WO2017061639A1

WO2017061639A1 - User context based motion counting method, sensor device and wearable device performing same

Info

Publication number: WO2017061639A1
Application number: PCT/KR2015/010532
Authority: WO
Inventors: 보글마이클; 권성호; 이종성
Original assignee: 주식회사 스탠딩에그
Priority date: 2015-10-06
Filing date: 2015-10-06
Publication date: 2017-04-13

Abstract

A user context based motion counting method, a sensor device and a wearable device performing the same are provided. The user context based motion counting method comprises the steps of: extracting at least one feature from sensor data; identifying a user context on the basis of the extracted feature; determining a motion counting algorithm on the basis of the identified user context; and counting a motion of the user on the basis of the sensor data using the determined motion counting algorithm.

Description

User context based motion counting method, sensor device and wearable device performing the same

The present invention relates to a method of counting a user's motion, and more particularly, to a user context-based motion counting method, a sensor device for performing the same, and a wearable device.

A wearable device represents a computer system provided in a form that can be worn on a user's body. The wearable device may measure a user's movement using various sensors and provide a measurement result to the user. The wearable device may manually input the user's exercise motion and count the number of times the received motion motion is repeated. In addition, the wearable device may monitor the movement of the user at a specific cycle to provide additional information such as calorie consumption and activity time.

An object of the present invention is to provide a user context-based motion counting method that can more accurately count a user's motion.

Another technical problem of the present invention is to provide a user context-based motion counting method that can count a user's motion more intuitively and conveniently.

Another technical problem of the present invention is to provide a sensor device and a wearable device for performing the user context based motion counting method.

The technical problems of the present invention are not limited to the above-mentioned problems, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method of counting motion based on a user context, extracting at least one feature from sensor data, and generating a user context based on the extracted feature. Identifying, determining a motion counting algorithm based on the identified user context, and counting a user's motion based on the sensor data using the determined motion counting algorithm.

In some embodiments of the present invention, the extracted feature may include at least one statistical value of the sensor data.

In some embodiments of the present disclosure, identifying the user context may identify a user context based on the extracted feature using a K-Nearest Neighborhood (KNN) algorithm.

In some embodiments of the invention, determining the motion counting algorithm comprises determining an autocorrelation parameter of the motion counting algorithm based on the identified user context, wherein the first autocorrelation parameter is determined for a first user context. The parameter may be determined, and a second autocorrelation parameter may be determined for the second user context.

In addition, the autocorrelation parameter may correspond to a repetition period of a peak included in the sensor data.

In some embodiments of the invention, determining the motion counting algorithm comprises: determining a first motion counting algorithm for a first user context based on the identified user context, and a second motion for a second user context The counting algorithm can be determined.

In some embodiments of the present invention, the motion counting algorithm includes filtering of sensor data of a predetermined frequency range of the sensor data, and determining the motion counting algorithm is based on the identified user context. The filtering setting may be determined, but the first filtering setting may be determined for the first user context, and the second filtering setting may be determined for the second user context.

In some embodiments of the present invention, the method may further comprise acquiring the sensor data from an acceleration sensor or a gyro sensor.

The sensor device according to another aspect of the present invention for solving the above technical problem, may perform any one of the above-described user context-based motion counting method.

The wearable device according to another aspect of the present invention for solving the above technical problem may perform any one of the above-described user context based motion counting method.

Other specific details of the invention are included in the detailed description and drawings.

According to the user context-based motion counting method of the present invention, since the user context is identified and a more suitable motion counting algorithm is determined for each identified user context, the motion of the user can be counted more accurately.

In addition, in order to identify a user context, since a user's direct input or selection is not required separately, more intuitive and user convenience may be enhanced.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a flowchart schematically illustrating a user context based motion counting method according to an embodiment of the present invention.

2 is a diagram schematically illustrating identifying a user context using the KNN algorithm.

3 is a diagram schematically illustrating determining autocorrelation parameters based on a user context.

4 is a diagram schematically illustrating determining filtering settings for sensor data based on a user context.

5 is a block diagram schematically illustrating a wearable device for performing a user context based motion counting method according to an embodiment of the present invention.

FIG. 6 is a diagram schematically illustrating a state in which the wearable device of FIG. 5 is worn on a user's body.

7 is a block diagram schematically illustrating a wearable device for performing a user context based motion counting method according to another embodiment of the present invention.

8 is a block diagram schematically illustrating a sensor hub for performing a user context based motion counting method according to an embodiment of the present invention.

9 is a block diagram schematically illustrating a sensor hub for performing a user context based motion counting method according to another embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be embodied in various different forms, and the present embodiments merely make the disclosure of the present invention complete, and are common in the art to which the present invention pertains. It is provided to fully inform those skilled in the art of the scope of the invention, which is defined only by the scope of the claims. Like reference numerals refer to like elements throughout.

Although the first, second, etc. are used to describe various elements, components and / or sections, these elements, components and / or sections are of course not limited by these terms. These terms are only used to distinguish one element, component or section from another element, component or section. Therefore, the first device, the first component, or the first section mentioned below may be a second device, a second component, or a second section within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used in a sense that can be commonly understood by those skilled in the art. In addition, terms that are defined in a commonly used dictionary are not ideally or excessively interpreted unless they are specifically defined clearly.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, "comprises" and / or "comprising" does not exclude the presence or addition of one or more other components in addition to the mentioned components.

Referring to FIG. 1, in step S10, sensor data is obtained from an acceleration sensor, a gyro sensor, or at least one sensor coupled thereto. Sensor data can be acceleration data in three axis directions (e.g., X axis, Y axis, Z axis) or three axis directions (e.g., Pitch axis, Yaw axis, Roll axis) The angular velocity data may include, but is not limited thereto. According to an embodiment, a process such as predetermined sampling or quantization of sensor data may be provided.

Subsequently, at step S20, at least one feature is extracted from the sensor data. A characteristic represents an aspect or component in which sensor data is distinguished from other sensor data. The feature may represent an entire portion of the sensor data or may represent a portion of the sensor data. For example, the feature may include at least one statistical value of sensor data. For example, the statistical value may include a maximum value, a minimum value, a median value, an average value, an interquartile range (IQR) value, a root mean square (RMS) value, and the like of a predetermined number of (sampled) sensor data. However, the present invention is not limited thereto. The feature may be extracted by any processing on the sensor data.

Then, in step S30, the user context is identified based on the extracted feature. The user context represents a situation of the user or information defining the same. The user context may be associated with the type of action that may specify the movement of the user. For example, the user context may be classified into various movement operations such as push up, biceps curls, and rowing, but is not limited thereto.

For example, the user context may be identified using the KNN algorithm. 2 is a diagram schematically illustrating identifying a user context using the KNN algorithm. 2, a plurality of features are processed using the KNN algorithm. By the KNN algorithm, a plurality of features may be compared to previously classified or learned features and classified into groups having the same or similar trends. By combining the classification results, a specific user context can be derived. Settings regarding the k value and the distance between neighboring features may be variously adjusted according to embodiments. Although not explicitly illustrated, weights of high importance, distance, and the like for some features may be applied. A detailed description of the KNN algorithm will be omitted since it may obscure the subject matter of the present invention.

On the other hand, any machine learning algorithm that is not illustrated may be used for identification of the user context.

Then, in step S40, a motion counting algorithm is determined based on the identified user context. At this time, autocorrelation parameters of the motion counting algorithm may be determined together based on the identified user context. The autocorrelation parameter may be determined differently according to the user context. That is, the first autocorrelation parameter may be determined for the first user context, and the second autocorrelation parameter may be determined for the second user context. The autocorrelation parameter may be used to remove noise or counting errors within the motion counting algorithm.

For example, the autocorrelation parameter may correspond to the repetition period of the peaks included in the sensor data. The peak may be a factor or factor for counting the user's movement as described below. 3 is a diagram schematically illustrating determining autocorrelation parameters based on a user context. Referring to FIG. 3, a plurality of peaks p0 to p2 may be included in the sensor data of a predetermined time range. Peaks can be selected based on their amplitude and the time interval between previous peaks. On the other hand, the repeating pattern of the peak may vary depending on the user context (user's motion). For example, if a user performs a push up operation, and the repetition period of the peak, that is, the time interval between the previous peak and the current peak is 100 msec, considering the repetition pattern of the general peak appearing in the push up operation However, the current peak is presumed to be due to noise or error and is not used for counting. In the example shown in FIG. 3, if the reference time interval for the repetition period of the peak is tr1, then p1 may be selected as the peak for counting, but if the reference time interval is tr2, p1 next to p0 may be selected. Ignored and p2 may be selected as the peak for counting.

At least some motion counting algorithms may include filtering for sensor data in a predetermined frequency range of the sensor data to remove noise or counting errors. And, when the motion counting algorithm is determined, filtering settings may be determined together based on the identified user context. Filtering settings may be determined differently according to the user context. That is, the first filtering setting may be determined for the first user context, and the second filtering setting may be determined for the second user context.

4 is a diagram schematically illustrating determining filtering settings for sensor data based on a user context. Referring to FIG. 4, a plurality of filtering settings that define a predetermined pass band is shown. In the example shown in FIG. 4, it is confirmed that the pass bands of fl1 and fh1 are set to cutoff frequencies according to the first filtering setting, and when fl2 and fh2 are set to cutoff frequencies according to the second filtering setting. Can be. In FIG. 4, the band pass filter is illustrated as an example, but is not limited thereto, and the filtering setting may be provided in a substantially same manner for the high pass filter and the low pass filter.

On the other hand, a separate motion counting algorithm can be determined based on the identified user context (in the case of a given motion moving in a completely different way than other actions). In this case, a first motion counting algorithm may be determined for the first user context, and a second motion counting algorithm may be determined for the second user context.

Subsequently, in step S50, the movement of the user is counted based on the sensor data using the determined motion counting algorithm. For example, the sensor data of the axis having the largest change among the sensor data of the three axes may be selected, and the peak in the sensor data of the selected axis may be counted. The method of counting movements may be variously modified according to specific embodiments.

Referring to FIG. 5, the wearable device 100 may include an acceleration sensor 110, a gyro sensor 120, a storage unit 130, an input unit 140, an output unit 150, a controller 160, and a power supply unit 170. ).

The acceleration sensor 110 detects acceleration in three axis directions (for example, X axis, Y axis, and Z axis). Although only one acceleration sensor is illustrated in FIG. 5, the present invention is not limited thereto, and a plurality of acceleration sensors capable of detecting acceleration in one axial direction may be provided. The acceleration sensor 110 may transmit acceleration data in three axes to the controller 160.

The gyro sensor 120 detects angular velocities in three axis directions (for example, a pitch axis, a yaw axis, and a roll axis). Although only one gyro sensor is illustrated in FIG. 5, the present invention is not limited thereto, and a plurality of gyro sensors capable of detecting an angular velocity in one axial direction may be provided. The gyro sensor 120 may transmit angular velocity data in three axes to the controller 160.

The storage unit 130 stores various data and commands. The storage unit 130 may store various software modules including system software for operating the wearable device 100 and applications in which a user context-based motion counting method according to an embodiment of the present invention is implemented. The storage unit 130 may be a random access memory (RAM), read only memory (ROM), erasable-programmable ROM (EPROM), electrically EPROM (EEPROM), flash memory, a removable disk, or well known in the art. Any type of computer readable recording medium may be included.

The input unit 140 receives various information from the user. The input unit 140 may include various input means such as keys, buttons, switches, wheels, and touch pads.

The output unit 150 notifies the user of various kinds of information. The output unit 150 may output information in the form of text, video or audio. To this end, the output unit 150 may include a display module 151 and a speaker module 152. The display module 151 may be a liquid crystal display (LCD), a thin film transistor (TFT) LCD, an organic light emitting diode (OLED), a flexible display, a three-dimensional display, an electronic ink display, or a technique well known in the art. It may be provided in any form.

The controller 160 controls other components to control the overall operation of the wearable device 100. The controller 160 may perform various software modules including system software for operating the wearable device 100 and applications in which a user context-based motion counting method according to an embodiment of the present invention is implemented. The controller 160 may filter the sensor data transmitted from the sensors 110 and 120, including the high pass filter 161, the low pass filter 162, and the band pass filter 163. The controller 160 may filter the sensor data by selecting at least one filter among the plurality of filters 161 ˜ 163. Meanwhile, the high pass filter 161, the low pass filter 162, and the band pass filter 163 may be provided as independent components to the outside of the controller 160 or may be internal components of the respective sensors 110 and 120. May be provided.

The power supply unit 170 supplies power required for the operation of the acceleration sensor 110, the gyro sensor 120, the storage 130, the input unit 140, the output unit 150, and the controller 160. The power supply unit 170 may include a built-in battery or convert power supplied from the outside into a power suitable for the above components.

Meanwhile, since the components illustrated in FIG. 5 are not essential, the wearable device 100 may be modified to include more components or fewer components.

Referring to FIG. 6, the wearable device 100 described with reference to FIG. 5 may be worn on a user's body. For example, the wearable device 100 may be worn on an extremity such as an arm or a leg of the user, or may be worn on the user's torso, but is not limited thereto, and may include other parts (head, hands, feet, etc.) It may be worn on at least part of the user's body. Alternatively, the wearable device 100 may be provided inside any item (hat, gloves, shoes, etc.) worn on the user's body.

7 is a block diagram schematically illustrating a wearable device for performing a user context based motion counting method according to another embodiment of the present invention. For convenience of description, duplicated descriptions of components that are the same as the wearable device 100 described with reference to FIG. 5 will be omitted.

Referring to FIG. 7, the wearable device 200 further includes a wireless communication unit 230 and a vibrator 260 as compared with the wearable device 100 described with reference to FIG. 5.

The wireless communication unit 230 may wirelessly communicate with an external device (such as a server or a user terminal). The wireless communication unit 230 may wirelessly communicate with an external device by using a wireless communication method such as mobile communication, WiBro, Wi-Fi, Bluetooth, Zigbee, ultrasonic wave, infrared ray, or RF (Radio Frequency). have. The wireless communication unit 230 may transmit data and / or information received from the external device to the controller 280, and may transmit data and / or information transmitted from the controller 280 to the external device. To this end, the wireless communication unit 230 may include a mobile communication module, a short-range communication module and the like.

The vibrator 260 may perform a vibration notification for notifying the user of various kinds of information.

The wearable device 200 may transmit / receive various types of information such as sensor data, a user context, a motion counting algorithm, and a motion counting result with various servers or user devices (not shown).

The wearable device according to the embodiment of the present invention may be provided as any computer system wearable on the body of the user, which is not illustrated.

Referring to FIG. 8, the sensor hub 1000 may include a processing device 1100, a MEMS device 1200, and an application specific integrated circuit (ASIC) device 1300. The MEMS device 1200 may be an acceleration sensor or a gyro sensor, but is not limited thereto. The ASIC device 1300 may process the sensing signal of the MEMS device 1200. The processing device 1100 may function as a coprocessor for professionally performing sensor data processing on behalf of the application processor. The processing device 1100 may execute various software modules including an application in which a user context-based motion counting method according to an embodiment of the present invention is implemented.

Referring to FIG. 9, the sensor hub 2000 may include a plurality of

MEMS devices

2200 and 2400 and a plurality of ASIC devices 2300 and 2500. The first MEMS device 2200 may be an acceleration sensor, and the second MEMS device 2400 may be a gyro sensor, but is not limited thereto. The plurality of ASIC devices 2300 and 2500 may process sensing signals of the

corresponding MEMS devices

2200 and 2400, respectively. The processing device 2100 may function as a coprocessor for professionally performing sensor data processing on behalf of the application processor. The processing device 2100 may execute various software modules including an application in which the user context-based motion counting method according to an embodiment of the present invention is implemented. Unlike shown, three or more MEMS devices and ASIC devices may be provided within the sensor hub 2000.

The method described in connection with an embodiment of the present invention may be implemented as a software module performed by a processor. The software module may reside in RAM, ROM, EPROM, EEPROM, flash memory, hard disk, removable disk, CD-ROM, or any form of computer readable recording medium well known in the art. .

Although the embodiments of the present invention have been described above with reference to the accompanying drawings, those skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing the technical spirit or essential features thereof. You will understand that. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive.

Claims

Extracting at least one feature from the sensor data;

Identifying a user context based on the extracted feature;

Determining a motion counting algorithm based on the identified user context; And

Counting a user's motion based on the sensor data, using the determined motion counting algorithm.
The method of claim 1,

The extracted feature is,

And at least one statistical value of the sensor data.
The method of claim 1,

Identifying the user context,

A user context based motion counting method for identifying a user context based on the extracted feature using a K-Nearest Neighborhood (KNN) algorithm.
The method of claim 1,

Determining the motion counting algorithm,

Determine an autocorrelation parameter of a motion counting algorithm based on the identified user context, determine a first autocorrelation parameter for a first user context, and determine a second autocorrelation parameter for a second user context A user context based motion counting method.
The method of claim 4, wherein

The autocorrelation parameter is

A user context based motion counting method corresponding to a repetition period of a peak included in the sensor data.
The method of claim 1,

Determining the motion counting algorithm,

Determine a first motion counting algorithm for a first user context based on the identified user context, and determine a second motion counting algorithm for a second user context.
The method of claim 1,

The motion counting algorithm,

Filtering the sensor data of a predetermined frequency range among the sensor data;

Determining the motion counting algorithm,

Determining the filtering setting based on the identified user context, determining a first filtering setting for a first user context, and determining a second filtering setting for a second user context.
The method of claim 1,

And acquiring the sensor data from an acceleration sensor or a gyro sensor.
A sensor device for performing the user context based motion counting method of any one of claims 1 to 8.
The wearable device of claim 1, wherein the wearable device performs the user context based motion counting method of claim 1.