Method for determining activity state

Abstract

This method for determining an activity state comprises: deriving acceleration values based on a reference coordinate system from an acceleration sensor; regarding a case where the maximum value among the magnitudes of acceleration based on the X-, Y-, and Z-axis directions constituting each acceleration value is maintained at 1.2 g or less for a reference time as a static state, and determining a start posture for dynamic activity; receiving angular velocity values based on the reference coordinate system from a gyro sensor to generate a movement signal related to walking information of a subject; when there is no movement signal for a predetermined time or while a movement signal is being generated, receiving three-axis angular velocity values based on the reference coordinate system from the gyro sensor to track angle change values for the torso of the subject; and determining the exercise type of the subject using the angle change values.

Description

How to determine activity status

[0001] Embodiments of the present invention relate to a method for determining an activity state. More specifically, embodiments of the present invention relate to a method for determining an activity state, which can determine a body's activity state, for example, the type of exercise, by measuring dynamic posture.

[0002] Dynamic postural measurement is a technique for measuring changes in the human body's posture during exercise or daily life. This dynamic postural measurement analyzes postural changes over time and is used to assess balance, motor skills, and postural control abilities.

[0003] For example, data obtained through the above dynamic posture measurement can be used to analyze athletes' posture and movements, evaluate exercise effectiveness, develop rehabilitation programs, etc., and in the medical field, it can be used for diagnosing balance disorders, correcting posture, and rehabilitating treatment. Furthermore, in the industrial field, the data can be used for analyzing workers' postures, improving work environments, and designing ergonomic products.

[0004] The need for dynamic posture measurement is growing, particularly in the gaming and entertainment sectors, where content that connects virtual and real environments is crucial. Wearable devices, in particular, enable continuous dynamic posture measurement during daily life, making them ideal for assessing an individual's health and daily calorie consumption.

[0005] Meanwhile, the latest wearable devices include a 9-axis inertial measurement unit (IMU), such as a 3-axis accelerometer sensor, a 3-axis gyroscope, and a 3-axis magnetometer sensor.

[0006] The above acceleration sensor can measure the number of steps, running speed, jump height, etc. However, since the acceleration sensor is continuously affected by gravity along the direction of gravity perpendicular to the surface of the Earth, it does not provide information about horizontal movement parallel to the direction of gravity.

[0007] For example, when the human body rotates, an accelerometer sensor, like a gyroscope sensor that measures angular velocity, only measures temporary changes in acceleration due to the movement. Therefore, rotation in the transverse plane of the body, which is horizontal to the ground, does not cause a change in gravitational acceleration, and thus cannot be measured by an accelerometer. Consequently, many dynamic situations, such as tennis, table tennis, golf, and boxing, involve rotation of the waist in a transverse plane parallel to the ground, and an accelerometer cannot provide information about this.

[0008] The above gyroscope is a sensor that measures the rotational angular velocity in three axes. Through this, it can detect changes in the direction of the human body and changes in the body posture. A three-axis gyroscope can obtain angular velocities ωx, ωy, and ωz about the x, y, and z axes. Therefore, by integrating the angular velocities, the rotation angles θx, θy, and θz in the x, y, and z directions can be obtained as follows.

[0009] θx = ∫ωx dt, θy = ∫ωy dt, θz = ∫ωz dt

[0010] If the rotation angle is known, the rotation information of the human body from the initial posture to the current posture can be derived based on the rotation angles θx, θy, and θz of the rotation matrix for each axis.

[0011] However, when measuring human posture using a gyro sensor, there is a problem in that it is impossible to accurately determine the current posture at the time of measurement because there is no information about the initial posture when the measurement began.

[0012] That is, when estimating human posture using data acquired from a gyro sensor, accumulated errors may occur due to noise, and a Kalman filter is used to correct this.

[0013] The Kalman filter is an algorithm that optimally estimates noisy measurements based on a state-space model. It consists of a prediction stage and an update stage. The Kalman gain is used to weight the measured and predicted values, thereby compensating for errors. Appropriately setting the initial state and error covariance can prevent long convergence times or divergence. However, in real-world situations, it's difficult to determine the initial attitude, making it difficult for the Kalman filter or gyro sensor to set initial values ​​for an appropriate initial state.

[0014] Furthermore, the state equations and measurement equations used in Kalman filters are simplified models of real-world systems. Therefore, when the error between the model and the real system is large, estimation performance deteriorates. In particular, modeling errors increase when nonlinearity and model uncertainty are present.

[0015] Furthermore, error covariances arising from measurement systems are generally assumed to be constant over time. However, in reality, they can vary over time, requiring an estimation of error covariances that are appropriate for the situation. Gyroscopes experience drift over time due to accumulated bias errors, a problem that cannot be avoided even with interval-based Kalman filters.

[0016] In summary, in order to perform real-time attitude estimation using a gyro sensor, Kalman prediction, update operation, and continuous correction for errors must be performed at set time intervals, which increases the computational load of the gyro sensor.

[0017] Even the MEMS sensors that make up a gyroscope measure rotation by continuously physically driving a micrometer-sized yoke that vibrates in the x, y, and z directions. Therefore, it is inevitable that the sensor will consume significant power while measuring the amount of rotation.

[0018] Furthermore, because the change in rotation is measured based on integration, the constant CPU operation consumes a lot of power. In particular, the reason why wearable devices with limited battery capacity do not continuously measure the change in posture using a gyro sensor is because the starting position of the measurement obtained through integration by driving the gyroscope cannot be known, resulting in a decrease in measurement value. Furthermore, the sensor and CPU operation consume significant power.

[0019] Meanwhile, geomagnetic sensors are not commonly used to measure dynamic changes in human posture.

[0020] This is because internal body movements and muscle activity can alter the surrounding magnetic field, affecting sensor measurements. Furthermore, the presence of metallic objects or electronic devices nearby can distort the magnetic field, making accurate position measurements difficult. In particular, the strength and direction of the magnetic field of geomagnetic sensors can vary significantly depending on the usage environment and location on the Earth, making accurate attitude measurements difficult. Due to these characteristics, the output of geomagnetic sensors is known to be significantly less consistent than that of accelerometers or gyroscopes.

[0021] Therefore, there is a problem that it is difficult to accurately estimate posture with only a geomagnetic sensor for rapid posture changes or fast movements. Therefore, in a dynamic posture measurement system, it is necessary to use a geomagnetic sensor not alone, but together with sensors such as an accelerometer and gyroscope to estimate posture more accurately.

[0022] Meanwhile, Korean Patent Registration No. 10-2556675 is known as prior art in the related field.

[0023] Figure 1 is a drawing defining the coronal, sagittal, and transverse planes based on the human body.

[0024] Referring to Figure 1, the coronal plane (or frontal plane - a cross-section plane that divides the front and back of the body, also known as the frontal plane), the saggital plane (a cross-section plane that divides the body symmetrically into the right and left sides), and the transverse plane (a cross-section plane that divides the body into the top and bottom) are defined based on the human body.

[0025] Figure 2 is a graph showing the trend of acceleration values ​​over time for each ball game using a conventional acceleration sensor.

[0026] Referring to Figure 2, when tracking changes in acceleration values ​​for each type of exercise using an acceleration sensor included in a wearable device, it is difficult to distinguish between the types of exercise because the differences in frequency and amplitude for each type of exercise are not clear.

[0027] Figure 3 is a graph showing the trend of acceleration values ​​that change as the type of exercise is changed using a conventional acceleration sensor.

[0028] Referring to Figure 3, it can be seen that the relative difference in exercise intensity can be simply confirmed through the change in acceleration value measured by the acceleration sensor.

[0029] Embodiments of the present invention provide a method for determining an activity state, which can determine an activity state of a body by determining an initial posture using an acceleration sensor and measuring a dynamic posture that changes from the initial posture using a gyro sensor and an acceleration sensor.

[0030] In a method for determining an activity state using an electronic device including an acceleration sensor and a gyro sensor that can be fixed to a torso area of ​​a subject according to embodiments of the present invention and have the same local coordinate system, a reference coordinate system is defined by a microprocessor as an X-axis direction extending from the right shoulder to the left shoulder when the subject is standing normally, a Y-axis direction opposite to gravity, and a Z-axis direction generated by a vector multiplication between the X-axis direction and the Y-axis, and from the acceleration sensor, the microprocessor derives an acceleration value according to the reference coordinate system, and when the maximum value of the magnitudes of the acceleration values ​​according to each of the X, Y, and Z-axis directions constituting each of the acceleration values ​​is maintained at 1.2 g or less for a reference time, the microprocessor determines a starting posture for dynamic activity, and when the starting posture is determined, the microprocessor receives an angular velocity value according to the reference coordinate system from the gyro sensor, and generates a movement signal regarding gait information of the subject, and when the movement signal is absent for a certain period of time or while the movement signal is being generated, The microprocessor receives a three-axis angular velocity value according to the reference coordinate system from the gyro sensor, tracks the angular change value of the subject's torso, and determines the type of movement of the subject using the angular change value.

[0031] In one embodiment of the present invention, in order to determine the starting posture, the angle formed by the three-axis acceleration vector with respect to the reference coordinate system and the Z-axis direction can be calculated.

[0032] In one embodiment of the present invention, the step of determining the starting posture may determine the posture of the subject as standing, semi-lateral, sitting, (semi) Fowler's position, supine, or prone.

[0033] In one embodiment of the present invention, the gait information of the subject can be defined as a gait unit by the transition between (+) / (-) of the angular velocity value based on the Y-axis.

[0034] Here, the walking information of the subject can be used to check the walking characteristics of the subject by using the time difference between the (+) / (-) of the angular velocity value based on the Y-axis.

[0035] In addition, the electronic device is additionally equipped with a geomagnetic sensor, and the point in time when the magnetic field value has a peak value in the direction having the smallest angle with the direction of the Earth's magnetic field lines among the directions of the geomagnetic sensor can be determined as a change in the walking direction.

[0036] In one embodiment of the present invention, in order to track the angular change value of the body part of the subject,

[0037] The angle formed by the three-axis acceleration vector with respect to the above reference coordinate system with respect to the Z-axis direction defines the amount of rotation on the sagittal plane, the angle formed by the three-axis acceleration vector with respect to the above reference coordinate system with respect to the X-axis direction defines the amount of rotation on the coronal plane, and when the size of the peak of the three-axis angular velocity with respect to the above reference coordinate system is 2 / π or greater, the amount of rotation on the cross-section of the torso of the subject can be defined using the value of the peak and the convergence time value for the peak to converge to 0.

[0038] Here, the angle change value may include the minimum and maximum values ​​of the rotation amount according to the sagittal plane and coronal plane, respectively, and the rotation amount of the cross-section.

[0039] In addition, the above gait information includes a cadence value, and the time from the end point of the starting posture until the rotation angle value on the cross-section of the torso becomes 0 is defined as the required time, and the step of determining the type of exercise of the subject using the angle change value can further use the cadence value and the required time.

[0040] Here, the step of determining the type of movement of the subject using the angle change value may be performed by displaying the minimum and maximum values ​​of the rotation amount according to each of the sagittal and coronal planes and the rotation amount of the cross-section on a virtual three-dimensional rectangular coordinate system having an x-axis, a y-axis, and a z-axis within the required time to form an initial rectangular parallelepiped.

[0041] In addition, a transformed rectangular parallelepiped is formed by moving each of the coordinate points constituting the vertices of the initial rectangular parallelepiped along a diagonal line connecting the vertices corresponding to the maximum and minimum values ​​of the initial rectangular parallelepiped by a size of (the time required × the average value of the cadence), and the transformed rectangular parallelepiped can be compared with the reference rectangular parallelepiped for each exercise type in terms of position.

[0042] According to the embodiments of the present invention described above, after specifying the initial posture of the subject, it is possible to determine the activity state by calculating the amount of rotation of the torso in the specified posture.

[0043] Figure 1 is a drawing defining the coronal, sagittal, and transverse planes based on the human body.

[0044] Figure 2 is a graph showing the trend of acceleration values ​​over time for each ball game using a conventional acceleration sensor.

[0045] Figure 3 is a graph showing the trend of acceleration values ​​that change as the type of exercise is changed using a conventional acceleration sensor.

[0046] FIG. 4 is a flowchart for explaining a method for determining an activity state using an electronic device according to embodiments of the present invention.

[0047] FIG. 5 is a drawing for explaining a step of correcting the local coordinate system of the sensor to derive an acceleration value according to the reference coordinate system of FIG. 4.

[0048] Figure 6 is a drawing for explaining the sagittal plane angle for each exercise type with respect to the starting posture of Figure 4.

[0049] Fig. 7 is a graph for explaining the measurement values ​​of the gyro sensor and the geomagnetic sensor in the step of generating the movement signal of Fig. 4.

[0050] Figure 8 is a graph for explaining the change in body angle of Figure 4 by exercise type.

[0051] Figure 9 is a graph illustrating a virtual three-dimensional rectangular coordinate system for explaining the steps of forming a transformed rectangular parallelepiped from an initial rectangular parallelepiped.

[0052] Figure 10 is a graph depicting a virtual three-dimensional rectangular coordinate system based on a standard rectangular solid for each sport.

[0053] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. The present invention can be modified in various ways and can take various forms, and thus specific embodiments are illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific disclosed form, but it should be understood that all modifications, equivalents, and substitutes included in the spirit and technical scope of the present invention are included. In describing each drawing, similar reference numerals are used to indicate similar components. In the attached drawings, the dimensions of structures are shown larger than actual size to ensure clarity of the present invention.

[0054] While terms like "first" and "second" may be used to describe various components, these components should not be limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, a first component could be referred to as a "second component," and similarly, a second component could also be referred to as a "first component."

[0055] The terminology used in this application is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0056] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0057] FIG. 4 is a flowchart for explaining a method for determining an activity state using an electronic device according to embodiments of the present invention.

[0058] Referring to FIG. 4, a method for determining activity status is provided using an electronic device that can be affixed to the torso of a subject and includes an acceleration sensor and a gyro sensor with identical local coordinate systems. An example of the electronic device is a mobile phone including a microprocessor. The following description will be based on a mobile phone.

[0059] Both the Android OS and the iOS system that runs the iPhone use the same local coordinate system. That is, when the phone is placed on a flat surface with the screen facing up, the direction of the display from the floor is the z-axis, and the counterclockwise rotation around the z-axis is the yaw momentum. Also, when the phone is held upright, the direction from left to right is the x-axis, and the counterclockwise rotation around the x-axis is called roll. At this time, the direction from bottom to top of the phone when held upright is the y-axis, and the amount of rotation around the y-axis is defined as pitch. Also, the direction of increase in rotational momentum for each axis is the direction pointed by the remaining four fingers when the positive direction of the axis is aligned with the thumb and grasped.

[0060] That is, the acceleration sensor is 9.8 m / sec toward the center of the Earth, i.e., in the direction of gravity. 2 It is subjected to gravitational acceleration. Thus, the mobile device is fixed near the chest of the standing subject, and if the subject is standing, it is subjected to 9.8 m / sec in the direction opposite to the direction of gravity. 2 Because the acceleration of is in effect.

[0061] Therefore, using an acceleration sensor, the gravitational acceleration (α) in the x-axis, y-axis, and y-axis directions according to the local coordinate system of the sensor can be measured without the shaking of the standing subject. x , α y , α z ) always has a magnitude of 1G (9.8m / sec) 2 ) and the direction is -Z, which is a vector. That is, α of the stationary accelerometer x , α y , α z The magnitude of the vector sum (the square root of the sum of the squares of the three values) always converges to the value 9.8.

[0062] First, a microprocessor defines a reference coordinate system that includes the direction extending from the right shoulder to the left shoulder when the subject is standing normally as the X-axis direction, the direction opposite to the gravity as the Y-axis direction, and the axis direction generated by the vector multiplication between the X-axis direction and the Y-axis as the Z-axis direction.

[0063] At this time, the opposite direction of gravity refers to a line close to a vertical line using the mastoid process of the temporal bone, the center of the acromion, the back of the second sacral vertebra, the front of the knee joint, and the front of the ankle joint as reference points on the sagittal plane of the human body.

[0064] Next, from the acceleration sensor, the microprocessor derives an acceleration value according to the reference coordinate system (S120).

[0065] The above acceleration sensor measures acceleration values ​​for each of the x-axis, y-axis, and z-axis directions according to the local coordinate system. The microprocessor converts the measured acceleration values ​​into acceleration values ​​according to the reference coordinate system.

[0066] To this end, the three-axis directions (x-axis, y-axis, and z-axis directions) constituting the local coordinate system can be corrected to match the three-axis directions (X-axis, Y-axis, and Z-axis directions) constituting the reference coordinate system.

[0067] FIG. 5 is a drawing for explaining a step of correcting the local coordinate system of the sensor to derive an acceleration value according to the reference coordinate system of FIG. 4.

[0068] Referring to FIG. 4 and FIG. 5, the attachment location of the mobile phone is illustrated as the chest area (a) and the left upper arm (b).

[0069] If the above mobile phone is attached to the chest area (a), it is mounted on the chest area with the x-axis according to the local coordinate system horizontal to the ground. At this time, an error of 0 to 15 degrees occurs depending on the chest inclination, so when mounting, x e , y e , ze It can be compensated by aligning it with the X-axis, Y-axis, and Z-axis, respectively. To this end, the angle between the vector opposite to gravity (Y-axis direction), which is the sum of the accelerometer vectors in a stable state, is calculated, and the calculated angle is compensated by rotation around the X-axis.

[0070] If the above mobile phone is mounted on the left upper arm (b), y e The method of correcting the value for the Y-axis is the same as that for the chest area (a), and x e The vector value is the value about the -Z axis, z e When a vector value is converted to a value regarding the (+)X axis, it can be converted to an acceleration value according to the reference coordinate system.

[0071] Conversely, for the right upper arm, y e The method of correcting the Y-axis is the same as when equipping the chest area, and x e The vector value is a value about the Z axis, z e Vector values ​​can be corrected to values ​​along the (-)X axis.

[0072] Next, if the maximum value of each of the above-mentioned conversion acceleration values ​​is maintained at 1.2g or less for a reference time, the microprocessor determines the starting posture for dynamic activity (S130). At this time, the static posture on the sagittal plane can be determined as the starting posture.

[0073] That is, there are movements in which the maximum values ​​of the acceleration values ​​in the X, Y, and Z directions are less than 1.2 g and the starting posture has a similar angle in the sagittal plane to that of a static posture. In this case, an example of a certain period of time can be set to 3 seconds or more.

[0074] For example, the starting posture can be set when the maximum acceleration value of each of the X-axis, Y-axis, and Z-axis directions remains at a state of less than 1.2g for 3 seconds or more.

[0075] Figure 6 is a drawing for explaining the sagittal plane angle for each exercise type with respect to the starting posture of Figure 4.

[0076] Referring to FIGS. 4 and 6, there are exercises in which the maximum value of the acceleration values ​​along the X-axis, Y-axis, and Z-axis directions is less than 1.2 g and the starting posture is at a similar angle on the sagittal plane to the static posture. That is, an angle on the sagittal plane of about 60 to 70 degrees can be considered the starting posture for ball sports such as table tennis, tennis, and badminton.

[0077] At this time, the angle of the body part on the saddle plane can be calculated as follows.

[0078] First, set the starting position, start timing from the starting point, and count the time taken.

[0079] For this, the accelerometer value A measured by the acceleration sensor at the starting position g =(X g , Y g , Z g ) is the angle θ formed by the unit vector Z(0,0,1) in the Z-axis direction. 시상면 silver,

[0080] θ 시상면 =cos -1 ([A g · Z(0,0,1)] / [|A g |×|Z(0,0,1)|])

[0081] Save with .

[0082] For example, in the prone position, θ 시상면 is close to 0 radians, and θ in the normal standing posture 시상면 is close to π / 2 radian, and θ in the supine position 시상면 is close to π radian.

[0083] In one embodiment of the present invention, the starting posture can be determined as the posture of the subject being measured, such as standing, semi-lateral, sitting, (semi) Fowler's position, supine, or prone. In this regard, detailed descriptions are provided in Korean Patent No. 10-2491165, “Method for Determining the Posture of a Nursing Patient and Method for Obtaining Data on Physical Activity Signs of a Patient Using the Same,” which the applicant of the present invention applied for and was granted patent for, and Korean Patent No. 10-2528684 (Title: Method for Determining the Posture of a Subject).

[0084] Next, when the starting posture is determined, a movement signal regarding the walking information of the subject is generated from the angular velocity value according to the reference coordinate system from the gyro sensor (S140). At this time, the movement signal includes information regarding the movement according to the walking of the subject after the starting posture.

[0085] That is, the microprocessor can extract information about the walking state by processing the acceleration value obtained using the gyro sensor.

[0086] Generally, the basis of dynamic postural changes is walking. For dynamic postural changes that do not involve moving from a standing position, cadence is defined as the time in seconds divided by the time required.

[0087] For example, in table tennis and badminton, short distances of 1 to 5 meters are traveled at a fast speed with a cadence (steps per minute) of 200 or more, with 3 or fewer steps per stroke, while in tennis, distances of 1 to 15 meters, which are longer than table tennis, are traveled with 3 to 7 steps, which is different from table tennis and badminton.

[0088] Meanwhile, the walking information of the subject is defined as a walking unit by the transition between (+) / (-) of the angular velocity value based on the Y-axis.

[0089] The signal from the gyro sensor attached to the fuselage is determined to be a right foot stepping state when the angular velocity measurement value with respect to the Y-axis is a positive signal. On the other hand, when the angular velocity measurement value is a negative signal, it is determined to be a left foot stepping state.

[0090] Meanwhile, the walking information of the subject can be used to check the walking characteristics of the subject by using the time difference between the (+) / (-) of the angular velocity value based on the Y-axis.

[0091] For example, by measuring N consecutive gait signals,

[0092] Average time between right foot signal peak (+) and left foot signal peak (-): t +to- ,

[0093] Average time between left foot signal peak (-) and right foot signal peak (+): t -to+

[0094] When defined as,

[0095] t +to- < t -to+ During back walking, the left foot supports the body for a longer period of time.

[0096] t +to- > t -to+ It can be seen that the right foot supports the body for a longer time during walking.

[0097] Therefore, the healthier a person is, the more likely it is that the time spent on the left and right feet while walking is the same or nearly the same, but if the balance of walking is disrupted due to disease or pain, the above walking characteristics can be used as an index to determine whether there is an abnormality in the left or right foot, or whether there is an abnormality in the nervous system required to move the left or right foot.

[0098] In one embodiment of the present invention, the electronic device may further include a geomagnetic sensor.

[0099] Fig. 7 is a graph for explaining the measurement values ​​of the gyro sensor and the geomagnetic sensor in the step of generating the movement signal of Fig. 4.

[0100] Referring to FIGS. 4 and 7, the point in time when the magnetic field value peaks in the direction having the smallest angle with the direction of the Earth's magnetic field lines among the directions of the geomagnetic sensor can be determined as a change in walking direction.

[0101] In analyzing information related to the direction change of the above gyro sensor and geomagnetic sensor, the gyro sensor value at the time when a high-inclination direction change signal is detected by the geomagnetic sensor can be linked.

[0102] At this time, if the magnetic field value of the gyro sensor has a large positive peak (right foot), the direction is changed counterclockwise by moving the right foot around the left foot as the axis.

[0103] On the other hand, if the magnetic field value of the gyro sensor has a large negative peak (left foot), it can be determined that the direction is changed clockwise by moving the left foot around the right foot.

[0104] Next, when the movement signal is absent for a certain period of time or the movement signal is being generated, the microprocessor receives a three-axis acceleration value according to the reference coordinate system from the gyro sensor and tracks the angular change value for the body part of the subject (S150).

[0105] More specifically, when the movement signal ends and the stroke state is entered, table tennis completes the stroke with a rotation centered on the transverse plane while restraining the movement of the sagittal and coronal planes while bending the upper body to hit the small ball with the small table tennis paddle. In badminton, the height of the net is higher than the player's height, so a unique stroke in which the upper body is bent backward to create an angle of attack for the badminton paddle is observed in a significant proportion. On the other hand, in tennis, a larger rotation in the transverse plane and a change in the body angle in the coronal plane are observed than in table tennis or badminton.

[0106] In order to automatically determine the above exercise type, the amount of rotation in the sagittal, coronal and transverse planes can be measured.

[0107] At this time, the angle formed by the three-axis acceleration vector with respect to the reference coordinate system and the Z-axis direction defines the amount of rotation on the sagittal plane.

[0108] To obtain the amount of rotation on the above plane, the accelerometer value A measured by the acceleration sensor g =(X g , Y g , Z g ) is the angle θ formed by the unit vector Z(0,0,1) in the Z-axis direction. 시상면 silver,

[0109] θ 시상면 =cos -1 ([A g · Z(0,0,1)] / [|A g |×|Z(0,0,1)|])

[0110] Save with .

[0111] Additionally, the angle formed by the three-axis acceleration vector with respect to the above reference coordinate system and the X-axis direction defines the amount of rotation on the coronal plane.

[0112] The change in the angle of the body part on the above-mentioned plane is the accelerometer value A g =(X g , Y g , Z g ) is the angle θ formed by the unit vector X(1,0,0) in the X-axis direction 관상면 am.

[0113] θ 관상면 =cos -1 ([A g · X(1,0,0)] / [|A g |×|X(1,0,0)|])

[0114] Save with .

[0115] Meanwhile, when the size of the peak of the magnitude of the three-axis angular velocity vector for the reference coordinate system is 2 / π or greater, the rotation amount on the cross-section of the body part of the subject can be defined using the value of the peak and the convergence time value for the peak to converge to 0.

[0116] For the ying, the signal R(x) is the 3-axis rotation angular velocity of the gyro sensor. r , y r , z r ) If the signal with the largest |R| value is greater than (2 / π) radian / sec, the area of ​​the rectangle obtained by multiplying the peak height by the time until the peak becomes 0 can be approximated as the rotation value on the cross section of the fuselage.

[0117] Additionally, the gyro sensor 3-axis rotation angular velocity signal R(x) measured every 1 msec after the peak r , y r , z r ) is the integral value multiplied by 2 from the peak point of the signal with the largest amplitude to the time when the signal becomes 0, and the rotation value (θ) on the cross section of the body is 횡단면 ) is used.

[0118] Therefore, most dynamic posture changes occur within seconds, so rotation value correction using a Kalman filter or similar is not required.

[0119] Meanwhile, the time required for a dynamic posture change is defined as the time from the end point of the starting posture until the signal of the reference axis for rotation of the body part drops to 0.

[0120] Figure 8 is a graph for explaining the change in body angle of Figure 4 by exercise type.

[0121] Referring to Figures 4 and 8, it can be confirmed that in canoeing (sections A and C), signals of axes forming the coronal plane (xy plane) and the transverse plane (xz plane) are evenly distributed, and in the case of rowing (section B), the vector of the x-axis, which is the transverse axis, is fixed, and there is flexion and extension of the upper body due to changes in the sagittal plane (yz plane).

[0122] Thereafter, the microprocessor determines the exercise type of the subject using the angle change value (S160).

[0123] To this end, the first rectangular parallelepiped for dynamic posture change is defined by reflecting the change in the body angle in the sagittal, coronal, and transverse planes and positioning it on a virtual three-dimensional orthogonal coordinate system composed of the x, y, and z axes.

[0124]

[0125] Figure 9 is a graph illustrating a virtual three-dimensional rectangular coordinate system for explaining the steps of forming a transformed rectangular parallelepiped from an initial rectangular parallelepiped.

[0126] Referring to Figure 9, the maximum value (Max_θ) for the angle change on the sagittal plane (YZ plane according to the reference coordinate system of the present invention) 시상면 ) and minimum value (Min_θ 시상면 ) is displayed on the x-axis of the virtual rectangular coordinate system.

[0127] In addition, the maximum angle change value (Max_θ) on the coronal plane (XY plane according to the reference coordinate system of the present invention) 관상면 ) and minimum value (Min_θ 관상면 ) is displayed on the y-axis.

[0128] Furthermore, the rotation value (0) and θ of the body part on the cross section (XZ plane according to the reference coordinate system of the present invention) 횡단면 is displayed on the z-axis. This creates an initial rectangular solid with six vertices in a virtual three-dimensional rectangular coordinate system.

[0129] Thereafter, a transformed rectangular parallelepiped is formed by moving each of the coordinate points constituting the vertices of the initial rectangular parallelepiped along a diagonal line connecting the vertices corresponding to the maximum and minimum values ​​of the initial rectangular parallelepiped by a size equal to (the time required × the average value of the cadence).

[0130] Afterwards, the minimum vertex of the rectangular solid (Min_θ 시상면, Min_θ 관상면 , 0) and the maximum vertex (Max_θ 시상면 , Max_θ 관상면 , θ 횡단면 ) is moved along the diagonal line connecting the coordinates of each vertex and the size of (time required (sec) × Cd (average cadence value within the measurement time / 60) Hz) to form a transformed rectangular parallelepiped. At this time, the cadence can be multiplied by an appropriate number to increase the resolution of the distinction between movements.

[0131] Referring again to Figure 9, (Max_θ) on the x-axis 시상면 ) = 1.92 radian, (Min_θ 시상면 ) = 0.87 radians, on the y-axis (Max_θ 관상면 ) = 1.92 radian, (Min_θ 시상면 ) = 1.2 radian,

[0132] θ on the z-axis 횡단면 = 0.83 radian,

[0133] The dynamic posture change is defined as [average cadence (Avg_Cadence = 2.1 Hz) × time required].

[0134]

[0135] The standard rectangular solid for each of the above sports is listed in Table 1 below. The angle is in radians.

[0136] Distinction Max_θ 시상면 Min_θ 시상면 Max_θ 관상면 Min_θ 관상면 θ 횡단면Average Cd (Hz) Time (sec) Plank 0.35 0.09 1.74 1.40 0.09 0.00 9 0.01 Push-ups 0.52 0.09 1.74 1.40 0.09 0.29 0.01 Burpees 1.60 0.35 1.74 1.40 0.09 0.23 0.8 Dips 1.55 0.61 1.66 1.48 0.05 0.58 0.01 Crunches 3.35 2.53 0.26 0.00 0.87 1.35 0.01 Lunges 1.78 1.29 1.74 1.40 0.09 0.45 0.5 squats 0.70 1.59 1.74 1.400 09 0.65 0.01 sit-ups 3.35 2.53 0.21 0.00 0.05 0.70 01 billiards 1.400 47 1.74 1.400 07 1.610 bowling 1.401 05 2.04 1.400 35 2.05 5.6 rowing 2.01 1.05 1.66 1.48 0.090 .451Canoeing1.661.221.831.310.610.651Running1.571.131.661.480.3520.2Walking1.571.261.661.480.141.35300Cycling1.240.031.781.360.050.1300Golf1.571.202.011.312.271.35225 Tennis 1.571.052.091.221.602.64.2 Table tennis 1.571.051.661.480.50.50.5 Badminton 1.571.062.40.80.30.60.7 Jumping rope 1.541.131.661.480.092.90.3 Figure skating 2.620.522.440.876.282.75240 Speed ​​skating 1.400.171.951.360.172.590

[0137] Figure 10 is a graph depicting a virtual three-dimensional rectangular coordinate system based on a standard rectangular solid for each sport.

[0138] By comparing the above-described transformed rectangular solid with the reference rectangular solid for each exercise type in terms of position, the microprocessor can determine the exercise type of the subject.

[0139] A device according to embodiments of the present invention may include a processor, a memory for storing and executing program data, permanent storage such as a disk drive, a communication port for communicating with an external device, a user interface device such as a touch panel, a key, a button, etc. Methods implemented as software modules or algorithms may be stored on a computer-readable recording medium as computer-readable codes or program instructions executable on the processor. Here, the computer-readable recording medium includes a magnetic storage medium (e.g., read-only memory (ROM), random-access memory (RAM), floppy disk, hard disk, etc.) and an optical reading medium (e.g., CD-ROM, DVD: Digital Versatile Disc)). The computer-readable recording medium may be distributed to computer systems connected through a network, so that the computer-readable code may be stored and executed in a distributed manner. The medium may be readable by a computer, stored in a memory, and executed by a processor.

[0140] Embodiments of the present invention may be represented by functional block configurations and various processing steps. These functional blocks may be implemented by various hardware and / or software components that perform specific functions. For example, embodiments may employ integrated circuit components, such as memory, processing, logic, and look-up tables, that may perform various functions under the control of one or more microprocessors or other control devices. Similarly, just as the components of the present invention may be implemented by software programming or software components, embodiments may be implemented by programming or scripting languages, such as C, C++, Java, and assembler, including various algorithms implemented by a combination of data structures, processes, routines, or other programming components. Functional aspects may be implemented by algorithms that execute on one or more processors. Furthermore, embodiments may employ conventional techniques for electronic configuration, signal processing, and / or data processing. Terms such as "mechanism," "element," "means," and "composition" can be used broadly and are not limited to mechanical or physical structures. These terms can also encompass a series of software routines, such as those associated with a processor.

[0141] The specific implementations described in the embodiments are merely exemplary and do not limit the scope of the embodiments in any way. For the sake of brevity of the specification, descriptions of conventional electronic components, control systems, software, and other functional aspects of the systems may be omitted. In addition, the lines connecting or connecting members between components illustrated in the drawings are merely exemplary functional connections and / or physical or circuit connections, and may be replaced or represented as various additional functional connections, physical connections, or circuit connections in an actual device. In addition, unless specifically mentioned as "essential," "important," etc., a component may not be absolutely necessary for the application of the present invention.

[0142] Embodiments according to the present invention can be applied to technologies for measuring changes in the human body's posture during exercise or daily life that require dynamic posture measurement, such as smart watches and mobile phones.

[0143] The dynamic posture measurement method according to embodiments of the present invention can be used for analyzing athletes' posture and movements, assessing exercise effectiveness, developing rehabilitation programs, etc., and can also be applied in the medical field to diagnosing balance disorders, correcting posture, and rehabilitating treatment. Furthermore, the data acquired using the dynamic posture measurement method can be utilized in industrial fields to analyze workers' postures, improve work environments, and design ergonomic products.

Claims

1. A method for determining an activity state using an electronic device including an acceleration sensor and a gyro sensor that can be fixed to a body part of a subject and have the same local coordinate system, A step for defining a reference coordinate system in which the direction extending from the right shoulder to the left shoulder when the subject is standing normally is defined as the X-axis direction, the direction opposite to gravity is defined as the Y-axis direction, and the axis direction generated by the vector multiplication between the X-axis direction and the Y-axis is defined as the Z-axis direction; A step of deriving an acceleration value according to the reference coordinate system from the acceleration sensor; A step of determining a starting posture for dynamic activity by viewing a static state in which the maximum value of the acceleration values ​​in each of the X, Y and Z axis directions constituting each of the above acceleration values ​​is maintained at 1.2g or less for a reference time; When the starting posture is determined to be the above, a step of receiving an angular velocity value according to the reference coordinate system from the gyro sensor and generating a movement signal regarding the walking information of the subject; A step of receiving a three-axis angular velocity value according to the reference coordinate system from the gyro sensor and tracking an angular change value for the body part of the subject when the movement signal is absent for a certain period of time or while the movement signal is being generated; and A step of determining the type of movement of the subject using the above angle change value; A method for determining an activity state including:

2. In the first paragraph, the step of determining the starting posture is: A method for determining an activity state, characterized in that the angle formed by a three-axis acceleration vector with respect to the above reference coordinate system is calculated with respect to the Z-axis direction.

3. A method for determining an activity state, characterized in that in the first paragraph, the step of determining the starting posture includes determining the posture of the subject as standing, semi-lateral, sitting, (semi) Fowler's position, supine, or prone.

4. A method for determining an activity state in the first paragraph, characterized in that the walking information of the subject is defined as a walking unit by the transition between (+) / (-) of the angular velocity value based on the Y-axis.

5. In the fourth paragraph, the gait information of the subject is a method for determining an activity state characterized in that the gait characteristics of the subject are confirmed by using the time difference between the (+) / (-) of the two-foot transition of the angular velocity value based on the Y-axis.

6. In the fourth paragraph, the electronic device further includes a geomagnetic sensor, A method for determining an activity state, characterized in that the point in time when the magnetic field value peaks in the direction having the smallest angle with the direction of the Earth's magnetic field lines among the directions of the above-mentioned geomagnetic sensor is determined as a change in walking direction.

7. In the first paragraph, the step of tracking the angle change value of the body part of the subject is, The angle formed by the three-axis acceleration vector for the above reference coordinate system with the Z-axis direction is used to calculate the amount of rotation on the sagittal plane. The angle formed by the three-axis acceleration vector for the above reference coordinate system with the X-axis direction is used to calculate the amount of rotation on the coronal plane, A method for determining an activity state, characterized in that when the size of the peak of the three-axis angular velocity with respect to the above reference coordinate system is 2 / π or greater, the amount of rotation on the cross-section of the body part of the subject is calculated using the value of the peak and the convergence time value for the peak to converge to 0.

8. A method for determining an activity state in the 7th paragraph, characterized in that the angle change value includes the minimum and maximum values ​​of the rotation amount according to each of the sagittal plane and the coronal plane and the rotation amount of the cross-section.

9. In the 8th paragraph, the gait information includes a cadence value, and the time from the end point of the starting posture until the rotation angle value on the cross-section of the body part becomes 0 is defined as the required time, A method for determining an activity state, characterized in that the step of determining the type of exercise of the subject using the above angle change value further uses the cadence value and the required time.

10. In the 9th paragraph, the step of determining the type of movement of the subject using the angle change value includes the step of forming an initial rectangular parallelepiped by displaying the minimum and maximum values ​​of the rotation amount according to each of the sagittal plane and the coronal plane and the rotation amount of the cross section on a virtual three-dimensional rectangular coordinate system having an x-axis, a y-axis, and a z-axis within the required time. A method for determining an activity state.

11. In the 10th paragraph, a step of forming a transformed rectangular parallelepiped by moving each of the coordinate points constituting the vertices of the initial rectangular parallelepiped along a diagonal line connecting the vertices corresponding to the maximum and minimum values ​​of the initial rectangular parallelepiped by a size of (the required time × the average value of the cadence); and A method for determining an activity state, characterized by including a step of comparing the transformed rectangular solid with a reference rectangular solid for each exercise type in terms of position.

Images