A body posture recognition method and device

Abstract

The embodiment of the present invention discloses a method and device for posture recognition. The method includes: obtaining the q-axis current value generated when the motor drives the treadmill; determining the characteristic data of the treadmill in each current cycle according to the q-axis current value, wherein the characteristic data of the treadmill includes multiple characteristic moments and the q-axis current values ​​corresponding to the multiple characteristic moments, and the multiple characteristic moments constitute a current cycle of the motor completing the load mutation process; according to the heavy load stability start moment and the heavy load stability end moment in the multiple characteristic moments, the posture of the athlete on the treadmill is identified. Through the present invention, the technical problem of using the sensor device of the treadmill to detect the posture of the athlete on the treadmill in the related technology, resulting in high detection cost, is solved, and the technical effect of no need for sensor devices, simpler implementation process, reduced failure rate and detection cost, and improved detection accuracy is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of treadmill data processing, and in particular to a body posture recognition method and device. Background Art

[0002] When a user uses a treadmill, it is necessary to obtain the user's current step count, speed, mileage, user posture and other data in real time, and display them through a display panel, APP, etc. and use them as the data basis for big data analysis. However, in the prior art, for the detection of human posture on a treadmill, a grating or pressure-sensitive sensor is usually installed on the treadmill pedal to detect human movements and perform posture tests. In order to improve the detection accuracy, a large number of sensors are installed, which not only increases the detection cost, but also makes the treadmill's detection system too complicated. In complex application scenarios, once a sensor fails, the treadmill's detection results will inevitably be inaccurate.

[0003] To address the above-mentioned problems, no effective solutions have been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a body posture recognition method and device to at least solve the technical problem in the related art that the body posture of an athlete on the treadmill is detected by using a sensor device of the treadmill, resulting in high detection costs.

[0005] According to one aspect of an embodiment of the present invention, a body posture recognition method is provided, comprising: obtaining a q-axis current value generated when a motor drives a treadmill; determining characteristic data of the treadmill in each current cycle based on the q-axis current value, wherein the characteristic data of the treadmill includes a plurality of characteristic moments and q-axis current values ​​corresponding to the plurality of characteristic moments, and the plurality of characteristic moments constitute one current cycle in which the motor completes a load mutation process; and identifying the body posture of an athlete on the treadmill based on a heavy load stabilization start moment and a heavy load stabilization end moment among the plurality of characteristic moments.

[0006] Optionally, when the characteristic moment is the starting moment of the load mutation, characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value is greater than or equal to a first current threshold of the motor, wherein the first current threshold is the product of the q-axis current value of the motor in a no-load state and a first predetermined multiple; when the q-axis current value is less than the first current threshold, continuing to judge whether the q-axis current value is greater than or equal to the first current threshold until the q-axis current value is greater than or equal to the first current threshold; when the q-axis current value is greater than or equal to the first current threshold, determining that the motor is at the starting moment of the load mutation, and using the q-axis current value as the q-axis current value corresponding to the starting moment of the load mutation.

[0007] Optionally, when the characteristic moment is the first load peak moment or the second load peak moment, characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment; when the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, continuing to judge whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; when the q-axis current value at the next moment is less than the q-axis current value at the previous moment, determining that the motor is at the first load peak moment or the second load peak moment, wherein, when the motor is at the first load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the first load peak moment; or, when the motor is at the second load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the second load peak moment.

[0008] Optionally, when the characteristic moment is the start moment of heavy-load stabilization, characteristic data of the treadmill in each current cycle are determined according to the q-axis current value, including: judging whether the q-axis current value is within the first predetermined current range of the motor for multiple consecutive times; determining that the motor is at the start moment of heavy-load stabilization when the q-axis current value is within the first predetermined current range for multiple consecutive times, and using the q-axis current value as the q-axis current value corresponding to the start moment of heavy-load stabilization; wherein, the first predetermined current range is determined according to a first average current value and an error current value, and the first average current value is calculated based on the q-axis current values ​​at multiple moments when the motor is in a heavy-load stable state.

[0009] Optionally, when the characteristic moment is the end moment of heavy-load stability, characteristic data of the treadmill in each current cycle are determined according to the q-axis current value, including: judging whether the q-axis current value is greater than or equal to the second current threshold of the motor, wherein the second current threshold is the product of the average q-axis current value corresponding to the motor in the heavy-load stable state and a second predetermined multiple; when the q-axis current value is less than the second current threshold, continuing to judge whether the q-axis current value is greater than or equal to the second current threshold until the q-axis current value is greater than or equal to the second current threshold; when the q-axis current value is greater than or equal to the second current threshold, determining that the motor is at the end moment of heavy-load stability, and using the q-axis current value as the q-axis current value corresponding to the end moment of heavy-load stability.

[0010] Optionally, when the characteristic moment is a no-load moment, characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value is within a second predetermined current range of the motor for multiple consecutive times; determining that the motor is at the no-load moment when the q-axis current value is within the second predetermined current range for multiple consecutive times, and using the q-axis current value as the q-axis current value corresponding to the no-load moment; wherein, the second predetermined current range is determined based on a second average current value and an error current value, and the second average current value is calculated based on the q-axis current values ​​of the motor in a no-load state in multiple current cycles.

[0011] Optionally, according to the heavy load stabilization start moment and the heavy load stabilization end moment among the multiple characteristic moments, the body posture of the athlete on the treadmill is identified, including: calculating the duration from the heavy load stabilization start moment to the heavy load stabilization end moment; when the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is greater than a first time threshold, determining that the body posture of the athlete on the treadmill is a fall; wherein the first time threshold is determined based on a benchmark library when the motor drives the treadmill.

[0012] Optionally, based on the heavy load stabilization start moment and the heavy load stabilization end moment among the multiple characteristic moments, the body posture of the athlete on the treadmill is identified, including: when the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is less than or equal to a first time threshold and the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is greater than a second time threshold, it is determined that the body posture of the athlete on the treadmill is slow walking; wherein the first time threshold and the second time threshold are both determined based on a benchmark library when the motor drives the treadmill, and the first time threshold is greater than the second time threshold.

[0013] Optionally, based on the heavy load stabilization start moment and the heavy load stabilization end moment among the multiple characteristic moments, the body posture of the athlete on the treadmill is identified, including: when the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is less than a second time threshold, determining that the body posture of the athlete on the treadmill is jogging; wherein the second time threshold is determined based on a benchmark library when the motor drives the treadmill.

[0014] Optionally, according to the heavy load stabilization start moment and the heavy load stabilization end moment among the multiple characteristic moments, the body posture of the athlete on the treadmill is identified, including: obtaining a first ratio and a second ratio, wherein the first ratio is the ratio of the q-axis current value corresponding to the first load peak moment to the q-axis current value corresponding to the heavy load stabilization moment, and the second ratio is the ratio of the q-axis current value corresponding to the second load peak moment to the q-axis current value corresponding to the heavy load stabilization moment; when the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is less than a second time threshold, the first ratio is greater than a first predetermined threshold, and the second ratio is greater than a second predetermined threshold, it is determined that the body posture of the athlete on the treadmill is a high jump; wherein the second time threshold, the first predetermined threshold and the second predetermined threshold are all determined according to a benchmark library when the motor drives the treadmill.

[0015] Optionally, the benchmark library when the motor drives the treadmill is obtained by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical benchmark library, wherein the benchmark library when the motor drives the treadmill is obtained by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical benchmark library, including: calculating the average value of the characteristic data of the treadmill in multiple current cycles to obtain average characteristic data, wherein the average characteristic data includes multiple characteristic moments and average q-axis current values ​​corresponding to the multiple characteristic moments; performing least squares calculation on the average characteristic data and multiple groups of the typical characteristic data respectively to obtain the minimum calculation result; and determining the benchmark library when the motor drives the treadmill according to the minimum calculation result.

[0016] According to another aspect of an embodiment of the present invention, a body posture recognition device is also provided, including: an acquisition module, used to acquire the q-axis current value generated when a motor drives a treadmill; a determination module, used to determine the characteristic data of the treadmill in each current cycle based on the q-axis current value, wherein the characteristic data of the treadmill includes multiple characteristic moments and the q-axis current values ​​corresponding to the multiple characteristic moments, and the multiple characteristic moments constitute one current cycle in which the motor completes a load mutation process; an identification module, used to identify the body posture of the athlete on the treadmill based on the heavy load stabilization start moment and the heavy load stabilization end moment among the multiple characteristic moments.

[0017] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute any one of the method steps described above.

[0018] In an embodiment of the present invention, the q-axis current value generated when the motor drives the treadmill is obtained; according to the q-axis current value, the characteristic data of the treadmill in each current cycle is determined, wherein the characteristic data of the treadmill includes multiple characteristic moments and the q-axis current values ​​corresponding to the multiple characteristic moments, and the multiple characteristic moments constitute a current cycle of the motor completing the load mutation process; according to the heavy load stability start moment and the heavy load stability end moment in the multiple characteristic moments, the body posture of the athlete on the treadmill is identified. That is to say, the embodiment of the present invention uses the q-axis current value generated when the motor drives the treadmill to determine the characteristic data of the treadmill in each current cycle, and then identifies the body posture of the athlete on the treadmill according to the heavy load stability start moment and the heavy load stability end moment in the characteristic data of the treadmill, thereby solving the technical problem of using the sensor device of the treadmill to detect the body posture of the athlete on the treadmill in the related art, resulting in a high detection cost, and achieving the technical effect of no need for sensor devices, simpler implementation process, reduced failure rate and detection cost, and improved detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 A flowchart of a posture recognition method provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of characteristic data of a treadmill within a current cycle provided by an optional embodiment of the present invention;

[0022] Figure 3A flowchart of determining a reference library when a motor drives a treadmill provided for an optional embodiment of the present invention;

[0023] Figure 4 A flowchart of a posture recognition method provided by an optional embodiment of the present invention;

[0024] Figure 5 A schematic diagram of a body posture recognition device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second" and the like in the specification, claims and drawings of the present invention are used to distinguish different objects rather than to limit a specific order.

[0027] According to one aspect of an embodiment of the present invention, a body posture recognition method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] Figure 1 A flow chart of a posture recognition method provided by an embodiment of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0029] Step S102, obtaining the q-axis current value generated when the motor drives the treadmill;

[0030] When the motor drives the treadmill, the q-axis current value can be collected in real time or according to a predetermined time period. The predetermined time period can be flexibly set according to the specific application scenario, and the time interval represented is relatively short, for example, 0.05ms, 0.15ms, 5ms, etc.

[0031] In addition, when the athlete is exercising on the treadmill, his / her movements at different times act on the treadmill, causing the load of the treadmill to change continuously, which will cause the q-axis current value generated when the motor drives the treadmill to change continuously. It should be noted that when the athlete is exercising on the treadmill, the q-axis current value has a periodic change law.

[0032] Step S104, determining characteristic data of the treadmill in each current cycle according to the q-axis current value, wherein the characteristic data of the treadmill includes a plurality of characteristic moments and the q-axis current values ​​corresponding to the plurality of characteristic moments, and the plurality of characteristic moments constitute a current cycle of the motor completing a load mutation process;

[0033] Optionally, an adaptive learning algorithm is used to process the q-axis current value generated when the motor drives the treadmill, and characteristic data of the treadmill in each current cycle is automatically identified.

[0034] The above-mentioned multiple characteristic moments include but are not limited to the load mutation start moment, the first load peak moment, the heavy load stability start moment, the heavy load stability moment, the heavy load stability end moment, the second load peak moment and the no-load moment. For example, the load mutation start moment, the first load peak moment, the heavy load stability start moment, the heavy load stability moment, the heavy load stability end moment, the second load peak moment and the no-load moment can constitute a current cycle of the motor completing the load mutation process; for another example, the load mutation start moment, the first load peak moment, the heavy load stability start moment, the heavy load stability end moment, the second load peak moment and the no-load moment can also constitute a current cycle of the motor completing the load mutation process. It should be noted that the load mutation start moment is the start moment of a current cycle of the motor completing the load mutation process, and the no-load moment is the end moment of a current cycle of the motor completing the load mutation process. In addition, the characteristic data of the treadmill in each current cycle corresponds to the movement of the athlete on the treadmill one by one, that is, the movement of the athlete at different times on the treadmill will have a characteristic moment corresponding to the q-axis current value corresponding to it. It should be noted that multiple moments can be spaced between every two characteristic moments.

[0035] Step S106, identifying the body posture of the athlete on the treadmill according to the heavy load stabilization start time and the heavy load stabilization end time among the multiple characteristic moments.

[0036] In an embodiment of the present invention, the q-axis current value generated when the motor drives the treadmill is obtained; according to the q-axis current value, the characteristic data of the treadmill in each current cycle is determined, wherein the characteristic data of the treadmill includes multiple characteristic moments and the q-axis current values ​​corresponding to the multiple characteristic moments, and the multiple characteristic moments constitute a current cycle of the motor completing the load mutation process; according to the heavy load stability start moment and the heavy load stability end moment in the multiple characteristic moments, the body posture of the athlete on the treadmill is identified. That is to say, the embodiment of the present invention uses the q-axis current value generated when the motor drives the treadmill to determine the characteristic data of the treadmill in each current cycle, and then identifies the body posture of the athlete on the treadmill according to the heavy load stability start moment and the heavy load stability end moment in the characteristic data of the treadmill, thereby solving the technical problem of using the sensor device of the treadmill to detect the body posture of the athlete on the treadmill in the related art, resulting in a high detection cost, and achieving the technical effect of no need for sensor devices, simpler implementation process, reduced failure rate and detection cost, and improved detection accuracy.

[0037] In an optional embodiment, the above-mentioned multiple characteristic moments may be, in sequence, the load mutation start moment, the first load peak moment, the heavy load stabilization start moment, the heavy load stabilization end moment, the second load peak moment and the no-load moment.

[0038] Figure 2 A schematic diagram of characteristic data of a treadmill within a current cycle provided by an optional embodiment of the present invention, such as Figure 2 As shown, t1, t2, t3, t4, t5, t6 and t7 constitute a current cycle of a completed load mutation process, wherein t1 is the load mutation starting moment of this current cycle, t2 is the first load peak value (i.e., the athlete lands on the treadmill, q-axis current peak value) moment of this current cycle, t3 is the heavy load stability start moment of this current cycle (i.e., the athlete is stationary on the treadmill), t4 is the heavy load stability moment of this current cycle, t5 is the heavy load stability end moment of this current cycle, t6 is the second load peak value (i.e., the athlete takes off on the treadmill, q-axis current peak value) moment of this current cycle, and t7 is the no-load (the athlete is completely detached from the treadmill) moment of this current cycle. Through the above seven characteristic moments and the corresponding q-axis current values, the calculation of running steps, speed, posture, and mileage can be realized. In addition, t0 is the no-load (the athlete is completely detached from the treadmill) moment of the previous current cycle.

[0039] In an optional embodiment, when the characteristic moment is the starting moment of the load mutation, the characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value is greater than or equal to the first current threshold of the motor, wherein the first current threshold is the product of the q-axis current value of the motor in a no-load state and a first predetermined multiple; when the q-axis current value is less than the first current threshold, continuing to judge whether the q-axis current value is greater than or equal to the first current threshold until the q-axis current value is greater than or equal to the first current threshold; when the q-axis current value is greater than or equal to the first current threshold, determining that the motor is at the starting moment of the load mutation, and using the q-axis current value as the q-axis current value corresponding to the starting moment of the load mutation.

[0040] In order to avoid false triggering, it is necessary to determine whether the q-axis current value is greater than or equal to the first current threshold of the motor. If the q-axis current value is less than the first current threshold, it will continue to determine whether the q-axis current value is greater than or equal to the first current threshold until the q-axis current value is greater than or equal to the first current threshold; if the q-axis current value is greater than or equal to the first current threshold, it can be determined that the motor is at the start of a sudden load change, and the q-axis current value is used as the q-axis current value corresponding to the start of the sudden load change. It should be noted that the above-mentioned first current threshold is the product of the q-axis current value of the motor in the no-load state and the first predetermined multiple, and the first predetermined multiple can be set according to actual needs. For example, the first predetermined multiple can be 1.1, 1.2, 1.5, etc.

[0041] In the above-mentioned embodiment of the present application, the q-axis current value is compared with the first current threshold value, and the comparison result is used to determine whether the motor is at the starting moment of the load mutation and the q-axis current value corresponding to the characteristic moment, thereby effectively avoiding the misjudgment of the characteristic moment caused by the q-axis current value being too small.

[0042] In an optional embodiment, when the characteristic moment is the first load peak moment or the second load peak moment, the characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment; when the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, continuing to judge whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment, until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; when the q-axis current value at the next moment is less than the q-axis current value at the previous moment, determining that the motor is at the first load peak moment or the second load peak moment, wherein, when the motor is at the first load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the first load peak moment; or, when the motor is at the second load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the second load peak moment.

[0043] It should be noted that the first load peak moment and the second load peak moment use the same judgment method. For example, after determining that the motor is at the start of the load mutation, it can be determined whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment; if the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, it can be determined whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment, until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; if the q-axis current value at the next moment is less than the q-axis current value at the previous moment, it can be determined that the motor is at the first load peak moment, and the q-axis current value at the previous moment is used as the q-axis current value corresponding to the first load peak moment. For another example, after determining that the motor is at the end of heavy-load stability, it is possible to determine whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment; if the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, it is possible to continue to determine whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment, until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; if the q-axis current value at the next moment is less than the q-axis current value at the previous moment, it can be determined that the motor is at the second load peak moment, and the q-axis current value at the previous moment is used as the q-axis current value corresponding to the second load peak moment.

[0044] In the above-mentioned embodiment of the present application, by comparing the q-axis current value at the next moment with the q-axis current value at the previous moment, the first load peak moment or the second load peak moment can be accurately determined.

[0045] In an optional embodiment, when the characteristic moment is the start moment of heavy load stabilization, characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value is in the first predetermined current range of the motor for multiple consecutive times; determining that the motor is in the start moment of heavy load stabilization when the q-axis current value is in the first predetermined current range for multiple consecutive times, and using the q-axis current value as the q-axis current value corresponding to the start moment of heavy load stabilization; wherein the first predetermined current range is determined according to the first average current value and the error current value, and the first average current value is calculated based on the q-axis current values ​​at multiple moments when the motor is in the heavy load stable state.

[0046] Optionally, after determining that the motor is at the second load peak moment, it is necessary to determine whether the q-axis current value is within the first predetermined current range of the motor for multiple consecutive times; if the q-axis current value is within the first predetermined current range for multiple consecutive times, it can be determined that the motor is at the beginning of heavy load stabilization, and the q-axis current value is used as the q-axis current value corresponding to the beginning of heavy load stabilization; it should be noted that the above-mentioned first predetermined current range is determined based on the first average current value and the error current value, and the first average current value is the average value calculated based on the q-axis current values ​​at multiple moments when the motor is in the heavy load stable state, and the above-mentioned error current value can be set according to the needs of the application scenario, and will not be repeated here one by one.

[0047] In the above-mentioned embodiment of the present application, whether the motor is at the heavy load stabilization start time is determined by judging whether the q-axis current value is in the first predetermined current range for multiple times in a row, thereby accurately calculating the heavy load stabilization start time.

[0048] In an optional embodiment, when the characteristic moment is the end moment of heavy-load stability, the characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: judging whether the q-axis current value is greater than or equal to the second current threshold of the motor, wherein the second current threshold is the product of the average q-axis current value corresponding to the motor in the heavy-load stable state and the second predetermined multiple; when the q-axis current value is less than the second current threshold, continuing to judge whether the q-axis current value is greater than or equal to the second current threshold until the q-axis current value is greater than or equal to the second current threshold; when the q-axis current value is greater than or equal to the second current threshold, determining that the motor is at the end moment of heavy-load stability, and using the q-axis current value as the q-axis current value corresponding to the end moment of heavy-load stability.

[0049] In order to more accurately determine the end time of the motor being in a heavy-load stable state, it is necessary to determine whether the q-axis current value is greater than or equal to the second current threshold of the motor; if the q-axis current value is less than the second current threshold, it is possible to continue to determine whether the q-axis current value is greater than or equal to the second current threshold until the q-axis current value is greater than or equal to the second current threshold; if the q-axis current value is greater than or equal to the second current threshold, it is possible to determine that the motor is in the end time of heavy-load stability, and use the q-axis current value as the q-axis current value corresponding to the end time of heavy-load stability. It should be noted that the above-mentioned second current threshold is the product of the average q-axis current value corresponding to the motor being in a heavy-load stable state and the second predetermined multiple; the average q-axis current value corresponding to the motor being in a heavy-load stable state is obtained by averaging the q-axis current values ​​at multiple moments when the motor is in a heavy-load stable state, and the second predetermined multiple can be set according to actual needs, for example, the second predetermined multiple can be 1.1, 1.2, 1.5, etc.

[0050] In the above-mentioned embodiment of the present application, the q-axis current value is compared with the second current threshold, and the comparison result is used to determine whether the motor is at the end of the load mutation and the q-axis current value corresponding to the characteristic moment, so as to more accurately obtain the end of the heavy load stability.

[0051] In an optional embodiment, when the characteristic moment is a no-load moment, characteristic data of the treadmill in each current cycle is determined based on the q-axis current value, including: judging whether the q-axis current value is within the second predetermined current range of the motor for multiple consecutive times; determining that the motor is at a no-load moment when the q-axis current value is within the second predetermined current range for multiple consecutive times, and using the q-axis current value as the q-axis current value corresponding to the no-load moment; wherein the second predetermined current range is determined based on a second average current value and an error current value, and the second average current value is calculated based on the q-axis current values ​​of the motor in a no-load state for multiple current cycles.

[0052] In order to accurately calculate the end time of a current cycle in which the motor completes the load mutation process, it is necessary to determine whether the q-axis current value is within the second predetermined current range of the motor for multiple consecutive times; if the q-axis current value is within the second predetermined current range for multiple consecutive times, it can be determined that the motor is at a no-load moment, and the q-axis current value is used as the q-axis current value corresponding to the no-load moment; it should be noted that the above-mentioned second predetermined current range is determined based on the second average current value and the error current value, and the second average current value is calculated based on the q-axis current values ​​of the motor in a no-load state in multiple current cycles, and the above-mentioned error current value can be set according to the needs of the application scenario, and will not be repeated here one by one; in addition, the value of the second average current value in the first current cycle is the q-axis current value of the motor that is always in a no-load state.

[0053] In the above-mentioned implementation mode of the present application, whether the motor is in the no-load moment is determined by judging whether the q-axis current value is in the second predetermined current range for multiple times in a row, thereby accurately calculating the no-load moment.

[0054] In an optional embodiment, when the characteristic moment is the end moment of overload stabilization, the above method also includes: calculating a third average current value based on the q-axis current value between the start moment of overload stabilization and the end moment of overload stabilization and the duration that the motor is in the overload stable state; replacing the first average current value with the third average current value, and updating the first predetermined current range of the motor.

[0055] Optionally, all q-axis current values ​​from the start time of heavy load stabilization to the end time of heavy load stabilization and the number of samplings from the start time of heavy load stabilization to the end time of heavy load stabilization can be obtained, and then all q-axis current values ​​are divided by the number of samplings to obtain a third average current value, and the third average current value is used to replace the first average current value to achieve the update of the first predetermined current range of the motor. In the next current cycle, the updated first predetermined current range can be used to determine whether the motor is at the start time of heavy load stabilization.

[0056] In the above-mentioned implementation mode of the present application, since the current cycle is repeatable, by updating the first predetermined current range of the motor, the characteristic moment of the next current cycle can be determined more accurately.

[0057] In an optional embodiment, the characteristic data of the treadmill is compared with the typical characteristic data of the treadmill in the typical benchmark library to obtain the benchmark library when the motor drives the treadmill, including: calculating the average value of the characteristic data of the treadmill in multiple current cycles to obtain average characteristic data, wherein the average characteristic data includes multiple characteristic moments and average q-axis current values ​​corresponding to the multiple characteristic moments; performing least squares calculation on the average characteristic data and multiple groups of typical characteristic data respectively to obtain the minimum calculation result; and determining the benchmark library when the motor drives the treadmill according to the minimum calculation result.

[0058] Optionally, the q-axis current values ​​corresponding to multiple characteristic moments in multiple current cycles can be averaged to obtain average characteristic data. For example, the q-axis current values ​​corresponding to the load mutation start moment in multiple current cycles can be averaged to obtain the average q-axis current value corresponding to the load mutation start moment. The average q-axis current values ​​at other characteristic moments can be calculated in sequence to obtain the average characteristic data. Furthermore, the average characteristic data can be least squared with multiple groups of typical characteristic data to obtain the minimum calculation result; finally, the reference library when the motor drives the treadmill can be determined based on the minimum calculation result.

[0059] In the above-mentioned embodiment of the present application, as the q-axis current value generated when the motor drives the treadmill is acquired, the reference library when the motor drives the treadmill can be determined in real time and the reference library can be updated.

[0060] Optionally, N groups of typical benchmark library storage space can be allocated to store typical characteristic data. When the treadmill is started, the characteristic data of the treadmill in the first M current cycles are sampled and least squares calculations are performed on the N groups of stored typical characteristic data. The value with the smallest result is selected as the benchmark library for this run, and the benchmark library is updated in real time according to the current running results, so that the treadmill can quickly enter the optimal operating state.

[0061] Figure 3 A flowchart of determining a reference library when a motor drives a treadmill is provided for an optional embodiment of the present invention, such as Figure 3 As shown, typical characteristic data of N groups of treadmills are stored in the typical benchmark library, and the typical characteristic data of the treadmill group I in the typical benchmark library are read; the characteristic data of the treadmill and the typical characteristic data of the treadmill group I are subjected to least squares calculation to obtain the calculation result RES[I]; it is determined whether I==N is true; if the determination is not true, I++ is performed, and the typical characteristic data of the next group of treadmills in the typical benchmark library is continued to be read; if the determination is true, the minimum calculation result MIN(RES[I]) is selected; finally, the benchmark library when the motor drives the treadmill is determined according to the minimum calculation result.

[0062] The typical characteristic data of the treadmill in the above-mentioned typical benchmark library is established according to the athlete's weight, running and jumping height, running speed, etc. The typical characteristic data of the treadmill includes multiple characteristic moments and typical q-axis current values ​​corresponding to multiple characteristic moments. The benchmark library when the above-mentioned motor drives the treadmill is obtained in real time by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical benchmark library. The benchmark library includes multiple characteristic moments and q-axis current values ​​corresponding to multiple characteristic moments, and these benchmark library data are used as reference standards to realize the detection of human body movement posture. For example, a reference threshold is set by setting a reference threshold through multiple characteristic moments and q-axis current values ​​corresponding to multiple characteristic moments in the benchmark library, and the characteristic data of the treadmill obtained subsequently is compared with the reference threshold, and the human body movement posture is determined according to the comparison result, wherein the human body movement posture is also called posture, and the human body movement posture includes but is not limited to falling, slow walking, fast running, high jumping, etc.

[0063] In an optional embodiment, the athlete's posture on the treadmill is identified based on the heavy-load stabilization start moment and the heavy-load stabilization end moment among multiple characteristic moments, including: calculating the duration from the heavy-load stabilization start moment to the heavy-load stabilization end moment; when the duration from the heavy-load stabilization start moment to the heavy-load stabilization end moment is greater than a first time threshold, determining that the athlete's posture on the treadmill is a fall; wherein the first time threshold is determined based on a benchmark library when the treadmill is driven by a motor.

[0064] Optionally, firstly, the duration from the start time of heavy load stabilization to the end time of heavy load stabilization needs to be calculated; then, it is determined whether the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is greater than a first time threshold; if the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is greater than the first time threshold, it is determined that the athlete's posture on the treadmill is a fall; if the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is less than or equal to the first time threshold, it is determined whether the athlete's posture on the treadmill is a slow walk or a fast run. Among them, the first time threshold is determined according to the reference library when the motor drives the treadmill.

[0065] In order to further determine whether the athlete's posture on the treadmill is a fall, the following test standards can be used:

[0066]

[0067] Among them, N represents the number of revolutions of the motor from the start of heavy-load stabilization to the end of heavy-load stabilization, K represents the speed ratio of the motor to the belt, L represents the single-sided length of the belt, D represents the diameter of the belt drive pulley, and π represents pi.

[0068] The above test standards can be used to more accurately determine whether the athlete has fallen on the treadmill. If the above test standards are met, it indicates that the athlete has fallen on the treadmill, and if the above test standards are not met, it indicates that the athlete has not fallen on the treadmill.

[0069] In the above-mentioned embodiment of the present application, by comparing the duration from the start moment of heavy-load stabilization to the end moment of heavy-load stabilization with the first time threshold, if the duration from the start moment of heavy-load stabilization to the end moment of heavy-load stabilization is greater than the first time threshold, it indicates that the duration from the start moment of heavy-load stabilization to the end moment of heavy-load stabilization is too long or continues forever, thereby more accurately determining that the athlete's posture on the treadmill is a fall.

[0070] After determining that the athlete's body posture on the treadmill is a fall, the motor is controlled to stop driving the treadmill, thereby effectively ensuring the athlete's personal safety when exercising on the treadmill.

[0071] In an optional embodiment, the athlete's posture on the treadmill is identified based on the heavy load stabilization start moment and the heavy load stabilization end moment among multiple characteristic moments, including: when the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is less than or equal to the first time threshold and the duration from the heavy load stabilization start moment to the heavy load stabilization end moment is greater than the second time threshold, then it is determined that the athlete's posture on the treadmill is slow walking; wherein the first time threshold and the second time threshold are both determined based on a benchmark library when the treadmill is driven by a motor, and the first time threshold is greater than the second time threshold.

[0072] In an optional embodiment, the athlete's posture on the treadmill is identified based on the heavy-load stabilization start moment and the heavy-load stabilization end moment among multiple characteristic moments, including: when the duration from the heavy-load stabilization start moment to the heavy-load stabilization end moment is less than a second time threshold, it is determined that the athlete's posture on the treadmill is jogging; wherein the second time threshold is determined based on a benchmark library when the motor drives the treadmill.

[0073] Optionally, it is first necessary to calculate the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization; then determine whether the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is greater than a first time threshold; if the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is less than or equal to the first time threshold, then determine whether the athlete's posture on the treadmill is slow walking, and further, determine whether the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is greater than a second time threshold; if the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is greater than the second time threshold, then determine that the athlete's posture on the treadmill is slow walking; if the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is less than the second time threshold, then determine that the athlete's posture on the treadmill is fast running; if the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is equal to the second time threshold, then it indicates that the athlete's posture on the treadmill remains unchanged.

[0074] Optionally, the second time threshold may be set to be the duration from the start time of heavy load stabilization to the end time of heavy load stabilization in the reference library.

[0075] In the above-mentioned embodiment of the present application, by comparing the duration from the start time of heavy-load stabilization to the end time of heavy-load stabilization with the first time threshold and the second time threshold in turn, the athlete's posture on the treadmill is accurately determined as slow walking or fast running according to the comparison result.

[0076] It should be noted that the above determination of whether the athlete's posture on the treadmill is slow walking or fast running is achieved under the condition that the athlete's posture on the treadmill does not fall.

[0077] In an optional embodiment, the body posture of the athlete on the treadmill is identified according to the start moment of heavy load stabilization and the end moment of heavy load stabilization in multiple characteristic moments, including: obtaining a first ratio and a second ratio, wherein the first ratio is the ratio of the q-axis current value corresponding to the first load peak moment to the q-axis current value corresponding to the heavy load stabilization moment, and the second ratio is the ratio of the q-axis current value corresponding to the second load peak moment to the q-axis current value corresponding to the heavy load stabilization moment; when the duration from the start moment of heavy load stabilization to the end moment of heavy load stabilization is less than the second time threshold, the first ratio is greater than the first predetermined threshold, and the second ratio is greater than the second predetermined threshold, it is determined that the body posture of the athlete on the treadmill is a high jump; wherein the second time threshold, the first predetermined threshold and the second predetermined threshold are all determined according to the benchmark library when the motor drives the treadmill.

[0078] Optionally, after determining that the athlete's posture on the treadmill is sprinting, the first ratio and the second ratio are respectively obtained. If the first ratio is greater than the first predetermined threshold and the second ratio is greater than the second predetermined threshold, it is determined that the athlete's posture on the treadmill is high jumping; the above-mentioned first predetermined threshold can be set to the ratio of the q-axis current value corresponding to the first load peak moment in the benchmark library to the q-axis current value corresponding to the heavy-load stable moment; the above-mentioned second predetermined threshold can be set to the ratio of the q-axis current value corresponding to the second load peak moment in the benchmark library to the q-axis current value corresponding to the heavy-load stable moment.

[0079] In the above-mentioned embodiment of the present application, it is possible to accurately determine that the athlete's body posture on the treadmill is high jumping based on determining that the athlete's body posture on the treadmill is fast running.

[0080] In addition, if the first ratio is equal to or less than the first predetermined threshold and the second ratio is equal to or less than the second predetermined threshold, it is determined that the athlete's body position on the treadmill remains unchanged or is not a high jump.

[0081] Figure 4 A flowchart of a posture recognition method provided by an optional embodiment of the present invention, such as Figure 4As shown, the method includes the following steps: reading multiple characteristic moments, wherein the multiple characteristic moments are load mutation start moment t1, first load peak moment t2, heavy load stability start moment t3, heavy load stability moment t4, heavy load stability end moment t5, second load peak moment t6 and no-load moment t7 in sequence; judging whether t5-t3 is greater than a first time threshold t_limit; if t5-t3 is greater than t_limit, judging that the athlete's posture on the treadmill is a fall, the speed of the treadmill is set to 0, and a fault shutdown is displayed; if t5-t3 is less than or equal to t_limit, limit, then continue to judge whether t5-t3 is greater than the second time threshold t_low; if t5-t3 is greater than t_low, then it is judged that the athlete's posture on the treadmill is slow walking; if t5-t3 is less than t_low, then it is judged that the athlete's posture on the treadmill is fast running; t5-t3 is equal to t_low, indicating that the state remains unchanged; after judging that the athlete's posture on the treadmill is fast running, judge whether Q(t2) / Q(t4) is greater than the first predetermined threshold M and whether Q(t6) / Q(t4) is greater than the second predetermined threshold N, then it is judged that the athlete's posture on the treadmill is high jumping.

[0082] According to another aspect of the embodiment of the present invention, a body posture recognition device is also provided. Figure 5 A schematic diagram of a body posture recognition device provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, the posture recognition device includes: an acquisition module 52, a determination module 54 and a recognition module 56. The posture recognition device is described in detail below.

[0083] An acquisition module 52 is used to acquire the q-axis current value generated when the motor drives the treadmill;

[0084] A determination module 54, connected to the acquisition module 52, is used to determine characteristic data of the treadmill in each current cycle according to the q-axis current value, wherein the characteristic data of the treadmill includes a plurality of characteristic moments and the q-axis current values ​​corresponding to the plurality of characteristic moments, and the plurality of characteristic moments constitute a current cycle in which the motor completes a load mutation process;

[0085] The identification module 56 is connected to the determination module 54 and is used to identify the body posture of the athlete on the treadmill according to the heavy load stabilization start time and the heavy load stabilization end time among the multiple characteristic moments.

[0086] It should be noted here that the above-mentioned acquisition module 52, determination module 54 and identification module 56 correspond to steps S102 to S106 in the method embodiment, and the examples and application scenarios implemented by the above-mentioned modules are the same as those of the corresponding steps, but are not limited to the contents disclosed in the above-mentioned method embodiment.

[0087] In an embodiment of the present invention, the posture recognition device uses the q-axis current value generated when the motor drives the treadmill to obtain; according to the q-axis current value, the characteristic data of the treadmill in each current cycle is determined, wherein the characteristic data of the treadmill includes multiple characteristic moments and the q-axis current values ​​corresponding to the multiple characteristic moments, and the multiple characteristic moments constitute a current cycle of the motor completing the load mutation process; according to the heavy load stability start time and the heavy load stability end time in the multiple characteristic moments, the posture of the athlete on the treadmill is identified. That is to say, the embodiment of the present invention uses the q-axis current value generated when the motor drives the treadmill to obtain, determine the characteristic data of the treadmill in each current cycle, and then identify the posture of the athlete on the treadmill according to the heavy load stability start time and the heavy load stability end time in the characteristic data of the treadmill, thereby solving the technical problem of using the sensor device of the treadmill to detect the posture of the athlete on the treadmill in the related art, resulting in a high detection cost, and achieving the technical effect of no need for sensor devices, simpler implementation process, reduced failure rate and detection cost, and improved detection accuracy.

[0088] In an optional embodiment, the multiple characteristic moments include at least: a load mutation start moment, a first load peak moment, a heavy load stabilization start moment, a heavy load stabilization end moment, a second load peak moment and a no-load moment.

[0089] In an optional embodiment, the above-mentioned determination module 54 includes: a first judgment unit, used to judge whether the q-axis current value is greater than or equal to the first current threshold of the motor when the characteristic moment is the starting moment of the load mutation, wherein the first current threshold is the product of the q-axis current value of the motor in the no-load state and a first predetermined multiple; a first processing unit, used to continue to judge whether the q-axis current value is greater than or equal to the first current threshold when the q-axis current value is less than the first current threshold, until the q-axis current value is greater than or equal to the first current threshold; a second processing unit, used to determine that the motor is at the starting moment of the load mutation when the q-axis current value is greater than or equal to the first current threshold, and use the q-axis current value as the q-axis current value corresponding to the starting moment of the load mutation.

[0090] In an optional embodiment, the above-mentioned determination module 54 includes: a second judgment unit, which is used to judge whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment when the characteristic moment is the first load peak moment or the second load peak moment; a third processing unit, which is used to continue to judge whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment when the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; a fourth processing unit, which is used to determine whether the motor is at the first load peak moment or the second load peak moment when the q-axis current value at the next moment is less than the q-axis current value at the previous moment, wherein when the motor is at the first load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the first load peak moment; or, when the motor is at the second load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the second load peak moment.

[0091] In an optional embodiment, the above-mentioned determination module 54 includes: a third judgment unit, used to judge whether the q-axis current value is in the first predetermined current range of the motor for multiple consecutive times when the characteristic moment is the start moment of heavy load stabilization; a fifth processing unit, used to determine that the motor is at the start moment of heavy load stabilization when the q-axis current value is in the first predetermined current range for multiple consecutive times, and use the q-axis current value as the q-axis current value corresponding to the start moment of heavy load stabilization; wherein the first predetermined current range is determined based on the first average current value and the error current value, and the first average current value is calculated based on the q-axis current values ​​at multiple moments when the motor is in the heavy load stable state.

[0092] In an optional embodiment, the above-mentioned determination module 54 includes: a fourth judgment unit, which is used to judge whether the q-axis current value is greater than or equal to the second current threshold of the motor when the characteristic moment is the end moment of overload stability, wherein the second current threshold is the product of the average q-axis current value corresponding to the motor in the overload stable state and the second predetermined multiple; a sixth processing unit, which is used to continue to judge whether the q-axis current value is greater than or equal to the second current threshold when the q-axis current value is less than the second current threshold, until the q-axis current value is greater than or equal to the second current threshold; a seventh processing unit, which is used to determine that the motor is at the end moment of overload stability when the q-axis current value is greater than or equal to the second current threshold, and use the q-axis current value as the q-axis current value corresponding to the end moment of overload stability.

[0093] In an optional embodiment, the above-mentioned determination module 54 includes: a fifth judgment unit, used to judge whether the q-axis current value is in the second predetermined current range of the motor for multiple consecutive times when the characteristic moment is a no-load moment; an eighth processing unit, used to determine that the motor is in a no-load moment when the q-axis current value is in the second predetermined current range for multiple consecutive times, and use the q-axis current value as the q-axis current value corresponding to the no-load moment; wherein the second predetermined current range is determined based on the second average current value and the error current value, and the second average current value is calculated based on the q-axis current value of the motor in a no-load state for multiple current cycles.

[0094] In an optional embodiment, the above-mentioned identification module 56 includes: a first calculation unit, used to calculate the duration from the start time of heavy load stabilization to the end time of heavy load stabilization; a first determination unit, used to determine that the athlete's posture on the treadmill is a fall when the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is greater than a first time threshold; wherein the first time threshold is determined based on a benchmark library when the motor drives the treadmill.

[0095] In an optional embodiment, the above-mentioned identification module 56 includes: a second determination unit, which is used to determine that the athlete's posture on the treadmill is slow walking when the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is less than or equal to the first time threshold and the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is greater than the second time threshold; wherein the first time threshold and the second time threshold are both determined according to a benchmark library when the motor drives the treadmill, and the first time threshold is greater than the second time threshold.

[0096] In an optional embodiment, the above-mentioned identification module 56 includes: a third determination unit, which is used to determine that the athlete's posture on the treadmill is jogging when the duration from the start time of heavy load stabilization to the end time of heavy load stabilization is less than a second time threshold; wherein the second time threshold is determined based on a benchmark library when the motor drives the treadmill.

[0097] In an optional embodiment, the above-mentioned identification module 56 includes: an acquisition unit, used to acquire a first ratio and a second ratio, wherein the first ratio is the ratio of the q-axis current value corresponding to the first load peak moment to the q-axis current value corresponding to the heavy-load stabilization moment, and the second ratio is the ratio of the q-axis current value corresponding to the second load peak moment to the q-axis current value corresponding to the heavy-load stabilization moment; a fourth determination unit, used to determine that the athlete's posture on the treadmill is a high jump when the duration from the start moment of the heavy-load stabilization to the end moment of the heavy-load stabilization is less than the second time threshold, the first ratio is greater than the first predetermined threshold, and the second ratio is greater than the second predetermined threshold; wherein the second time threshold, the first predetermined threshold and the second predetermined threshold are all determined based on the benchmark library when the motor drives the treadmill.

[0098] In an optional embodiment, the benchmark library when the motor drives the treadmill is obtained by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical benchmark library, wherein the above-mentioned device includes: a first calculation module, used to calculate the average value of the characteristic data of the treadmill in multiple current cycles to obtain average characteristic data, wherein the average characteristic data includes multiple characteristic moments and average q-axis current values ​​corresponding to the multiple characteristic moments; a second calculation module, used to perform least squares calculation on the average characteristic data and multiple groups of typical characteristic data respectively to obtain the minimum calculation result; a processing module, used to determine the benchmark library when the motor drives the treadmill according to the minimum calculation result.

[0099] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute any one of the above method steps.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention.

Claims

1. A posture recognition method, characterized in that: include: Get the q-axis current value generated when the motor drives the treadmill; Determine characteristic data of the treadmill in each current cycle according to the q-axis current value, wherein the characteristic data of the treadmill includes a plurality of characteristic moments and the q-axis current values ​​corresponding to the plurality of characteristic moments, and the plurality of characteristic moments constitute one current cycle of the motor completing a load mutation process; According to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, identifying the body posture of the athlete on the treadmill, the body posture including falling, slow walking, fast running and high jumping; Wherein, when the characteristic moment is the heavy load stabilization start moment, determining the characteristic data of the treadmill in each current cycle according to the q-axis current value, including: judging whether the q-axis current value is within the first predetermined current range of the motor for multiple consecutive times; determining that the motor is at the heavy load stabilization start moment when the q-axis current value is within the first predetermined current range for multiple consecutive times, and using the q-axis current value as the q-axis current value corresponding to the heavy load stabilization start moment; wherein, the first predetermined current range is determined according to a first average current value and an error current value, and the first average current value is calculated according to the q-axis current values ​​at multiple moments when the motor is in the heavy load stable state; When the characteristic moment is the end moment of heavy-load stability, characteristic data of the treadmill in each current cycle are determined according to the q-axis current value, including: judging whether the q-axis current value is greater than or equal to the second current threshold of the motor, wherein the second current threshold is the product of the average q-axis current value corresponding to the motor in the heavy-load stable state and a second predetermined multiple; when the q-axis current value is less than the second current threshold, continuing to judge whether the q-axis current value is greater than or equal to the second current threshold until the q-axis current value is greater than or equal to the second current threshold; when the q-axis current value is greater than or equal to the second current threshold, determining that the motor is at the end moment of heavy-load stability, and using the q-axis current value as the q-axis current value corresponding to the end moment of heavy-load stability.

2. The method according to claim 1, characterized in that When the characteristic moment is the load mutation start moment, characteristic data of the treadmill in each current cycle is determined according to the q-axis current value, including: Determine whether the q-axis current value is greater than or equal to a first current threshold of the motor, wherein the first current threshold is the product of the q-axis current value of the motor in a no-load state and a first predetermined multiple; When the q-axis current value is less than the first current threshold, continue to determine whether the q-axis current value is greater than or equal to the first current threshold until the q-axis current value is greater than or equal to the first current threshold; When the q-axis current value is greater than or equal to the first current threshold, it is determined that the motor is at the load mutation starting moment, and the q-axis current value is used as the q-axis current value corresponding to the load mutation starting moment.

3. The method according to claim 1, characterized in that When the characteristic moment is the first load peak moment or the second load peak moment, determining characteristic data of the treadmill in each current cycle according to the q-axis current value includes: Determine whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment; If the q-axis current value at the next moment is greater than or equal to the q-axis current value at the previous moment, then continue to determine whether the q-axis current value at the next moment is less than the q-axis current value at the previous moment, until the q-axis current value at the next moment is less than the q-axis current value at the previous moment; When the q-axis current value at the next moment is less than the q-axis current value at the previous moment, it is determined that the motor is at the first load peak moment or the second load peak moment, wherein, when the motor is at the first load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the first load peak moment; or, when the motor is at the second load peak moment, the q-axis current value at the previous moment is used as the q-axis current value corresponding to the second load peak moment.

4. The method according to claim 1, characterized in that When the characteristic moment is a no-load moment, determining characteristic data of the treadmill in each current cycle according to the q-axis current value includes: Determining whether the q-axis current value is within a second predetermined current range of the motor for multiple consecutive times; When the q-axis current value is within the second predetermined current range for multiple times in a row, determining that the motor is at the no-load moment, and using the q-axis current value as the q-axis current value corresponding to the no-load moment; The second predetermined current range is determined based on a second average current value and an error current value, and the second average current value is calculated based on q-axis current values ​​of the motor in a no-load state during multiple current cycles.

5. The method according to claim 1, characterized in that According to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, identifying the body posture of the athlete on the treadmill comprises: Calculating the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization; In the case where the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization is greater than a first time threshold, it is determined that the athlete's posture on the treadmill is a fall; Wherein, the first time threshold is determined according to a reference library when the motor drives the treadmill.

6. The method according to claim 1, characterized in that According to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, identifying the body posture of the athlete on the treadmill comprises: In the case where the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization is less than or equal to the first time threshold and the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization is greater than the second time threshold, it is determined that the body posture of the athlete on the treadmill is slow walking; The first time threshold and the second time threshold are both determined according to a reference library when the motor drives the treadmill, and the first time threshold is greater than the second time threshold.

7. The method according to claim 2, characterized in that According to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, identifying the body posture of the athlete on the treadmill comprises: When the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization is less than a second time threshold, it is determined that the athlete's body posture on the treadmill is fast running; Wherein, the second time threshold is determined according to a reference library when the motor drives the treadmill.

8. The method according to claim 2, characterized in that: According to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, identifying the body posture of the athlete on the treadmill comprises: Obtain a first ratio and a second ratio, wherein the first ratio is a ratio of a q-axis current value corresponding to a first load peak moment to a q-axis current value corresponding to a heavy load stable moment, and the second ratio is a ratio of a q-axis current value corresponding to a second load peak moment to a q-axis current value corresponding to the heavy load stable moment; In the case where the duration from the start time of the heavy load stabilization to the end time of the heavy load stabilization is less than a second time threshold, the first ratio is greater than a first predetermined threshold, and the second ratio is greater than a second predetermined threshold, it is determined that the athlete's posture on the treadmill is a high jump; The second time threshold, the first predetermined threshold and the second predetermined threshold are all determined according to a reference library when the motor drives the treadmill.

9. The method according to any one of claims 5 to 8, characterized in that The reference library when the motor drives the treadmill is obtained by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical reference library, wherein the reference library when the motor drives the treadmill is obtained by comparing the characteristic data of the treadmill with the typical characteristic data of the treadmill in the typical reference library, including: Calculating the average value of the characteristic data of the treadmill in multiple current cycles to obtain average characteristic data, wherein the average characteristic data includes multiple characteristic moments and average q-axis current values ​​corresponding to the multiple characteristic moments; Performing least square calculation on the average characteristic data and the plurality of groups of typical characteristic data respectively to obtain the minimum calculation result; The reference library when the motor drives the treadmill is determined according to the minimum calculation result.

10. A posture recognition device, characterized in that: include: An acquisition module is used to obtain the q-axis current value generated when the motor drives the treadmill; A determination module, configured to determine characteristic data of the treadmill in each current cycle according to the q-axis current value, wherein the characteristic data of the treadmill includes a plurality of characteristic moments and the q-axis current values ​​corresponding to the plurality of characteristic moments, and the plurality of characteristic moments constitute one current cycle in which the motor completes a load mutation process; an identification module, for identifying the posture of the athlete on the treadmill according to the heavy load stabilization start time and the heavy load stabilization end time among the plurality of characteristic moments, wherein the posture includes falling, slow walking, fast running and high jumping; Wherein, the determination module includes: a third judgment unit, used for judging whether the q-axis current value is within the first predetermined current range of the motor for multiple times in a row when the characteristic moment is the heavy load stability start moment; a fifth processing unit, used for determining that the motor is at the heavy load stability start moment when the q-axis current value is within the first predetermined current range for multiple times in a row, and using the q-axis current value as the q-axis current value corresponding to the heavy load stability start moment; wherein, the first predetermined current range is determined based on a first average current value and an error current value, and the first average current value is calculated based on the q-axis current values ​​at multiple moments when the motor is in a heavy load stable state; The determination module includes: a fourth judgment unit, used to judge whether the q-axis current value is greater than or equal to the second current threshold of the motor when the characteristic moment is the end moment of heavy load stability, wherein the second current threshold is the product of the average q-axis current value corresponding to the motor being in the heavy load stable state and a second predetermined multiple; a sixth processing unit, used to continue to judge whether the q-axis current value is greater than or equal to the second current threshold when the q-axis current value is less than the second current threshold, until the q-axis current value is greater than or equal to the second current threshold; a seventh processing unit, used to determine that the motor is at the end moment of heavy load stability when the q-axis current value is greater than or equal to the second current threshold, and use the q-axis current value as the q-axis current value corresponding to the end moment of heavy load stability.

Images