Pedometer with Accelerometer and Foot Motion Distinguishing Method

Abstract

A method for distinguishing a foot motion of a user by placing a pedometer at a foot of a user includes the steps of collecting an accelerating data from an accelerometer in a real time manner; filtering the accelerating data via a smoothing filter and a Kalman filter; generating a step data that represents number of steps taken by the user in response to the accelerating data through the smoothing filter; generating an activity data that represents a foot motion of the user in response to the accelerating data through the Kalman filter; and combining the step data and the activity data to form a resulted data that distinguishes the foot motion with step count of the user.

Claims

1. A method for distinguishing a foot motion of a user by placing a pedometer at a foot of a user, which comprises the steps executed by a computerize device of: (a) collecting an accelerating data from an accelerometer of said pedometer in a real time manner; (b) filtering said accelerating data via a smoothing filter and a Kalman filter; (c) generating a step data that represents number of steps taken by the user in response to said accelerating data through said smoothing filter; (d) generating an activity data that represents a foot motion of the user in response to said accelerating data through said Kalman filter; and (e) combining said step data and said activity data to form a resulted data that distinguishes the foot motion with step count of the user.

2. The method as recited in claim 1 wherein, in the step (b), said accelerating data from said accelerometer is duplicated to form two sets of identical accelerating data that a first set of accelerating data is processed via said smoothing filter and a second set of accelerating data is processed via said Kalman filter.

3. The method as recited in claim 1 wherein, in the step (c), said step data generated by averaging two consequent accelerating data via said smoothing filter to smooth said accelerating data.

4. The method as recited in claim 2 wherein, in the step (c), said step data generated by averaging two consequent accelerating data via said smoothing filter to smooth said accelerating data.

5. The method, as recited in claim 1, wherein the step (a) further comprises the steps of: (a.1) extracting said accelerating data from X axis, Y axis, and Z axis to obtain X value, Y value, and Z value of said accelerating data respectively, wherein said X axis refers to a foot motion in a forward direction, said Y axis refers to a foot motion in a left-and-right direction, and said Z axis refers to a foot motion in an elevated direction; (a.2) comparing each of said X values of said accelerating data with a predetermined X threshold; and (a.3) determining a one step motion user when the previous X value is smaller than said X threshold and the following X value is larger than said X threshold.

6. The method, as recited in claim 4, wherein the step (a) further comprises the steps of: (a.1) extracting said accelerating data from X axis, Y axis, and Z axis to obtain X value, Y value, and Z value of said accelerating data respectively; (a.2) comparing each of said X values of said accelerating data with a predetermined X threshold; and (a.3) determining a one step motion user when the previous X value is smaller than said X threshold and the following X value is larger than said X threshold.

7. The method, as recited in claim 5, wherein said accelerometer is set at 25 Hz that said accelerometer is set at 25 Hz that no more than 5 steps are counted within 25 consequent X values, such that 8 or more consequent X values are collected for counting one step motion.

8. The method, as recited in claim 6, wherein said accelerometer is set at 25 Hz that said accelerometer is set at 25 Hz that no more than 5 steps are counted within 25 consequent X values, such that 8 or more consequent X values are collected for counting one step motion.

9. The method, as recited in claim 5, wherein the step (d) further comprises the steps of: (d.1) determining a resultant acceleration in response to said X value, Y value, and Z value of said accelerating data; and (d.2) determining an activity of walking, jogging, and running of the user by said activity data related to a value of said resultant acceleration, wherein the activity of walking has smaller value of said resultant acceleration, and the activity of running has bigger value of said resultant acceleration.

10. The method, as recited in claim 8, wherein the step (d) further comprises the steps of: (d.1) determining a resultant acceleration in response to said X value, Y value, and Z value of said accelerating data; and (d.2) determining an activity of walking, jogging, and running of the user by said activity data related to a value of said resultant acceleration, wherein the activity of walking has smaller value of said resultant acceleration, and the activity of running has bigger value of said resultant acceleration.

11. The method, as recited in claim 9, wherein the step (d) further comprises the steps of: (d.3) forming a wave form in response to said resultant acceleration; and (d.4) comparing said wave form with a plurality of wave form configurations to distinguish a motion posture of the user.

12. The method, as recited in claim 10, wherein the step (d) further comprises the steps of: (d.3) forming a wave form in response to said resultant acceleration; and (d.4) comparing said wave form with a plurality of wave form configurations to distinguish a motion posture of the user.

13. The method, as recited in claim 5, wherein said accelerometer is initiated to automatically define said X axis, Y axis, and Z axis to obtain X value, Y value, and Z value of said accelerating data.

14. The method, as recited in claim 1, wherein said pedometer is automatically activated by a sensor selected from a group consisting of pressure sensor and motion sensor.

15. The method, as recited in claim 1, wherein said pedometer is embedded in a shoe sole of a shoe worn by the user at a mid portion thereof between toe and heel portions.

16. The method, as recited in claim 1, wherein said pedometer is held on a shoe vamp of a shoe worn by the user.

17. A pedometer, comprising: a casing adapted for being placing at a foot of a user; an accelerometer, which is received in said casing, that collects an accelerating data in a real time manner; a smoothing filter filtering said accelerating data to generate a step data that represents number of steps taken by the user; a Kalman filter filtering said accelerating data to generate an activity data that represents a foot motion of the user; and a processor combining said step data and said activity data to form a resulted data that distinguishes the foot motion with step count of the user.

18. The pedometer, as recited in claim 17, wherein said accelerating data from said accelerometer is duplicated to form two sets of identical accelerating data that a first set of accelerating data is processed via said smoothing filter and a second set of accelerating data is processed via said Kalman filter.

19. The pedometer, as recited in claim 18, further comprising a comparison module operatively linked to said processor, wherein said accelerating data is extracted to obtain X value, Y value, and Z value of said accelerating data along from X axis, Y axis, and Z axis respectively, wherein each of said X values of said accelerating data is compared with a predetermined X threshold via said comparison module for determining a one step motion user when the previous X value is smaller than said X threshold and the following X value is larger than said X threshold.

20. The pedometer, as recited in claim 19, wherein said processor further generates a resultant acceleration in response to said X value, Y value, and Z value of said accelerating data for determining an activity of walking, jogging, and running of the user by said activity data related to a value of said resultant acceleration, wherein the activity of walking has smaller value of said resultant acceleration, and the activity of running has bigger value of said resultant acceleration.

21. The pedometer, as recited in claim 20, wherein said comparison module further compares a wave form formed in response to said resultant acceleration with a plurality of wave form configurations for distinguishing a motion posture of the user.

22. The pedometer, as recited in claim 1, wherein said pedometer is automatically activated by a sensor selected from a group consisting of pressure sensor and motion sensor.

23. A shoe, comprising: a shoe body adapted to be worn on a foot of a user; and a pedometer, which is carried by said shoe body and positioned with respect to a center of gravity of the user, comprising: a casing adapted to be carried by said shoe body; an accelerometer, which is received in said casing, that collects an accelerating data in a real time manner during movement of said shoe body by the user; a smoothing filter filtering said accelerating data to generate a step data that represents number of steps taken by said shoe body by the user; a Kalman filter filtering said accelerating data to generate an activity data that represents a foot motion of said shoe body by the user; and a processor combining said step data and said activity data to form a resulted data that distinguishes the foot motion with step count of said shoe body of the user.

24. The shoe, as recited in claim 23, wherein said accelerating data from said accelerometer is duplicated to form two sets of identical accelerating data that a first set of accelerating data is processed via said smoothing filter and a second set of accelerating data is processed via said Kalman filter.

25. The shoe, as recited in claim 24, further comprising a comparison module operatively linked to said processor, wherein said accelerating data is extracted to obtain X value, Y value, and Z value of said accelerating data along from X axis, Y axis, and Z axis respectively, wherein each of said X values of said accelerating data is compared with a predetermined X threshold via said comparison module for determining a one step motion user when the previous X value is smaller than said X threshold and the following X value is larger than said X threshold.

26. The shoe, as recited in claim 25, wherein said processor further generates a resultant acceleration in response to said X value, Y value, and Z value of said accelerating data for determining an activity of walking, jogging, and running of the user by said activity data related to a value of said resultant acceleration, wherein the activity of walking has smaller value of said resultant acceleration, and the activity of running has bigger value of said resultant acceleration.

27. The shoe, as recited in claim 26, wherein said comparison module further compares a wave form formed in response to said resultant acceleration with a plurality of wave form configurations for distinguishing a motion posture of the user.

28. The shoe, as recited in claim 23, wherein said pedometer is automatically activated by a sensor selected from a group consisting of pressure sensor and motion sensor.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG. 1 is a graph illustrating an accelerating data along a X axis generated by a pedometer according to a preferred embodiment of the present invention.

[0033] FIG. 2 is a graph illustrating an accelerating data processed through a Kalman filter according to the above preferred embodiment of the present invention.

[0034] FIG. 3 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing running activity via a front soles portion of the user stepping on a ground surface.

[0035] FIG. 4 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing on-the-stairs activity.

[0036] FIG. 5 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing down-the-stairs activity.

[0037] FIG. 6 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing running activity via a full soles portion of the user stepping on a ground surface.

[0038] FIG. 7 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing running activity via a heel portion of the user stepping on a ground surface.

[0039] FIG. 8 is a graph illustrating the accelerating data processed through the Kalman filter according to the above preferred embodiment of the present invention, illustrating the wave form of the activity data representing jogging activity.

[0040] FIG. 9 is a flow chart of the foot motion distinguishing method according to the above preferred embodiment of the present invention.

[0041] FIG. 10 is a block diagram of the pedometer according to the above preferred embodiment of the present invention.

[0042] FIG. 10A is a block diagram of an alternative mode of the pedometer according to the above preferred embodiment of the present invention.

[0043] FIG. 11 illustrates the pedometer embedded in a shoe sole of a shoe according to the above preferred embodiment of the present invention.

[0044] FIG. 12 illustrates the pedometer held on a shoe vamp of a shoe according to the above preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0045] The following description is disclosed to enable any person skilled in the art to make and use the present invention. Preferred embodiments are provided in the following description only as examples and modifications will be apparent to those skilled in the art. The general principles defined in the following description would be applied to other embodiments, alternatives, modifications, equivalents, and applications without departing from the spirit and scope of the present invention.

[0046] A pedometer 10 with an accelerometer 20 and foot motion distinguishing arrangement, which not only accurately records number of steps taken by a user but also precisely distinguish the foot motion of the user to differentiate the user movement of walking, jogging and running.

[0047] As shown in FIG. 10, the pedometer 10 comprises a casing 11 adapted for being placing at a foot of a user, a processor 30 such as a microcontroller unit (MCU), and a comparison module 40. The casing 11 has a power source compartment 12 for receiving a power source, such as a replaceable watch battery or rechargeable watch battery, wherein the accelerometer 20 is received in the casing 30 to electrically linked to the power source in the power source compartment 12.

[0048] The accelerometer 20 is initiated to define a X axis, a Y axis, and a Z axis. The X axis refers to a foot motion in a forward direction. The Y axis refers to a foot motion in a left-and-right direction, wherein when the foot moves to the left direction, the value of Y axis is positive and when the foot moves to the right direction, the value of Y axis is negative. The Z axis refers to a foot motion in an elevated direction, wherein when the user elevates the foot, the value of Z axis is negative, and when the user lowers the foot, the value of Z axis is positive. It is worth mentioning that the accelerometer 20 is initiated to automatically define the X axis, Y axis, and Z axis to obtain X value, Y value, and Z value of the accelerating data. For example, when the pedometer 10 is horizontally supported, such as embedded in the shoe sole in FIG. 11, the accelerometer 20 is initiated to automatically define the X axis, Y axis, and Z axis with respect to the horizontal pedometer 10. When the pedometer 10 is inclinedly supported, such as being held on the shoe vamp as shown in FIG. 12, the accelerometer 20 can also be initiated to automatically define the X axis, Y axis, and Z axis with respect to the inclined pedometer 10.

[0049] According to the preferred embodiment, the pedometer 10 is automatically activated by a sensor 21. The sensor 21 can be a pressure sensor and/or a motion sensor. For example, when the pedometer 10 is embedded in the shoe sole, the pressure sensor can be used. Therefore, when the user applies pressures, such as walking or running, the pedometer 10 will be automatically activated to activate the accelerometer 20 for data collection. Likewise, when the pedometer 10 is held on the shoe vamp, the motion sensor can be used. Therefore, when the user moves his or her foot, such as walking or running, the pedometer 10 will also be automatically activated to activate the accelerometer 20 for data collection.

[0050] The method comprises the following steps, which are executed by a computerized device.

[0051] (1) Collect the accelerating data from the accelerometer 20 placed at one foot of the user in a real time manner. For example, the pedometer 10 can be embedded in the shoe sole, preferably at the mid portion thereof between the toe and heel portions. Alternatively, the pedometer 10 can be held on the shoe vamp via the shoe tie. It is worth mentioning that the pedometer 10 is only held at one foot of the user.

[0052] (2) Process and filter the collected accelerating data via a smoothing filter 31 and a Kalman filter 32 to smooth the collected accelerating data and to minimize deviation thereof. Accordingly, the accelerating data from the accelerometer 20 is duplicated by the processor 30 to form two sets of identical accelerating data via the processor 30. The first set of accelerating data is processed via the smoothing filter 31 and the second set of accelerating data is processed via the Kalman filter 32.

[0053] (3) Generate a step data that represents number of steps taken by the user in response to the collected accelerating data through the smoothing filter 31. Accordingly, the processor 30 will generate the step data after the accelerating data is filtered by the smoothing filter 31.

[0054] (4) Generate an activity data that represents a foot motion of the user in response to the collected accelerating data through the Kalman filter 32. Accordingly, the processor 30 will generate the activity data after the accelerating data is filtered by the Kalman filter 32.

[0055] (5) Combine the step data and the activity data via the processor 30 to form a resulted data that distinguishes the foot motion with step count of the user.

[0056] According to the preferred embodiment, the present invention uses Fused HMM algorithm to obtain two or more temporal sequences at the same time, so as to ensure the enhance the interaction and characterization for dynamically representing a state or condition at one moment for activity recognition. If one of the HMM fails to obtain the data, another HMM is able to pick the data to ensure the stability of the system. The present invention will obtain the data in response to the activity of the user and project for estimating the following data so as to make more accurate predictions. In other words, before the workout of the user is completed, the resulted data is already processed that distinguishes the foot motion with step count of the user.

[0057] In the step (3), the accelerating data is processed through the smoothing filter 31 to obtain smooth step data by the moving average method in a scale space, wherein every two consequent accelerating data will be averaged to from the smooth accelerating data. The size of the scale space is directly related to the smoothing effect. The smoothing effect will be enhanced by increasing the scale space. However, if the scale space is extremely large, marginal information will be lost through the smoothing process, such that the output value of the smoothing filter 31 will be vague. In addition, the wave form through the smoothing filter 31 will lag that it cannot distinguish the motion posture of the user. On the other hand, the accelerating data is processed through the smoothing filter 31 can be accurately determine number of steps taken by the user.

[0058] The accelerating data is extracted along three axes, i.e. X axis, Y axis, and Z axis. The values of X axis, referring to X values, are continuously collected in a real manner to form a temporal sequence of X value. A X threshold (Ax) is preset to compare with the X value via the comparison module 40, wherein each of X values in sequence is collected and compared with the X threshold. It is counted as one step motion of the user when the previous X value is smaller than the X threshold and the following X value is larger than the X threshold. In other words, when the user lifts his or her foot, the X value is smaller than the X threshold. When the user drops his or her foot, X value is larger than the X threshold. Therefore, it is counted as one step motion of the user when the user drops his or her foot and then drops his or her foot. Due to the error or noise during the data collection, the X values are collected that the values thereof are fluctuated close to the X threshold in response to one step motion of the user. Once the X values are compared with the X threshold, more than one step will be counted. In order to reduce the error, the data collection is configured that the accelerometer 20 is set at 25 Hz. Since normal human will not able to take five steps in one second, no more than 5 steps will be counted within 25 consequent X values. It is worth mentioning since the pedometer 10 is only located at one foot of the user, such as provided at one shoe worn by the user, the foot motion is counted to have no more than 3 steps when the user walks 5 steps. In other words, the step count will not be exceed 3 steps in one second, such that 8 or more consequent X values will be collected for counting one step motion. As a result, the pedometer 10 of the present invention can accurately count the footsteps of the user by using this algorithm.

[0059] In the step (4), the accelerating data is processed through the Kalman filter 32 to obtain the activity data by optimizing prediction that produces estimates of the current state variables, so as to smooth the wave form of the activity data. Therefore, the wave form of the activity data can be analyzed for distinguishing the motion posture of the user.

[0060] The activity data is analyzed to define an activity periodicity, wherein different activities have different activity periodicities. In addition, different wave forms represent different activities in each activity periodicity. As a result, each wave form can be distinguished for a particular motion posture of the user.

[0061] The value of the activity data is directly related to the intensity of the activity of the user. In other words, different values of the activity data are directly related to activity of walking, jogging, and running in response to a value of a resultant acceleration. Accordingly, the resultant acceleration can be determined via the processor 30 by the following formulas.

a=√{square root over ((a_x.sup.2+a_y.sup.2+a_z.sup.2))}

[0062] a refers to the resultant acceleration, a_x, a_y, and a_z refer to accelerating data from X axis, Y axis, and Z axis respectively.

[0063] An average resultant acceleration a′ is determined by averaging the values of a within a period. Therefore, the value of a′ will represents the activity of walking or running. In other words, the activity of walking will have smaller value of a′, and the activity of running will have bigger value of a′. After the activity of the user is distinguished, the characteristic value of the wave form is obtained by further data analysis. Therefore, the characteristic value of the wave form is classified to distinguish the motion posture of the user. Accordingly, the characteristic value of the wave form is obtained by an average value, a mean difference, a quartile deviation, a variation coefficient, and a skewness of the wave form in one period to distinguish the motion posture of the user.

[0064] In particular, a plurality of wave form configurations are pre-stored to compare the wave form in response to the resultant acceleration via the comparison module 40. FIG. 4 illustrates the wave form of the activity data representing running activity via a front soles portion of the user stepping on a ground surface. FIG. 6 illustrates the wave form of the activity data representing running activity via a full soles portion of the user stepping on the ground surface. FIG. 7 illustrates the wave form of the activity data representing running activity via a heel portion of the user stepping on the ground surface. Therefore, by comparing the wave form, the present invention is able to distinguish the running posture of the user. FIGS. 4 and 5 illustrate the wave form of the activity data representing on-the-stairs activity and down-the-stairs activity respectively. In other words, through the comparison of the wave form, the present invention is able to distinguish the up-the-stairs posture or down-the-stairs posture of the user. FIG. 8 illustrates the wave form of the activity data representing jogging activity, therefore, the present invention is able to distinguish the jogging posture of the user.

[0065] According to the preferred embodiment, the pedometer 10 of the present invention can be wirelessly linked to an electronic device E (as shown in FIGS. 11 and 12) such as mobile phone, a portable pedometer processing device, a tablet, or a computer via a wireless link, as shown in FIG. 10. The wireless link can be “WiFi”, “Bluetooth”, NFC, or other wireless connection. Therefore, the accelerating data can be processed in the pedometer 10, wherein the resulted data will then be wirelessly transmitted to the electronic device E.

[0066] Alternatively, the accelerating data from the accelerometer 20 can be directly transmitted to the electronic device E, wherein the accelerating data can be processed in the electronic device. Accordingly, the smoothing filter 31, the Kalman filter 32, and the comparison module 40 are provided in the electronic device E, as shown in FIG. 10A. Therefore, the accelerating data can be processed in a processing unit of the electronic device E. In the embodiment and its alternative mode, the resulted data will be displayed by a screen of the electronic device.

[0067] Referring to FIGS. 11 and 12, the pedometer 10 is preferred to be carried by a shoe S to be worn by the user, wherein the pedometer 10 is preferred to be installed at a position of the shoe S where no substantial bending or deforming of the shoe while walking, jogging and running activities of the user. For example, as shown in FIG. 11, a receiving cavity S11 is provided at a middle portion of a shoe sole S1 of the shoe S for receiving the pedometer 10 therein. Alternatively, as shown in FIG. 12, a container S22 is attached to the shoe string S21 of the shoe body S2 coupled on top of the shoe sole S1 of the shoe S, and the pedometer 10 is received in the container S22.

[0068] It is appreciated that when the user is a child wearing the shoe S, the pedometer 10 of the present invention not only can determine the activity of the child that whether he or she is walking, jogging and running, but also can record the activity of the child user for health monitoring and exercise planning by the parents. If a positioning device is further included in the pedometer 10, the parents may even track the location of the child user for security purpose.

[0069] In addition, when the user, who wears the shoe S carrying the pedometer 10 of the present invention, is an old man or woman or a patient required attention by the physician or his or her family member, the pedometer 10 may also installed with a positioning device and a communication device, wherein activities of the aged user or patient user can be monitored and reported by the pedometer 10 to the physician or family member who have the electronic device E communicating with the pedometer 10. The activity of walking, jogging and running of the aged user or patient user can be recorded and determined whether it is normal or whether there may be any risk to the aged user or patient user going to happen for physical education and safety purposes.

[0070] One skilled in the art will understand that the embodiment of the present invention as shown in the drawings and described above is exemplary only and not intended to be limiting.

[0071] It will thus be seen that the objects of the present invention have been fully and effectively accomplished. The embodiments have been shown and described for the purposes of illustrating the functional and structural principles of the present invention and is subject to change without departure from such principles. Therefore, this invention includes all modifications encompassed within the spirit and scope of the following claims.