Sensor-Based Athletic Activity Measurements

• US 20160354635A1
• Filed: 08/19/2016
• Published: 12/08/2016
• Est. Priority Date: 07/01/2011
• Status: Active Grant

• Current Assignee: Nike, Inc.
• Original Assignee: Nike, Inc.
• Inventors: Weast, Aaron B., Mullin, James M.
• Granted Patent: US 10,265,579 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61B 2503/10 Athletes
• CPC Class Codes: A61B 5/1036 Measuring load distribution, e.g. podologic studies
• CPC Class Codes: A61B 5/112 Gait analysis
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/486 Bio-feedback A61B5/375 takes precedence
• CPC Class Codes: A61B 5/6807 Footwear
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A63B 2024/0065 Evaluating the fitness, e.g. fitness level or fitness index
• CPC Class Codes: A63B 2220/20 Distances or displacements
• CPC Class Codes: A63B 2220/836 Sensors arranged on the body of the user
• CPC Class Codes: A63B 2230/75 calorie expenditure
• CPC Class Codes: A63B 24/0062 Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance

Abstract

Determining pace or speed based on sensor data may include determining an amount of contact time a user'"'"'s foot has with a workout surface such as the ground. Contact time may be determined by identifying samples in the sensor data that correspond to various events such as a heelstrike, a toe-off and a subsequent heelstrike. In one example, these events may be identified by determining a sequence of three sample values (e.g., a triplet) that exceeds corresponding thresholds. The validity of an identified triplet (e.g., heelstrike, toe-off and heelstrike) may be confirmed by determining whether a difference between a last event sample and a middle event sample is greater than a difference between the middle event sample and an initial event sample. Once confirmed, a contact time may be determined from the triplet. A linear or non-linear relationship may then be applied to the contact time to determine a speed or pace.

Claims

1. An athletic performance monitoring method comprising:

receiving foot-based sensor data during an athletic activity performed by a user;

processing the foot-based sensor data;

determining, based on the processing, a sequence of the foot-based sensor data that corresponds to a set of at least two consecutive heelstrike samples; and

determining a performance metric of the user based on the set of at least two consecutive heelstrike samples.

2. The athletic performance monitoring method of claim 1, wherein processing the foot-based data includes filtering a predetermined number of rolling standard deviations away from a rolling mean.

3. The athletic performance monitoring method of claim 1, wherein processing the foot-based data includes applying a fast fourier transform, and wherein determining the set of at least two consecutive heelstrike samples includes identifying locations of relative peak values in fast fourier transform data.

4. The athletic performance monitoring method of claim 1, wherein determining the performance metric includes determining a time between two consecutive heelstrike samples and an amplitude of the two consecutive heelstrike samples.

5. The athletic performance monitoring method of claim 1, wherein the performance metric is one of:

pace, distance, speed, and acceleration.

6. The athletic performance monitoring method of claim 1, further comprising determining a pace of the user based on an elapsed time between the two consecutive heelstrike samples.

7. The athletic performance monitoring method of claim 1, further comprising:

determining, based on the processing, a toe-off sample between the two consecutive heelstrike samples,

wherein the performance metric of the user is determined based on the two consecutive heelstrike samples and the toe-off sample.

8. The athletic performance monitoring method of claim 7, further comprising:

determining a first difference between a value of a second heelstrike sample and a value of the toe-off sample;

determining a second difference between a value of the toe-off sample and the value of a first heelstrike sample;

determining whether the first difference is greater than the second difference; and

in response to determining that the first difference is greater than the second difference, adding a sequence of three samples to a list.

9. The athletic performance monitoring method of claim 8, further comprising:

adding a plurality of three sample sequences to the list; and

determining a plurality of contact times of the user based on the plurality of three sample sequences in the list.

10. The athletic performance monitoring method of claim 9, further comprising:

determining a mean contact time of the plurality of contact times;

determining a standard deviation between the plurality of contact times; and

filtering out one or more contact times of the plurality of contact times that exceed a threshold number of standard deviations from the mean contact time.

11. The athletic performance monitoring method of claim 1, further comprising:

selecting a predefined linear relationship from a plurality of linear relationships, wherein each of the plurality of linear relationships corresponds to a different type of athletic activity; and

applying the predefined linear relationship to the sensor data to calculate the performance of the user.

12. The athletic performance monitoring method of claim 1, wherein determining the sequence of the foot-based sensor data that corresponds to a set of at least two consecutive heelstrike samples includes:

selecting a first sample from a plurality of samples;

determining whether a value of the first sample is at or above a first threshold sample value;

in response to determining that the value of the first sample is not at or above the first threshold sample value, selecting a subsequent sample in the plurality of samples; and

in response to determining that the value of the first sample is at or above the first threshold sample value;

determining a maximum sample value in a first sample window of a predefined size, wherein a lower bound of the first sample window corresponds to the first sample, and

identifying a sample corresponding to the maximum sample value as the first heelstrike sample.

13. The athletic performance monitoring method of claim 12, wherein in response to determining that the value of the first sample is at or above the threshold sample value, the method further comprises:

determining a maximum sample value in a second sample window; and

determining whether the maximum sample value of the second sample window is at or above a second threshold sample value; and

in response to determining that the maximum sample value of the second sample window is at or above the second threshold sample value, identifying a sample corresponding to the maximum sample value of the second sample window as a toe-off sample.

14. An athletic performance monitoring apparatus comprising:

a foot-based sensor; and

at least one processor configured to;

receive a signal generated by the foot-based sensor during athletic activity performed by a user, wherein the signal comprises a plurality of samples;

process the signal;

determine, based on the processing, a sequence of the foot-based sensor signal that corresponds to a set of at least two consecutive heelstrike samples; and

determine a performance metric of the user based on the set of at least two consecutive heelstrike samples.

15. The athletic performance monitoring apparatus of claim 14, wherein the processor is further configured to:

determine, based on the processing, a toe-off sample between the two consecutive heelstrike samples,

wherein the performance metric of the user is determined based on the two consecutive heelstrike samples and the toe-off sample.

16. The athletic performance monitoring apparatus of claim 15, wherein the processor is further configured to:

identify a plurality of three sample sequences, each of the three sample sequences including a first heelstrike sample, a toe-off sample and a second heelstrike sample;

add the plurality of three sample sequences to a list; and

determine a plurality of contact times of the user based on the plurality of three sample sequences in the list.

17. The athletic performance monitoring apparatus of claim 16, wherein the processor is further configured to:

determine a mean contact time of the plurality of contact times;

determine a standard deviation between the plurality of contact times; and

filter out one or more contact times of the plurality of contact times that exceed a threshold number of standard deviations from the mean contact time.

18. An athletic performance monitoring method comprising:

receiving foot-based sensor data during an athletic activity performed by a user;

processing the foot-based sensor data;

determining, based on the processing, a sequence of the foot-based sensor data that corresponds to a set of a heelstrike sample and a toe-off sample;

determining a contact time by determining a difference between the toe-off sample and the heelstrike sample; and

applying a predefined linear relationship to the contact time to calculate a pace of the user.

19. The athletic performance monitoring method of claim 18, wherein processing the foot-based data includes filtering a predetermined number of rolling standard deviations away from a rolling mean.

20. The athletic performance monitoring method of claim 18, wherein applying the predefined linear relationship includes selecting a predefined linear relationship from a plurality of linear relationships, wherein each of the plurality of linear relationships corresponds to a different type of athletic activity.