KR101872702B1 - System and method for walking activity prediction service for exercise motive - Google Patents

Abstract

The present invention relates to a system and method for predicting walking activity for motivation. The present invention relates to a sensor (3) used as a pedometer; And stores the measured values in the internal app DB 30 every time the sensor 3 measures the number of steps, and the stored values are stored in the view (View) A walking activity prediction app 10 provided to the user 1 through a screen; A current screen and a future prediction screen, and the data displayed on the future prediction screen can be calculated using the prediction filter 30 of the walking activity prediction app 10.
Thus, not only the walking exercise information data for the user's basic current and past but also the prediction data for the user should be given to the user to exercise, through the walking exercise activity prediction service, to provide the nudge effect And provides an effect that helps the user to motivate the exercise.
In addition, it provides motivation beyond providing the extracted data to the user, and provides the user with an effect to contribute to improvement of the health condition.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of predicting walking activity,

The present invention relates to a system and method for predicting walking activity for exercise motivation, and more particularly, to a method and system for estimating walking activity, And more particularly, to a system and method for estimating a walking activity amount for exercise motivation to provide motivation.

Modern people who are busy due to work and study are often sitting in a chair for a long time. Therefore, most modern people who are aware of situations that have a lack of activity and suffer from lack of exercise and have a bad health pattern are interested in health and exercise. It is emphasized.

Based on this social situation, the importance of the walking exercise, which is simple and easy to do anywhere, is becoming more important, and there is an increasing demand for an analysis of the degree of walking and how to do the walking beyond the simple walking have.

The development of hardware (H / W) and software (S / W) has led to the universalization of smart devices such as wearables and smart phones, and it has become possible to satisfy the requirements of users one by one.

With many consumer requirements and an environment that meets these requirements, applications for measuring and accessing individual athletic data ("apps") have become more popular.

However, most exercise-related app services are limited to current and historical data, and there are limitations that focus on measuring the current number of steps and showing historical measurement records.

Korean Patent Registration No. 10-1647361 entitled " APPARATUS AND METHOD FOR PREDICTING RESPIRATORY MOTION " Korean Patent Registration No. 10-1545258 entitled " BIOMARKER FOR PREDICTING OF SENSITIVITY TO EXERCISE " Korean Patent Laid-Open Publication No. 10-2005-0095326 "CELLULAR PHONE WITH THE SENSOR DETECTING QUANTITY OF MOTION" Korean Patent Registration No. 10-0195366 entitled "EXERCISE AMOUNT MEASURING DEVICE"

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to provide a method and apparatus for estimating a walking exercise activity data by using not only a walking exercise information data for a user's basic current and past, The present invention provides a system and method for predicting a walking activity amount for providing a motive for inducing a nudge effect and providing a user with motivation for exercise.

The present invention also provides a system and method for estimating a walking activity amount for motivating a user to motivate a user not only to provide extracted data to a user but also to contribute to improvement of a health condition to a user.

However, the objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

In order to achieve the above object, a walking activity estimation system for exercise motivation according to an embodiment of the present invention includes a sensor 3 used as a pedometer; And stores the measured values in the internal app DB 30 every time the sensor 3 measures the number of steps, and the stored values are stored in the view (View) A walking activity prediction app 10 provided to the user 1 through a screen; A current screen and a future prediction screen, and the data displayed on the future prediction screen can be calculated using the prediction filter 30 of the walking activity prediction app 10.

Here, in the walking activity estimation system for exercise motivation according to another embodiment of the present invention, the walking activity estimation app 10 may display a personal information screen, a target screen, a past screen , A current measurement screen, a future prediction screen, and a personal recording screen.

In addition, in the walking activity estimation system for exercise motivation according to another embodiment of the present invention, when receiving the period selection information input, the walking activity estimation application 10 adds the data corresponding to the prediction data to the period selection information If the period is included, a prediction calculation can be performed.

In order to achieve the above object, the method for predicting walking activity for exercise motivation according to the embodiment of the present invention is a method for predicting walking activity of a user (1) after receiving the personal information of the user (1) In an app DB (20) which is an internal database of the predictive app (10); A second step of storing the personal goal in the app DB 20 after the walking activity estimation application 10 receives the personal goal of the user 1; A third step in which the walking activity estimation app 10 starts measuring the number of steps of the user 1 through the control of the sensor 3; A fourth step of storing the time and the number of steps in the internal application DB 20 of the walking activity estimation app 10 every time the walking activity estimation application 10 recognizes the user 1 walking; When the walking activity estimation application 10 is requested by the user 1 with a predicted value, the walking activity estimation application 10 uses a prediction filter 30 using the Kalman filter in the walking activity estimation app 10, A fifth step of calculating a value; And a walking activity estimation app (10) outputting a measured value and a predicted value; . ≪ / RTI >

In order to achieve the above object, according to another aspect of the present invention, there is provided a method for predicting a walking activity amount for exercise motivation, the walking activity prediction application (10) A first step of confirming whether or not to input a target of the user (1) upon inputting; A second step of receiving the data selection information of the past, present, and future from the user 1 when the walking activity estimation application 10 inputs the goal as well as the personal information; A third step in which the walking activity estimation app 10 receives period information of desired data from the user 1 according to reception of the data selection information; And a fourth step of, when receiving the period selection information, proceeding to a prediction calculation process when the walking activity estimation application 10 includes a data period corresponding to prediction data in the period selection information; . ≪ / RTI >

Hereinafter, the method for predicting walking activity for exercise motivation according to another embodiment of the present invention may further include, after the fourth step, outputting a result based on the contents requested by the user (1) 5 < th > . ≪ / RTI >

The system and method for predicting the walking activity amount for exercise motivation according to the embodiment of the present invention can be applied not only to the user's basic current and past walking exercise information data through the walking exercise activity prediction service, Provides predictive data rather than a coercive message to induce a nudge effect, which provides the user with the motivation for motivation.

In addition, the system and method for predicting the walking activity for motivation in accordance with another embodiment of the present invention can provide motivation beyond providing the extracted data to the user and contribute to improvement of the health condition to the user to provide.

1 is a diagram illustrating a concept of a service of the walking activity prediction application for estimating walking activity in the system for predicting the walking activity for exercise motivation according to the embodiment of the present invention.
FIG. 2 is a flowchart showing a usage process of the walking activity estimation app 10 by the user 1 among the walking activity estimation methods for motivation for exercise according to the embodiment of the present invention.
3 is a diagram showing the system architecture of the walking activity estimation app 10 in the walking activity estimation system for motive motivation according to the embodiment of the present invention.
FIG. 4 is a flowchart illustrating an overall process of a method for predicting walking activity for exercise motivation according to an embodiment of the present invention. FIG.
FIG. 5 is a diagram for explaining a Kalman filter cycle for explaining a data prediction calculation process among the walking activity estimation methods for motional exercise according to the embodiment of the present invention shown in FIG. 4;
6 is a flowchart illustrating a method of predicting the walking activity of a user (1) from the mobile smart device (2) to the output unit (4) by the walking activity prediction application (10) A diagram showing an implemented UI screen (Display of body information).
FIG. 7 is a graph showing a UI screen showing the contents calculated by the walking activity estimation application 10 in the walking activity estimation system for motivation exercise according to an embodiment of the present invention, drawing.
FIG. 8 is a chart showing the past walking status daily chart UI implemented by the walking activity estimation application 10 in the walking activity estimation system for exercise motivation according to the embodiment of the present invention.
FIG. 9 is a UI display of current walking activity, which is a chart showing current walking statuses implemented by the walking activity prediction application 10 in the walking activity amount prediction system for exercise motivation according to the embodiment of the present invention.
FIG. 10 is a diagram showing a UI screen of a future walking status screen implemented by the walking activity prediction application 10 in a walking activity amount prediction system for motivation in accordance with an embodiment of the present invention, ).
FIG. 11 is a UI display of record in walking activity implemented by the walking activity prediction application 10 in the walking activity estimation system for exercise motivation according to the embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a detailed description of preferred embodiments of the present invention will be given with reference to the accompanying drawings. In the following description of the present invention, detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In the present specification, when any one element 'transmits' data or signals to another element, the element can transmit the data or signal directly to the other element, and through at least one other element Data or signal can be transmitted to another component.

1 is a diagram illustrating a concept of a service of a walking activity prediction application (hereinafter referred to as a 'walking activity prediction application') 10 in a walking activity estimation system for motivation in accordance with an embodiment of the present invention.

Referring to FIG. 1, a user 1 inputs information on a health state to use the walking activity estimation app 10, and then receives data on a walking exercise.

When the user 1 inputs personal information such as his / her key, body weight, and the number of daily goal steps, the walking activity prediction app 10 measures and predicts data to determine the past, present, and future factors And the user 1 gives an option to the desired part to provide information.

Further, the walking activity prediction app 10 provides the user 1 with a record extracted based on the measured data.

FIG. 2 is a flowchart showing a usage process of the walking activity estimation app 10 by the user 1 among the walking activity estimation methods for motivation in accordance with the embodiment of the present invention. Referring to FIG. 2, when the user 1 executes the walking activity estimation app 10 (S11), the user 1 first inputs the personal information of the user 1 (S12). The personal information input by the user 1 is stored in the app DB 20, which is an internal database of the walking activity prediction app 10 (S13).

After step S13, the walking activity estimation app 10 stores the contents of the personal goal input from the user 1 in the same manner as the user's personal goal in the application DB 20 (S14, S15).

After storing both personal information and other goal inputs, the walking activity prediction app 10 starts measuring the number of steps.

Accordingly, the walking activity estimation application 10 stores the time and the number of steps in the internal activity DB 20 of the walking activity estimation app 10 every time the user 1 recognizes the user's step (S16).

If the user 1 desires a predicted value (S17), a value is calculated through a prediction filter (30) using the Kalman filter in the walking activity prediction app (10) with the stored number of steps (S18, S19). When the calculation is completed, the walking activity estimation app 10 displays the measured value and the predicted value to the user 1 (S20 to S22).

FIG. 3 is a diagram illustrating a system architecture of the walking activity estimation app 10 in the walking activity estimation system for motive motivation according to the embodiment of the present invention.

Referring to FIG. 3, the walking activity prediction app 10 measures the number of steps using the pedometer sensor 3 provided by the mobile smart device 2 based on the Android OS. Every time the sensor 3 measures the number of steps, the walking activity estimation app 10 stores the measurement value in the internal app DB 30. [ These stored values are provided to the user 1 through a View screen. The view typically includes a past screen, a current screen, and a future prediction screen. The data displayed on the future prediction screen is calculated using the prediction filter 30 of the walking activity prediction app 10.

4 is a flowchart illustrating an overall process of a walking activity estimation method for motivation for exercise according to an embodiment of the present invention. Referring to FIG. 4, when the walking activity estimation app 10 is executed, it is first checked whether personal information of the user 1 is input (S10). When the personal information is inputted (S120), it is confirmed whether or not the user 1 inputs a target (S130).

If the user inputs not only the personal information but also the target (S140), the walking activity prediction app 10 receives the data selection information of past, present, and future from the user 1 (S150).

Upon receipt of the data selection information, the walking activity estimation app 10 receives period information of desired data from the user 1 (S160).

Upon receipt of the period selection information, the walking activity prediction app 10 proceeds with the prediction calculation process when the period selection information includes the data period corresponding to the prediction data (S170) (S180).

After step S180, the walking activity estimation app 10 outputs a result based on the contents requested by the user 1. [

FIG. 5 is a diagram for explaining a Kalman filter cycle for explaining step S180, which is a data predictive calculation process in the method of predicting walking activity for exercise motivation according to the embodiment of the present invention shown in FIG. Referring to FIG. 4, the prediction filter 30 applies a Kalman filter for data prediction, that is, to implement a predicted part of the number of steps. The Kalman filter can be divided into a prediction step and a correction step. (1) Project the state ahead, (2) Project the error The convariance ahead calculates the error changes for the state estimation and the estimation. Correction step (1), the predicted value of the calculate Compute the Kalman gain, (2) Update estimate with measurement z _k, (3) Update the error convariance the optimal Kalman gain (Kalman gain) value K, and the predicting step and the actual Use the error between measured values to inductively modify the previously obtained values.

In order to predict the movement of the dry-period motion, a Kalman filter widely used for robot control, data prediction, and path estimation is used to track the state of the linear dynamic system. This allows optimal statistical prediction from current momentum information accumulated over time.

 A predicted part of the number of steps is implemented using a Kalman filter, and FIG. 8 shows a process for a Kalman filter cycle. The Kalman filter can be divided into a prediction step and a correction step. In the prediction steps (1) and (2), error changes for the state estimation and estimation are calculated. In the correction steps (1), (2), and (3), the optimal Kalman gain value is calculated and the previously obtained value is corrected inductively using the error between the predicted value and the actual measured value in the prediction step.

은 특정 시간 k에서의 보정을 하기 전인 걸음 수 예측 값이다.

은 k-1에서의 걸음 수인 x_k-1, 이전 상태에 대한 상태 전이인 A, 사용자 입력에 의한 상태 전이인 B, 특정 시간에서 사용자 입력인 u_k를 통해 구할 수 있다. 예측 단계(2, 수학식 2)에서 공분산 P_k-을 추정한다. 본 발명에서의 예측은 상태가 전이 되지 않기 때문에 A = 1 이라 정의하고, 사용자 입력이 없기 때문에 u_k=0이라 정의한다. 수학식 1과 수학식 2는 다음과 같이 정의한다. The prediction value is estimated through the previous data in the prediction step (1, see equation (1)). Seeking silver

Is a step number prediction value before correction at a specific time k.

Is a B, the state transition by the state transition A, the user inputs to the step number x _k-1, k-1 from the previous state at a particular time is available through a user input of u _k. The covariance _Pk - is estimated in the prediction step (2, equation (2)). The prediction in the present invention defines A = 1 because the state does not transit, and u _k = 0 because there is no user input. Equations (1) and (2) are defined as follows.

예측 단계(2, 수학식 2)에서 공분산 P_k-을 추정하되, 예측은 상태가 전이 되지 않기 때문에 A = 1 이라 정의하고, 사용자 입력이 없기 때문에 u_k=0이라 정의되며, Q는 예측 노이즈 공분산으로 정의하고,
보정 단계 (3, 수학식 3)에서는 Pk-, H, R을 이용해 특정 시간 k에서의 칼만 게인 Kk을 구하고, H는 전달함수, R은 측정 노이즈 공분산으로 정의하고, 보정 단계 수학식 4에서 xk-와 Kk, zk를 통해 특정 시간 k에서의 구하려는 걸음 수 예측 값 xk을 구할 수 있다. 보정 단계 수학식 5에서 공분산 Pk을 보정한다. 수학식 3, 수학식 4, 수학식 5는 다음과 같이 정의한다.

In the prediction step (2, equation (2)), the covariance _Pk - is estimated. Since the state is not transitioned, A = 1 is defined and _uk = 0 because there is no user input. Covariance defined as covariance,
In the correction step (3, Equation 3), Kalman gain Kk at a specific time k is obtained by using Pk-, H and R, H is defined as a transfer function, R is defined as a measured noise covariance, - and Kk, zk, the predicted step number xk to be obtained at a specific time k can be obtained. Correction step Correct the covariance Pk in equation (5). Equations (3), (4) and (5) are defined as follows.

delete

Based on the above description, the environment in which the walking activity prediction application 10 for implementing the prediction service of the walking activity estimation application 10 is developed can be as shown in [Table 1] below.

Development language Development Tools Execution environment Java, XML Android Studio Samsung Galaxy
Note 4
Android 6.0
Mashmallow

6 is a diagram showing a user interface (UI) screen displaying the physical information of the user 1 from the mobile smart device 2 to the output unit 4 by the walking activity prediction app 10 body information. Referring to FIG. 6, in order to execute the walking activity estimation app 10 and calculate more accurate data, first, the user needs to input body information. The body information of the user 1 is basically a key and a weight.

Table 2 below shows an example of a calculation method of estimation.

stride Key (cm) * 0.4 Expected distance Stride (cm) * Number of steps (beam) Estimated consumption time Distance (m) / adult average walking speed (4km / h) Expected calorie consumption 0.0669 Kcal * Weight (kg) * Time (minutes)

The user 1 inputs personal information after inputting the personal information. The goal is the step number of the beam unit. Estimates can be calculated using the calculations in Table 2, by calculating estimates of expected distance, expected consumption time, and estimated calories burned according to the goal entered by the user (1).

According to this calculation, FIG. 10 shows the results of calculation of the walking activity estimation appara- tus 10 by the user (1) and the calculation method of the estimation value in the walking activity estimation system for exercise motivation according to the embodiment of the present invention Fig.

FIG. 8 is a chart showing the past walking status daily chart UI implemented by the walking activity prediction app 10 in the walking activity estimation system for motivation in accordance with the embodiment of the present invention. Referring to FIG. 8, after the user 1 sets the body information and the target, the walking activity prediction app 10 measures the number of steps through the pedometer sensor 3 and stores it in units of minutes. The stored data set is visually provided to the user 1 for easy viewing.

FIG. 9 is a UI display of current walking activity, which is a chart showing a walking current state implemented by the walking activity prediction application 10 in the walking activity amount prediction system for exercise motivation according to the embodiment of the present invention. Referring to FIG. 9, the user 1 can view data such as exercise time, consumed calories burned, and the like, while comparing the set target with data measured at the current time (one day).

FIG. 10 is a diagram showing a UI screen of a future walking status screen implemented by the walking activity prediction application 10 in a walking activity amount prediction system for motivation in accordance with an embodiment of the present invention, )to be. Referring to FIG. 13, the user 1 can check past and current data as well as future prediction data. Future forecast data is calculated by applying a Kalman filter. That is, as shown in FIG. 13, it is possible to show the estimated number of steps after one day, two days, and three days after the prediction using the Kalman filter.

FIG. 11 is a UI display of record in walking activity implemented by the walking activity prediction application 10 in the walking activity estimation system for exercise motivation according to the embodiment of the present invention. 11, the measured number of steps data is processed in a different form beyond simply providing data for each day, month, and hour, and the user 1 can calculate the activity amount contents such as the maximum number of steps per minute, the average number of steps per minute, Can be provided as a single statistical data.

Accordingly, the system and method for predicting the walking activity amount for the exercise motive can provide a service for predicting the walking exercise amount using the nudge concept. The user (1) is provided with measurement data and prediction data on his / her health state and physical activity through the walking activity estimation app (10) service. The data extracted from the walking activity prediction app 10 will motivate the user 1 not only to provide data to the user 1 but also contribute to improving the health state to the user 1. In the present invention, it is possible to expand not only the step number prediction but also the predicted data such as the consumed calories and the weight, so that the user 1 can show more motivation.

In another embodiment of the present invention, an artificial intelligence method such as Deep Learning may be used for more detailed and accurate prediction rather than Kalman filter based prediction.

The present invention can also be embodied as computer-readable codes on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored.

Examples of the computer-readable recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device and the like, and also implemented in the form of a carrier wave (for example, transmission over the Internet) .

The computer readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner. And functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers skilled in the art to which the present invention pertains.

As described above, preferred embodiments of the present invention have been disclosed in the present specification and drawings, and although specific terms have been used, they have been used only in a general sense to easily describe the technical contents of the present invention and to facilitate understanding of the invention , And are not intended to limit the scope of the present invention. It is to be understood by those skilled in the art that other modifications based on the technical idea of the present invention are possible in addition to the embodiments disclosed herein.

1: User
2: Mobile Smart Device
3: Sensor
4: Output section
10: Walking activity prediction app
20: App DB (application DB)
30: prediction filter

Claims (6)

delete delete delete A sensor 3 used as a pedometer; And
The sensor 3 measures the number of steps of the user 1 and the measured value is stored in the internal app DB 30 every time the sensor 3 measures the number of steps. To a user (1) through a walking activity estimation app (10); ≪ / RTI >
The walking activity prediction app 10 provides a personal information screen, a target screen, a past screen, a current measurement screen, a future prediction screen, and a personal record screen with a screen that the user 1 can select,
Wherein the view is divided into a past screen, a current screen, and a future prediction screen, and data displayed on a future prediction screen is calculated using a prediction filter (30) of a walking activity prediction application (10) A method for predicting walking activity using a walking activity prediction system,
A first step of storing the personal information in the app DB 20 which is an internal database of the walking activity prediction app 10 after the walking activity estimation app 10 receives the personal information of the user 1;
A second step of storing the personal goal in the app DB 20 after the walking activity estimation application 10 receives the personal goal of the user 1;
A third step in which the walking activity estimation app 10 starts measuring the number of steps of the user 1 through the control of the sensor 3;
A fourth step of storing the time and the number of steps in the internal application DB 20 of the walking activity estimation app 10 every time the walking activity estimation application 10 recognizes the user 1 walking;
When the walking activity estimation application 10 is requested by the user 1 with a predicted value, the walking activity estimation application 10 uses a prediction filter 30 using the Kalman filter in the walking activity estimation app 10, A fifth step of calculating a value; And
The Kalman filter for implementing the predicted part of the step number,
(1) Project the state ahead; (2) Project the error, which calculates the error change for the state estimation and the estimation, and the prediction step for calculating the error change for the state estimation and the estimation; and (1) Compute the Kalman (2) Update estimate with measurement z _k , (3) Calculate the optimal Kalman gain value K and inductively correct the previously obtained value using the error between the predicted value and the actual measured value at the prediction step And a correction step of inductively correcting the previously obtained value by using the error between the predicted value and the actual measured value at the prediction step,
In the prediction step (1, see equation 1)
(1)

The prediction value is estimated through the previous data,

Is the step number predicted value before the correction at the specific time k,

Is obtained in B, the state transition by the state transition A, the user inputs to the step number x _k-1, k-1 from the previous state at a particular time through the user input is u _k,
In the prediction step (2, equation 2)
(2)

A covariance P _k - but estimating, the prediction is defined as u _k = 0 Because of the state transition is not A = 1 because there is not defined, and the user input, Q is defined as a noise covariance predictions,
In the correction step (3, Equation 3)
(3)

The Kalman gain Kk at a specific time k is obtained by using Pk-, H, and R as follows: H is a transfer function, R is a measured noise covariance,
In the correction step (4)
(4)

The predicted step number xk to be obtained at a specific time k is obtained through xk-, Kk, and zk,
In the correction step (5)
(5)

The covariance Pk is corrected,
A sixth step of the walking activity prediction app 10 outputting a measured value and a predicted value; Wherein the step of estimating the walking activity amount comprises:
A sensor 3 used as a pedometer; And
The sensor 3 measures the number of steps of the user 1 and the measured value is stored in the internal app DB 30 every time the sensor 3 measures the number of steps. To a user (1) through a walking activity estimation app (10); ≪ / RTI >
The walking activity prediction app 10 provides a personal information screen, a target screen, a past screen, a current measurement screen, a future prediction screen, and a personal record screen with a screen that the user 1 can select,
Wherein the view is divided into a past screen, a current screen, and a future prediction screen, and data displayed on a future prediction screen is calculated using a prediction filter (30) of a walking activity prediction application (10) A method for predicting walking activity using a walking activity prediction system,
A first step of confirming whether or not the walking activity estimation application 10 inputs a target of the user 1 when the personal information is inputted according to confirmation of whether the user 1 inputs personal information;
A second step of receiving the data selection information of the past, present, and future from the user 1 when the walking activity estimation application 10 inputs the goal as well as the personal information;
A third step in which the walking activity estimation app 10 receives period information of desired data from the user 1 according to reception of the data selection information;
When receiving the period selection information, the walking activity prediction application 10 proceeds to a prediction calculation process when the period selection information includes a data period corresponding to the prediction data, and the prediction calculation process is performed in the walking activity prediction application 10 A fourth step of calculating a value through a prediction filter (30) using a Kalman filter; Lt; / RTI >
The Kalman filter for implementing the predicted part of the step number,
(1) Project the state ahead; (2) Project the error, which calculates the error change for the state estimation and the estimation, and the prediction step for calculating the error change for the state estimation and the estimation; and (1) Compute the Kalman (2) Update estimate with measurement z _k , (3) Calculate the optimal Kalman gain value K and inductively correct the previously obtained value using the error between the predicted value and the actual measured value at the prediction step And a correction step of inductively correcting the previously obtained value by using the error between the predicted value and the actual measured value at the prediction step,
In the prediction step (1, see equation 1)
(1)

The prediction value is estimated through the previous data,

Is the step number predicted value before the correction at the specific time k,

Is obtained in B, the state transition by the state transition A, the user inputs to the step number x _k-1, k-1 from the previous state at a particular time through the user input is u _k,
In the prediction step (2, equation 2)
(2)

A covariance P _k - but estimating, the prediction is defined as u _k = 0 Because of the state transition is not A = 1 because there is not defined, and the user input, Q is defined as a noise covariance predictions,
In the correction step (3, Equation 3)
(3)

The Kalman gain Kk at a specific time k is obtained by using Pk-, H, and R as follows: H is a transfer function, R is a measured noise covariance,
In the correction step (4)
(4)

The predicted step number xk to be obtained at a specific time k is obtained through xk-, Kk, and zk,
In the correction step (5)
(5)

The covariance Pk is corrected,
A fifth step in which the walking activity prediction app 10 outputs a result based on the contents requested by the user 1; And calculating a walking activity amount for the motivation.
delete

Images