: max =(1-C4)*Are, where Are max is the maximum energy threshold of angular velocity, C4 is the fifth coefficient, and Are is the maximum energy value of angular velocity in the current gait cycle.
[0079] In some embodiments, detecting gait events by analyzing gait waveforms comprises the following steps:
[0080] The multiple gait waveforms include an angular velocity waveform rotating around a first direction of the foot (along the lateral direction of the foot), an acceleration amplitude waveform, and an angular velocity energy waveform. Detecting gait events by comprehensively analyzing the gait waveforms includes the following steps:
[0081] Preset angular velocity reference point threshold,
[0082] The analysis area of the angular velocity waveform is determined by two consecutive gait segmentation points. The first peak or trough position that meets the angular velocity reference point threshold is selected as the first reference point in the determined angular velocity waveform analysis area. The acceleration amplitude waveform corresponding to several sampling points in the adjacent area on both sides of the first reference point is analyzed. The maximum or minimum acceleration amplitude peak value among several sampling points is selected.
[0083] When the maximum acceleration amplitude peak value among the sampling points is greater than the calculated acceleration amplitude peak value threshold value or the minimum acceleration amplitude peak value among the sampling points is less than the calculated acceleration amplitude peak value threshold value, the sampling point time corresponding to the maximum acceleration amplitude peak value is determined as the initial touchdown event.
[0084] When the acceleration amplitude waveforms of several sampling points in the vicinity of the first reference point do not have peaks, or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the first reference point or several sampling points in advance are selected as the initial touchdown event.
[0085] The angular velocity energy waveform of the area within the set time range after the reference point is selected, and the sampling point with the minimum angular velocity energy value is located in the selected area, and the located sampling point is used as the mid-stance event.
[0086] Select the angular velocity waveform of the area within the set time range after the located sampling point, select the second peak or trough position that meets the angular velocity reference point threshold as the second reference point, analyze the acceleration amplitude waveforms of several sampling points in the adjacent area on both sides of the second reference point, and select the largest or smallest acceleration amplitude peak value among the several sampling points.
[0087] When the maximum acceleration amplitude peak value is greater than the acceleration amplitude peak value threshold, the sampling point corresponding to the maximum acceleration amplitude peak value is determined as the foot-off event.
[0088] When the acceleration amplitude waveforms of several sampling points in the adjacent areas on both sides of the second reference point do not have peaks or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the second reference point or several delayed sampling points are selected as the foot-off event.
[0089] like Figure 2 As shown in the figure, on the angular velocity waveform, the segmentation points A and B are determined as the segmentation points of the two gait cycles, and the first peak position M is found. In order to reduce the violent fluctuation of the acceleration amplitude during the ground contact process and interfere with the judgment of the gait event, the acceleration amplitude within the gait cycle range can be low-pass filtered, such as a zero-delay bidirectional Butterworth low-pass filter, to remove high-frequency noise. Find the maximum peak position of the acceleration amplitude within the neighborhood of the peak of point M. If the peak value exceeds the threshold, this peak position is the time of the detected IC event.
[0090] Starting from the detected IC event, within the possible touchdown time range (e.g. 250ms), that is, the waveform after the M point, analyze the angular velocity energy (or angular velocity amplitude) waveform. The angular velocity energy is the sum of the squares of the three-axis angular velocities. Each point can be smoothed at several sampling points in the neighborhood. Find the minimum angular velocity energy within this time range as the detected MS event.
[0091] Starting from the detected MS event, within a certain time range (for example, 200ms), that is, after the MS event sampling point, continue to analyze the lateral angular velocity waveform to find the second peak position, that is, point N, and analyze the acceleration amplitude waveform in the neighborhood of the sampling points around point N. If the peak value exceeds the threshold, then this peak position is the time of the detected FO event. If such an acceleration amplitude peak value cannot be found, then the second peak position of the lateral angular velocity just found or this peak position lagging several sampling points is used as the time of the FO event.
[0092] Another embodiment of the present application also provides a gait detection system based on an inertial sensor, such as Figure 3 As shown, the detection system is used to perform a gait detection method.
[0093] Another embodiment of the present application further provides a storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the gait detection method.
[0094] Yet another embodiment of the present application provides a wearable motion detection device, which includes a gait detection system.
[0095] The above embodiments are only for illustrating the technical concept and features of the present application, and their purpose is to enable people familiar with the technology to understand the content of the present application and implement it accordingly, and they cannot be used to limit the protection scope of the present application. Any equivalent transformation or modification made according to the spirit of the present application shall be included in the protection scope of the present application.
Claims
1. A gait detection method based on inertial sensor, characterized in that: include: Read the acceleration and angular velocity data obtained by the inertial sensor in real time, and pre-process the read acceleration and angular velocity data. Preset the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold, In the acquired angular velocity data, continuous sampling points are selected in the angular velocity data in the first direction, and a plurality of angular velocity peak values greater than the maximum angular velocity threshold value or angular velocity valley values less than the minimum angular velocity threshold value are simultaneously selected from the continuous sampling points as pre-selected segmentation points, Selecting two consecutive pre-selected segmentation points from the pre-selected segmentation points, when the interval time value between the two consecutive pre-selected segmentation points is between the period minimum time threshold and the period maximum time threshold, and the maximum angular velocity energy of each sampling point between the two consecutive pre-selected segmentation points is greater than the angular velocity maximum energy threshold, performing motion recognition on the two consecutive pre-selected segmentation points that meet the requirements based on the motion recognition module, When the two consecutive pre-selected segmentation points are identified as being in a moving state, the two consecutive pre-selected segmentation points are used as gait cycle segmentation points, Based on the current motion state, the maximum angular velocity threshold or minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold are adaptively adjusted for the next gait cycle segmentation point analysis. Based on continuous sampling points on the gait cycle between two gait cycle segmentation points, a plurality of gait waveforms are formed, and gait events are detected by comprehensive analysis of the plurality of gait waveforms, wherein the gait events include an initial ground contact event, a foot-off event, and a mid-stance event.
2. The gait detection method based on inertial sensor according to claim 1, characterized in that: The sensor data acquired by the inertial sensor includes three-axis acceleration data and three-axis angular velocity data on the spatial axis, and the angular velocity energy threshold is the square sum of the three-axis angular velocity data.
3. The gait detection method based on inertial sensor according to claim 1, characterized in that: Preprocessing the read sensor data includes filtering the acceleration and angular velocity data by low-pass filtering.
4. The gait detection method based on inertial sensor according to claim 1, characterized in that: The first step of detecting whether the two gait cycle segmentation points are in a corresponding motion state after being determined as a cycle segmentation point includes the following: When the detection is the first step of the current motion state, the two gait cycle segmentation points are the first step of the current motion state, and the two gait cycle segmentation points are determined to be the determined gait cycle segmentation points. When it is detected that it is not the first step of the current motion state, the previous gait cycle between the two gait cycle segmentation points is selected for further comparison and verification.
5. The gait detection method based on inertial sensor according to claim 1, characterized in that: When the maximum angular velocity threshold or the minimum angular velocity threshold, the maximum cycle time threshold and the minimum cycle time threshold, and the maximum angular velocity energy threshold are adaptively adjusted in the current motion state, specifically including the following: The maximum angular velocity threshold value adjustment or the minimum angular velocity threshold value adjustment is to multiply the maximum angular velocity value or the minimum angular velocity value in the current gait cycle by the first coefficient or the second coefficient, The maximum cycle time threshold is adjusted to be multiplied by a third coefficient based on the cycle time value of the current gait cycle. The minimum cycle time threshold is adjusted to be multiplied by a fourth coefficient based on the cycle time value of the current gait cycle. The angular velocity maximum energy threshold is adjusted by multiplying the angular velocity maximum energy value of the current gait cycle by a fifth coefficient.
6. The gait detection method based on inertial sensor according to claim 5, characterized in that: When you choose to adjust the maximum angular velocity threshold in the current motion state, the calculation formula is: P max =(1-C0)*P1 Where P max is the maximum angular velocity threshold, (1-C0) is the first coefficient, P1 is the maximum angular velocity value in the current gait cycle, When you choose to adjust the minimum angular velocity threshold in the current motion state, the calculation formula is: P min =(1-C1)*P2 Where P min is the minimum angular velocity threshold, (1-C1) is the second coefficient, and P2 is the minimum angular velocity value in the current gait cycle; The calculation formula for adjusting the maximum time threshold of the cycle is: T max =(1-C2)*T Where, T max is the maximum time threshold of the cycle, (1-C2) is the third coefficient, T is the cycle time value of the current gait cycle, The calculation formula for adjusting the minimum period time threshold is: T min =(1+C3)*T Where, T min is the minimum time threshold of the cycle, C3 is the fourth coefficient, T is the cycle time value of the current gait cycle, The T max and T min Need to meet: T1>T max >T min >T0 Wherein, T1 is the preset maximum time threshold of the cycle, and T0 is the preset minimum time threshold of the cycle; The calculation formula for adjusting the angular velocity maximum energy threshold is: Are max =(1-C4)*Are Where, Are max is the maximum energy threshold of angular velocity, C4 is the fifth coefficient, and Are is the maximum energy value of angular velocity in the current gait cycle.
7. The gait detection method based on inertial sensor as claimed in claim 1, characterized in that: The multiple gait waveforms include an angular velocity waveform rotating around a foot in a first direction, an acceleration amplitude waveform, and an angular velocity energy waveform. Detecting gait events by comprehensively analyzing the gait waveforms includes the following steps: Preset angular velocity reference point threshold, The analysis area of the angular velocity waveform is determined by two consecutive gait segmentation points, and the first peak or trough position that meets the angular velocity reference point threshold is selected as the first reference point in the determined angular velocity waveform analysis area, and the acceleration amplitude waveforms corresponding to a plurality of sampling points in the adjacent areas on both sides of the first reference point are analyzed, and the maximum acceleration amplitude peak value among the plurality of sampling points is selected. When the maximum acceleration amplitude peak value among the sampling points is greater than the calculated acceleration amplitude peak value threshold value, the sampling point time corresponding to the maximum acceleration amplitude peak value is determined as the initial touchdown event, When the acceleration amplitude waveforms of several sampling points in the vicinity of the first reference point do not have peaks, or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the first reference point or several sampling points in advance are selected as the initial touchdown event. The angular velocity energy waveform of the area within the set time range after the reference point is selected, and the sampling point with the minimum angular velocity energy value is located in the selected area, and the located sampling point is used as the mid-stance event. Select the angular velocity waveform of the area within the set time range after the located sampling point, select the second peak or trough position that meets the angular velocity reference point threshold as the second reference point, analyze the acceleration amplitude waveforms of several sampling points in the adjacent area on both sides of the second reference point, and select the largest or smallest acceleration amplitude peak value among the several sampling points. When the maximum acceleration amplitude peak value is greater than the acceleration amplitude peak value threshold value, the sampling point time corresponding to the maximum acceleration amplitude peak value is determined as a foot-off event, When the acceleration amplitude waveforms of several sampling points in the adjacent areas on both sides of the second reference point do not have peaks or the maximum acceleration amplitude peak is less than the acceleration amplitude peak threshold, the second reference point or several delayed sampling points are selected as the foot-off event.
8. A gait detection system based on inertial sensors, characterized in that: The detection system is used to execute the gait detection method according to any one of claims 1 to 7.
9. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the gait detection method according to any one of claims 1 to 7.
10. A wearable motion detection device, characterized in that: The device comprises a gait detection system as claimed in claim 8.