JP2024055567A - Travel speed calculation system and travel speed calculation method - Google Patents

Abstract

[Problem] To estimate the walking speed of an analysis subject in an exercise test conducted in a gymnasium or the like in the subject's everyday life. [Solution] A moving speed calculation system 1 of the present invention calculates the walking speed of the analysis subject in an exercise test of physical function based on the moving speed of the analysis subject calculated from multiple pieces of measurement information measured in time series by a non-contact sensor 21. In detail, the walking speed of the analysis subject is calculated based on the approximate maximum value of the moving speed, the average value of the moving speed values equal to or greater than a predetermined threshold, or the moving speed during a moving period defined as a period during which a moving speed equal to or greater than a predetermined threshold is detected. [Selected Figure] Figure 1

Description

The present invention relates to a movement speed calculation system and a movement speed calculation method that detects the movement speed of a person using a sensor.

In recent years, advances in communication and sensor technologies have made it possible to measure human movements and behavior in everyday life. In particular, movement speed is known to correlate with human motor function, and simple measurement methods are needed.

Patent Document 1 is a background technology of this technical field. This patent document 1 states that "the exercise test evaluation system includes an acceleration sensor that acquires acceleration values of the feet of a user performing an exercise test, a heart rate sensor that measures the user's heart rate, and a processing circuit. The processing circuit acquires a first maximum acceleration value in a first time range and a second maximum acceleration value in a second time range after the first time range based on the acceleration values, and calculates a reliability score for the exercise test using the first maximum acceleration value and the second maximum acceleration value. If the reliability score is equal to or greater than a predetermined threshold, the processing circuit acquires the walking speed and heart rate of the user during the exercise test, and estimates the user's maximum oxygen intake using the walking speed and heart rate. If the reliability score is less than the predetermined threshold, the processing circuit estimates the maximum oxygen intake using the walking speed."

With this method, acceleration sensors are attached to the feet of the subject and the reliability of the exercise test is evaluated based on the measured maximum acceleration value, making it possible to properly analyze the walking speed during the exercise test and estimate maximum oxygen intake.

JP 2018-158087 A

According to the technology of Patent Document 1, an acceleration sensor is attached to the foot of the subject to be analyzed, and the reliability of a separately conducted exercise test is evaluated based on the measured maximum acceleration value, thereby making it possible to appropriately analyze the walking speed during the exercise test and estimate the maximum oxygen intake.
However, the method of Patent Document 1 is premised on an acceleration sensor that is attached to the foot of the person being analyzed, and there is a possibility that the person being analyzed may feel aversion to wearing a wearable sensor in their daily lives.

The objective of the present invention is to estimate the walking speed of a subject during an exercise test conducted in a gymnasium or other facility in the subject's everyday life.

To solve the above problem, the moving speed calculation system of the present invention calculates the walking speed of the subject in an exercise test of the subject's physical function based on the moving speed of the subject calculated from multiple pieces of measurement information measured in time series by a non-contact sensor.

According to the present invention, the walking speed in the exercise test is estimated using a non-contact sensor, so that the wearer does not feel uncomfortable about wearing a wearable sensor.
Furthermore, according to the present invention, the walking speed of the subject in the exercise test can be estimated in everyday life, so that the health condition and physical function of the subject can be estimated without carrying out a special exercise test.

FIG. 1 is a block diagram showing a first configuration of a travel speed calculation system. FIG. 2 is a block diagram showing a detailed configuration of a calculation unit. FIG. 13 is a conceptual diagram of a method for calculating a moving speed. 11 is a flowchart illustrating an operation of the travel speed calculation system. FIG. 13 is a block diagram showing a second configuration of the travel speed calculation system. 13 is an example of a method for calculating an approximately maximum value. 13 is an example of a method for calculating an approximately maximum value. 13 is an example of a method for calculating an approximately maximum value. 13 is a display example of an external terminal when the business operator is an insurance company. FIG. 13 is a block diagram showing a third configuration of the travel speed calculation system. 13 is an example of a method in which the calculation unit calculates a moving speed. 13 is an example of a display section for an individual who wants to watch over their elderly parents on an external terminal of a watching service that utilizes a moving speed calculation system. 13 is an example of a display unit for elderly parents on an external terminal of a monitoring service that utilizes a moving speed calculation system. 13 is an example of an interface for an external terminal of a monitoring service that utilizes a moving speed calculation system. FIG. 13 is a block diagram showing a fourth configuration of the travel speed calculation system. 13 is a second example of a display unit for elderly parents on an external terminal of a monitoring service utilizing a moving speed calculation system.

The following describes in detail an embodiment of the present invention with reference to the drawings. However, the present invention is not limited to the following embodiment, and various modifications and application examples within the technical concept of the present invention are also included in its scope.

Figure 1 is a block diagram showing a first configuration of a travel speed calculation system 1. The travel speed calculation system 1 is composed of an information acquisition unit 11, a calculation unit 12, a storage unit 13, and a sensor 21.

Sensor 21 has the function of measuring the moving speed of the subject of analysis or the location information required to calculate it, and representative examples include a radio wave sensor, a sensor using GPS, an image sensor, and a sensor group consisting of at least two sensors.

In this specification, radio wave sensors are defined as a general term for sensors that use radio waves to detect the presence, position, and movement of objects, such as microwave sensors, millimeter wave sensors, and TOF sensors. By using radio wave sensors, the subject of analysis can be measured as a point cloud without contact. Since each point in this point cloud has coordinate information, spatial resolution can be improved as the frequency of the radio wave sensor increases and the number of points increases. In addition, it is possible to calculate the speed of movement from the change in this coordinate information and the time required for it. Furthermore, because radio wave sensors do not infringe on privacy, they can be installed without any hesitation in spaces where people live, such as inside homes and facilities.

In this specification, GPS-based sensors include those that function independently, as well as those that are incorporated into wearable devices such as smartphones and smartwatches and can simultaneously acquire coordinate information and time. This sensor uses a satellite positioning system to acquire the coordinate information of the subject of analysis. It is possible to calculate the speed of movement from the change in this coordinate information and the time required for it.

In this specification, image sensors include cameras with the ability to continuously record still images, cameras with the ability to record video, web cameras, depth cameras, fisheye cameras, etc., as well as thermo viewers and cameras that have been processed to display a model of the human body as a skeleton, etc., to protect privacy. It is possible to obtain coordinate information of the person being analyzed on the screen from these images, and calculate the speed of movement from the change in coordinates and the time required for that.

Methods for acquiring coordinate information of the person being analyzed from an image include a method of defining coordinates based on a part of the human body such as the head, chest, or legs, and a method of treating the pixels representing the person being analyzed as a point cloud and calculating the center of gravity of the point cloud to obtain coordinate information. This center of gravity may be calculated only from the coordinates of the point cloud, or the center of mass taking mass into consideration may be used.

The information acquisition unit 11 acquires the measurement results of the sensor 21 and information 41 about the subject of analysis. The information 41 about the subject of analysis includes, for example, height, weight, date of birth, BMI, body fat percentage, visceral fat level, muscle mass, body water percentage, body age, and the results of a health survey conducted separately.

The information acquisition unit 11 may acquire the measurement results of the sensor 21 by directly communicating with the sensor 21 via, for example, Ethernet, wireless communication, etc. As another example, when sensor data is collected via a gateway to a locally installed PC, the information acquisition unit 11 acquires the sensor data by accessing the PC via a local network or the Internet.

As yet another example, when sensor data is collected directly or via a gateway on a publicly available rental server, the information acquisition unit 11 acquires the sensor data by accessing the server via the Internet. Alternatively, the sensor data may be stored on an external vendor's server and the data may be acquired in a file format such as csv. In this case, the information acquisition unit 11 may acquire information by implementing a function to read csv (Comma Separated Values) data using Python (registered trademark), Java (registered trademark), C language, or the like.

On the other hand, there are multiple methods for acquiring information 41 about the person being analyzed, such as a method in which the system administrator refers to the application documents filled out by the person being analyzed when applying for the service and inputs the information, or a method in which the information input by the system administrator is stored in the storage unit 13 and then referenced.

Furthermore, the information acquisition unit 11 may acquire information other than the measurement values, such as analysis results, in addition to measurement values such as movement speed and position information from the sensor 21. For example, the sensor 21 may measure the coordinate information of the person being analyzed and acquire the time at the same time, and calculate the movement speed, acceleration, and their variance, standard deviation, etc. from the change over time in the coordinate information within the sensor, and the information acquisition unit 11 may acquire these.

As shown in FIG. 2, the calculation unit 12 is composed of a movement speed calculation unit 12a and a maximum movement speed calculation unit 12b. The movement speed calculation unit 12a calculates the movement speed of the subject of analysis using the information received from the information acquisition unit 11, and the maximum movement speed calculation unit 12b uses the movement speeds over a certain period of time in the past to calculate an approximate maximum value of the walking speed in the physical function exercise test. In other words, the calculation unit 12 determines the walking speed of the subject of analysis in the physical function exercise test based on the movement speed of the subject of analysis calculated from multiple pieces of measurement information, and for example, the walking speed of the subject of analysis is set to an approximate maximum value of the movement speed.

A conceptual diagram of a method for calculating the moving speed is shown in Fig. 3. For example, when coordinate information of the analysis subject is received from the information acquisition unit 11, the moving speed calculation unit 12a calculates the moving speed by the following formula (1).

Here, the two orthogonal axes in the plane perpendicular to the vertical direction are the x-axis and y-axis. V is the moving speed, t is time, and the superscript xy of V indicates the speed components in the xy plane, and the subscript t of V indicates the value at time t. Also, xt and yt respectively represent the x-coordinate and y-coordinate at time t. Note that equation (1) may be a calculation formula for only one axis, either the x-axis or the y-axis, or the vertical direction may be the z-axis, and the moving speed may be calculated taking into account the speed component in the z-axis direction.

Next, the maximum moving speed calculation unit 12b calculates an approximate maximum value based on the moving speed over a certain period of time in the past. Here, the data on the moving speed over a certain period of time in the past may be stored inside the maximum moving speed calculation unit 12b and referenced, or the moving speed calculation unit 12a may transmit the result to the storage unit 13 every time it calculates a moving speed, and the maximum moving speed calculation unit 12b may reference the storage unit 13 as necessary.

The storage unit 13 stores at least one of the information acquired by the information acquisition unit 11 and the moving speed calculated by the calculation unit 12. The data may be stored for each subject of analysis, or may be classified by attributes such as the age, sex, residential area, and hobbies and preferences of the subject of analysis. The stored data may also be transmitted to the calculation unit 12, which may then perform a time series analysis of the data.

In this specification, time series analysis is defined as visualizing various types of data using graphs with time on the horizontal axis, calculating maximum, minimum, average, rate of change, moving average, weighted average, variance, standard deviation, etc. of the data values over time, and performing error analysis, polynomial approximation, and other operations. These time series analyses may also include comparing the analysis results with those of people who belong to the same category in terms of age, sex, residential area, lifestyle, occupation, medical history, etc.

Figure 4 is a flowchart explaining the operation of the travel speed calculation system 1. The travel speed calculation system 1 starts processing in response to an operation by an operator of the system (S1). Note that the operator is not particularly limited, and may be a system administrator, a business operator, a person to be analyzed, an individual, etc.

First, the information acquisition unit 11 acquires data from the sensor 21 and information 41 about the person being analyzed (S2), and transmits the acquired information to the calculation unit 12 and the storage unit 13 (S3). The calculation unit 12 calculates the moving speed using information received from at least one of the information acquisition unit 11 or the storage unit 13 (S4), and transmits the calculated moving speed to the storage unit 13 (S5).

The storage unit 13 transmits the movement speed calculated by the calculation unit 12 together with the past accumulated data of the subject of analysis or other people to the calculation unit 12 (S6). The calculation unit 12 uses the current movement speed and the past accumulated data to calculate the maximum movement speed (S7), and transmits the calculated maximum movement speed to the storage unit 13 (S8). The movement speed calculation system 1 may end immediately (S9), or may implement a loop process that returns to immediately after the start of processing (S1) by setting it in advance.

The following describes in detail the operation of the system, using FIG. 5, assuming that an insurance company utilizes the travel speed calculation system 1. Note that this embodiment may also be used for businesses that provide services and products to the elderly, such as day care providers and senior housing operators.

FIG. 5 is a block diagram showing a second configuration of the moving speed calculation system.
First, the business operator 63 prepares services and products that utilize the moving speed calculation system 1 stored in the server 53 managed by the system operator. This utilization method includes, in addition to utilizing the moving speed and its approximate maximum value, estimating various indices related to physical function, such as frailty level, locomotive syndrome level, sarcopenia, risk of falling, and the like, based on the results of time series analysis of the moving speed and its approximate maximum value.

When a subscription application for a service or product is made, the sensor 21 is sent to the subscriber 64's home. After installation, the sensor 21 secures power by connecting to a power source, using a battery, or by environmental power generation such as sunlight or vibration, and begins measurement. The sensor data is sent to the external server 51 directly or via a gateway.

In this embodiment, a specific example will be described in which a millimeter wave sensor, a type of radio wave sensor, is used as the sensor 21. The millimeter wave sensor is installed on a wall 1.8 m above the ground, and the power cord is connected to an outlet to ensure power. The measurement range is approximately 1.5 m left and right from the center of the sensor body, and 5 m deep, and it is desirable to install the sensor within this range in a location where multiple daily activities are performed. For example, by installing the sensor facing the kitchen in a living/dining room, it is possible to include relaxation, eating, meal preparation, and cooking within the measurement range.

As in this embodiment, by measuring both actions that involve relatively little movement (relaxing) and actions that involve movement (food preparation, cooking), it is possible to estimate the noise that is generated simply by the presence of a human from the former and remove the noise from the latter, which has the advantage of improving measurement accuracy.

A separate dedicated router may be installed to transmit data from the millimeter wave sensor to the external server 51. The type of sensor used may be changed depending on the service and product. For example, a combination of radio wave sensors, sensors using GPS, image sensors, human presence sensors, illuminance sensors, temperature, humidity and air pressure sensors, door opening/closing sensors, multiple vibration sensors, pressure sensors and wearable sensors may be used.

Here, some or all of the human presence sensor, illuminance sensor, and temperature, humidity, and air pressure sensor may be combined into one environmental sensor. In addition, data linking with other commercially available devices, such as a weighing scale or body composition scale that can be connected online, may also be included.

The external server 51 is, for example, a server of the manufacturer of the sensor 21. In this case, the sensor data is encrypted inside the sensor before being transmitted, and then decrypted in the external server 51, which has the advantage of ensuring secure information communication. In addition, the sensor data is stored in the external server 51 together with the measurement time, and can be acquired by the information acquisition unit 11 of the moving speed calculation system 1 via an API provided by the manufacturer.

The owner of the external server 51 is not limited, and it may be a server managed by a business operator, a server managed by a sensor manufacturer, or a server that can be rented to the public. If the external server 51 is a server managed by a business operator 63 or a server that can be rented to the public with a certain level of security, the travel speed calculation system 1 may be stored therein.

Also, the business operator and the system operator may store data in different environments, read the data into the travel speed calculation system 1, and connect within the system. However, these servers must be able to communicate with the travel speed calculation system 1 built in either the system operator's internal environment, the business operator's internal environment, or a cloud environment.

The information acquisition unit 11 of the movement speed calculation system 1 acquires data from the sensor 21 and information 41 of the person being analyzed within the movement speed calculation system 1, and transmits it to the calculation unit 12 and the storage unit 13. The storage unit 13 may receive information 41 (physical information) of the person being analyzed from the information acquisition unit 11, search for historical data such as past physical information and movement speed of the same person, and transmit this to the information acquisition unit 11 and the calculation unit 12. This allows the calculation unit 12 to perform time series analysis.

One method for acquiring information 41 of the subject of analysis is, for example, to write it down on an application form for a service or product. With this method, it is possible to acquire information 41 of the subject of analysis and at the same time obtain written consent for the handling of personal information. Other methods may also be used, such as allowing the information to be entered into the sensor 21, recording it on an application form for a service or product, allowing additional input from the personal page on the business's website, providing separate measuring equipment such as a grip strength meter or body composition scale, or acquiring the information in collaboration with a third party such as another business, local government, or non-profit organization.

The calculation unit 12 calculates the moving speed and its approximate maximum value from the information of the sensor 21 received from the information acquisition unit 11 and the information 41 of the subject of analysis. Specifically, first, the moving speed calculation unit 12a calculates the moving speed from the coordinate information of the subject of analysis received from the information acquisition unit 11 and the above formula (1). At this time, a threshold value may be set for the moving speed, and a value equal to or greater than the threshold is determined as moving, and a value equal to or less than the threshold is determined as stationary, and the moving speed may be calculated for a moving period. In other words, the walking speed of the subject of analysis can be calculated based on the moving speed during a moving period defined as a period during which a moving speed equal to or greater than a predetermined threshold is detected.

One method for determining this threshold is to first measure actions that do not involve movement, such as sitting or sleeping, and then consider the calculated movement speed to be noise, and use the maximum, minimum, average, standard deviation, and multiples of these, of the noise. For example, if the standard deviation of the movement speed calculated for sitting is 0.1 m/s, then it may be possible to evaluate noise at a 95% confidence interval by setting twice that value, or 0.2 m/s, as the threshold. Setting this threshold can reduce sensor noise and improve the measurement accuracy of movement speed.

Next, the maximum movement speed calculation unit 12b calculates the approximate maximum value based on the movement speed calculated by the movement speed calculation unit 12a. Methods for calculating the approximate maximum value include calculating the maximum value for a period such as an hour, day, week, month, or year, and examples are shown in Figures 6A, 6B, and 6C. In other words, the walking speed of the person being analyzed can be found based on the movement speed for each fixed period.

Figure 6A shows an example of calculating the approximate maximum value of the movement speed for each month. Since the approximate maximum value of the movement speed for each month is the same or has an increasing trend, the subject of the analysis may be determined to be healthy. This makes it possible to reduce the human and time costs required for the analysis.

On the other hand, FIG. 6B is an example in which the approximate maximum value of the moving speed is similarly calculated for each month, but since the approximate maximum value of the moving speed for each month has a downward trend, it may be possible to detect the possibility that the subject's physical function is declining. In this case, as shown in FIG. 6C, the month in which the approximate maximum value of the moving speed decreased the most may be focused on, and the period for calculating the approximate maximum value of the moving speed may be subdivided from monthly to weekly. This allows the change in the approximate maximum value of the moving speed to be confirmed in detail, and the decline in physical function can be more accurately estimated.

In addition, movement may be detected first using the threshold value, a maximum value may be calculated for each period during which movement is detected, and these maximum values may then be averaged again for periods such as hours, days, weeks, months, or years. Alternatively, the approximate maximum value may be calculated by averaging values above a certain percentage, such as the top 10%, of the calculated movement speeds. By using these average values to calculate the approximate maximum value, it is possible to smooth out abnormal values, variations due to measurement, and variations in the daily behavior of the subject of analysis, and to calculate an approximate maximum movement speed that more appropriately estimates physical function. In other words, the walking speed of the subject of analysis may be the average value of movement speed values above a predetermined threshold value.

When the sensor 21 includes at least two of a human presence sensor, an illuminance sensor, a temperature, humidity and air pressure sensor, a door opening/closing sensor and a vibration sensor installed in different locations, the movement speed of the subject of analysis may be calculated based on the difference in the time when these sensors react. For example, if a human presence sensor is installed in the living room and a door opening/closing sensor is installed in the toilet, when an elderly person sleeping in the bedroom passes the living room and moves to the toilet, the human presence sensor in the living room will react first, and then the toilet door opening/closing sensor will react. In this case, the movement speed can be calculated by dividing the distance between the living room and the toilet by the difference in the reaction time of the human presence sensor in the living room and the reaction time of the toilet door opening/closing sensor.

The calculation unit 12 transmits the calculated moving speed, its approximate maximum value, and the time series analysis results thereof to the storage unit 13 and the display unit 14 of the external terminal 61 as the walking speed in the physical function exercise test. In other words, the walking speed in the physical function exercise test of the subject is calculated based on the moving speed of the subject calculated from the measurement information. The storage unit 13 classifies and stores the results calculated by the calculation unit 12 for each subject and for categories such as the subject's age, sex, residential area, lifestyle, occupation, and medical history.

The display unit 14 displays the received results to the business operator 63 or the system operator 69. The display method may be different between the business operator 63 and the system operator 69.

For example, the business operator 63 may be shown the current movement speed of the person being analyzed, the approximate maximum movement speed, and the results of their time series analysis, as well as the results of conversion into information that is particularly required by each business operator. Figure 7 shows an example of the display on the external terminal 61 when the business operator is an insurance company. The insurance company may be provided with material for inferring the applicability of insurance for the current person being analyzed by showing the past insurance application records and movement speeds of other people, the approximate maximum movement speeds, and the results of their time series analysis.

Specifically, as shown in FIG. 7, the horizontal axis may represent the movement speed and its change over time, and the vertical axis may display the number of insurance applications. The horizontal axis may be selected to represent the average or approximate maximum value of the movement speed. In addition, the calculation period for the change over time may be selected from one day, one week, one month, three months, six months, or one year, or may be set arbitrarily. Based on this information, the insurance company can consider intervention measures to correct the movement speed of the subject of analysis, the approximate maximum value of the movement speed, and the time series analysis results of these in a direction that reduces the possibility of insurance application.

This intervention method includes, for example, recommendations such as "Take a 30-minute walk every day" or "There's a bazaar being held in the town, why don't you join it?". In addition, a day care service provider may provide information for considering the impact that the content of the day care service has on the subject by simultaneously showing the date on which the day care service was provided, the movement speed, the approximate maximum movement speed, and the results of a time series analysis of these.

Based on this information, the day service provider can consider the optimal program for each person being analyzed. Alternatively, if the provider has personnel with medical knowledge such as physical therapists, the moving speed of the person being analyzed, the approximate maximum moving speed, and the results of their time series analysis may be output as speed information from a walking test in which a person is assisted. Based on this information, the physical therapist or other personnel with medical knowledge may estimate various indices related to physical function, such as the degree of frailty, degree of locomotor syndrome, sarcopenia, and risk of falling.

For example, in addition to the current movement speed of the subject of analysis, the approximate maximum value of the movement speed, and the time series analysis results thereof, the system administrator 69 may be shown information regarding the operating state of the sensor 21, information regarding the battery status of the sensor 21, the sensor readings not processed by the calculation unit 12, and information regarding the date, time, number of times, duration, and content of the operator's reference to the system. Furthermore, the movement speed, the approximate maximum value of the movement speed, and the time series analysis results thereof may be shown as speed information of a walking test in which a person is present, and various indices regarding physical function, such as frailty level, locomotive syndrome level, sarcopenia, risk of falling, etc. may be predicted.

The business operator 63 can provide feedback to the system administrator about the display content. The means of feedback include the interface 62 of the external terminal 61, oral communication, email, posting via the system operator's homepage, etc. The system operator may change the display content based on the feedback from the business operator.

The following describes in detail the operation of the system using FIG. 8, assuming that an individual 66 who wants to watch over his/her elderly parent utilizes the movement speed calculation system 1. FIG. 8 is a block diagram showing a third configuration of the movement speed calculation system 1. Note that in the following description, content that overlaps with Example 1 will be omitted, and only the differences will be described.

First, an individual 66 who wants to keep an eye on his/her elderly parent subscribes to a monitoring service that utilizes the moving speed calculation system 1 stored in the server 53 managed by the system operator. When subscribing to this service, various types of sensors 21 are sent to the individual 66 who wants to keep an eye on his/her elderly parent according to the service plan. When the individual 66 who wants to keep an eye on his/her elderly parent installs the sensor 21 in the home of the elderly parent 67, the sensor 21 secures a power source and starts measuring, and transmits and stores the data to a server or the like.

In FIG. 8, the information acquisition unit 11 of the moving speed calculation system 1 exchanges information with the storage unit 13 and the sensor 21 provided in the external server 52 managed by the system administrator 69, and acquires the data of the sensor 21 and the information 41 of the subject of analysis into the moving speed calculation system 1. Here, the subject of analysis in this embodiment is an elderly parent 67.

The external server 51 may also store the behavior recognition unit 15 and transmit the behavior recognition results to the information acquisition unit 11. Furthermore, the method of acquiring the information 41 of the subject of analysis may be entered on a subscription application form for a service or product, or may be input by the individual 66 who wants to watch over their elderly parent or the elderly parent 67 himself/herself through the interface 62 of the external terminal 61.

The behavior recognition unit 15 has a function of recognizing the daily behavior of the subject of analysis within the home based on data from the sensors 21. Here, daily behavior includes types such as going out, sleeping, relaxing, eating, cooking, cleaning, etc. In addition, the recognition method may, for example, use a sensor that can detect the presence of a human being in a specific place, such as a human presence sensor, a millimeter wave sensor, or an image sensor, to associate the location with the behavior and recognize it; cluster data from at least two or more types of sensors 21, and associate the sensor characteristics that react in each cluster with the behavior and recognize it; or learn the association between the behavior self-reported by the subject of analysis and the sensor data by machine learning or the like, and create a classifier that classifies the sensor data into behavior.

Specific examples of behavior recognition methods using clustering include data belonging to a certain cluster that, if there is no reaction from the human presence sensor, indicates that the person is out, and if there is a reaction from the human presence sensor in the bedroom at night and the reading from the illuminance sensor is small, indicates that the person is asleep.

The calculation unit 12 acquires information from the information acquisition unit 11 or the storage unit 13 provided in the external server 52, and calculates the movement speed, the approximate maximum movement speed, and their time series analysis results from the coordinate information of the person being analyzed. The method by which the calculation unit 12 calculates the movement speed will be specifically explained below with reference to FIG. 9. In FIG. 9, t is time, and the smaller the subscript value, the earlier the time. Furthermore, L is a line connecting the coordinates at each time, and the subscript indicates the time of the two points connected.

First, the calculation unit detects linear movement using coordinate information for at least three points in the time direction. Specifically, using the three points of time t1, t2, and t3, it calculates the angle θ between a line L12 connecting the coordinates of times t1 and t2 and a line L23 connecting the coordinates of times t2 and t3, and detects linear movement if the value of θ is within a predetermined range.

The range of this value may be set arbitrarily, but may be set to, for example, 90 degrees or more and 270 degrees or less, provided that there is at least a forward component. Alternatively, a change in the direction of travel of 20 degrees or more may be defined as a turn, and the range of the θ value may be set to 160 degrees or more and 200 degrees or less, which is not defined as a turn.

Next, for this linear movement and the detected motion, the movement speed, the approximate maximum value of the movement speed, and the time series analysis results of these are calculated as the walking speed in the physical function exercise test. This makes it possible to evaluate the movement speed with a particular focus on the linear movement of the subject. In other words, the linear movement of the subject is detected using the measurement information from at least three temporally consecutive points, and the walking speed is calculated based on the movement speed during the linear movement period in which the linear movement was detected.

The 5m walking test, which is used as an index of physical function in physical measurement tests for elderly people, measures linear movement, so by having the calculation unit 12 also measure linear movement, there is an advantage in that the physical function of the subject of analysis can be more appropriately estimated. Note that linear movement may be detected using four or more consecutive points in the time direction.

For example, in FIG. 9, the angle φ may be calculated similarly to the angle θ, and linear movement may be detected if all of them are within the range of the above values. Alternatively, the angles θ and φ may be defined as the angle between a line connecting the start point and end point of the movement and a line L at each time, and linear movement may be detected if some or all of these acute angles are below a threshold value such as 20 degrees. Finally, the calculation unit 12 transmits the calculation result to the storage unit 13 provided in the external server 52.

The storage unit 13 is provided in an external server 52 that is different from the server in which the travel speed calculation system 1 is provided, and can exchange data with the information acquisition unit 11, the calculation unit 12, and the display unit 14 provided in the external terminal 61. This external terminal 61 may be a smart device, such as a smartphone or tablet, that can be used by an individual 66 who wants to watch over their elderly parent or the elderly parent 67.

Figure 10 shows an example of a display unit 14 for an individual 66 who wants to watch over their elderly parents on an external terminal 61 of a monitoring service that utilizes the moving speed calculation system 1. The individual 66 who wants to watch over their elderly parents can check the type of current daily activity, the moving speed, the approximate maximum value of the moving speed, and the time series analysis results thereof through the display unit 14. In other words, the moving speed calculation system 1 determines and displays the type of activity and walking speed of the person being analyzed.

The time series analysis result of the approximate maximum value of the moving speed here is the result of calculating the approximate maximum value of the moving speed for any fixed period such as one hour, one day, one week, one month, three months, half a year, one year, etc., and performing a time series analysis on these approximate maximum values. In addition, the method of displaying the results may be selected to be either by type of daily activity or by all daily activities.

Furthermore, based on the time series analysis results, recommendations for improving the physical function of the subject of analysis may be displayed. Specifically, as shown in FIG. 10, if the movement speed, the approximate maximum value of the movement speed, and the time series analysis results thereof are on a downward trend, a recommendation such as "Mr./Ms. XX doesn't seem to be in good spirits lately. Would you like to go out together this weekend?" may be displayed.

Note that in Figure 10, graphs are displayed for all actions as examples of movement speed, approximate maximum movement speed, and the results of their time series analysis, but similar graphs may be displayed for individual actions such as sleeping, eating, relaxing, bathing, using the toilet, and doing laundry.

Figure 11 shows an example of the display unit 14 for an elderly parent 67 on an external terminal 61 of a monitoring service that utilizes the movement speed calculation system 1. The behavior recognition results may be displayed in a scheduler format to the elderly parent 67 so that the parent can reflect on his or her own lifestyle habits. Also, recommendations regarding a healthy lifestyle may be displayed, taking into account changes in lifestyle habits and movement speed. For example, the recommendation may be something like, "Go outside today, leading a healthy lifestyle. Make sure to exercise tomorrow as well."

In addition, individuals who wish to watch over elderly parents can check the received results through the display unit 14, and can also provide feedback on their evaluation of the displayed results to the system operator, manager, business operator, etc. through the interface 62 of the external terminal 61. Figure 12 shows an example of the interface of the external terminal 61 for a watching service that utilizes the moving speed calculation system 1. This interface not only provides feedback on a 10-point evaluation of the displayed results, but can also change the information of the person being analyzed.

The following describes in detail the operation of the system, using FIG. 13, on the assumption that an individual 68 who wants to check his/her own physical function utilizes the movement speed calculation system 1. FIG. 13 is a block diagram showing a fourth configuration of the movement speed calculation system 1. Note that in the following description, content that overlaps with Examples 1 and 2 will be omitted, and only the differences will be described.

First, an individual 68 who wants to check his/her physical function subscribes to a health support service that utilizes the movement speed calculation system 1. The service starts by installing dedicated software in an external terminal 61 such as the individual's smart device, and data measured by the external terminal 61 or the sensor 21 is collected in the information acquisition unit 11 directly or via the external server 51. Here, the sensor 21 in this embodiment may include a wearable device.

This data includes coordinate information measured using a GPS, acceleration information measured using an acceleration sensor and the number of steps estimated using the same, posture information measured using a gyro sensor, direction information measured using a magnetic sensor, biometric information measured using a fingerprint sensor, camera, etc., air pressure information measured using an air pressure sensor, surrounding distance information measured using a LiDAR sensor, and illuminance information measured using an illuminance sensor. Note that this embodiment may target outdoor behavior in addition to indoor behavior.

The calculation unit 12 calculates the moving speed, the approximate maximum value of the moving speed, and the time series analysis results of the moving speed of the subject of analysis as the walking speed in the physical function exercise test based on the data collected by the information acquisition unit. Here, if the data is coordinate information measured using a GPS, the moving speed can be calculated using the formula (1). Note that an arbitrary threshold value, for example 2.0 m/s or more, may be set, and values above the threshold value may be excluded when calculating the approximate maximum value of the moving speed of the subject of analysis. By setting the threshold value appropriately, it is possible to detect the running state and exclude it from the analysis.

This has the advantage that only the walking movement of the subject can be analyzed, similar to the 5m walking test used as an indicator of physical function in physical measurement tests for elderly people, allowing for a more accurate estimation of the subject's physical function.

In addition, the acceleration, which is the change over time in the moving speed, may be calculated at the same time, and the moving speed during the period when the absolute value of the acceleration is below a certain threshold and movement is detected, may be averaged and set as the approximate maximum value. In other words, the walking speed of the subject of analysis is set as the average value of the moving speed during the period when the absolute value of the change in moving speed is below a predetermined threshold. This allows the moving speed to be calculated while reducing the effects of acceleration and deceleration, allowing for a more appropriate estimation of the subject's physical function. The calculated moving speed, the approximate maximum value of the moving speed, and the results of their time series analysis are stored in the external server 52 that stores the accumulation unit 13.

The storage unit 13 can send and receive data to and from an external terminal 61 of an individual 68 who wishes to check their own physical function. FIG. 14 shows a second example of the display unit of an external terminal 61 for a monitoring service that utilizes the movement speed calculation system 1. An individual who wishes to check their own physical function can check the approximate maximum, average, and minimum acceleration and movement speed, as well as their time series analysis results, movement route, number of steps, etc., through the display unit 14.

For example, in addition to displaying information on the number of steps taken, movement speed, acceleration, and movement route for each day, it is also possible to display the history of past exercise in graph form and display comments on how to maintain physical function. This display content may also include recommendations such as, "You have developed an exercise habit this month. Let's continue doing it in the future."

REFERENCE SIGNS LIST 1 Movement speed calculation system 11 Information acquisition unit 12 Calculation unit 12a Movement speed calculation unit 12b Maximum movement speed calculation unit 13 Storage unit 14 Display unit 15 Behavior recognition unit 21 Sensor 41 Information of analysis subject 51 External server 52 External server that stores the storage unit 53 Server managed by system operator 61 External terminal 62 Interface 63 Business operator 64 Service subscriber 66 Individual who wants to watch over elderly parent 67 Elderly parent 68 Individual who wants to check their own physical function 69 System operator

Claims (11)

A movement speed calculation system characterized by calculating the walking speed of a subject in an exercise test of physical function based on the movement speed of the subject calculated from multiple measurement information measured in time series by a non-contact sensor. The moving speed calculation system according to claim 1 ,
A movement speed calculation system, wherein the walking speed of the person being analyzed is approximately the maximum value of the movement speed. The moving speed calculation system according to claim 1 ,
A movement speed calculation system, characterized in that the walking speed of the person being analyzed is an average value of the movement speed values that are equal to or greater than a predetermined threshold. The moving speed calculation system according to claim 1 ,
A movement speed calculation system, characterized in that the walking speed of the person being analyzed is calculated based on the movement speed during a movement period, which is defined as a period during which the movement speed is detected to be equal to or greater than a predetermined threshold. The moving speed calculation system according to claim 1 ,
A movement speed calculation system, characterized in that the walking speed of the person to be analyzed is calculated based on the movement speed for each fixed period of time. The moving speed calculation system according to claim 1 ,
A movement speed calculation system, characterized in that the walking speed of the person being analyzed is an average value of the movement speed during a period in which the absolute value of the change in the movement speed is equal to or less than a predetermined threshold value. The moving speed calculation system according to claim 1 ,
A moving speed calculation system comprising: determining a type of behavior of the subject of analysis recognized from the measurement information; and determining the walking speed of the subject of analysis. The moving speed calculation system according to claim 1 ,
A movement speed calculation system characterized by detecting the linear movement of the subject of analysis using the measurement information of at least three temporally consecutive points, and calculating the walking speed based on the movement speed during the linear movement period in which the linear movement was detected. The moving speed calculation system according to claim 1 ,
A moving speed calculation system, comprising: a moving speed calculation unit that calculates the walking speed by excluding, as a running state, the moving speed whose absolute value is equal to or greater than a predetermined threshold value. The moving speed calculation system according to claim 1 ,
A moving speed calculation system comprising a display unit that displays a physical condition including the walking speed. The moving speed of the subject is calculated from multiple pieces of measurement information measured in time series by the non-contact sensor,
A method for calculating a moving speed, comprising the steps of: determining a walking speed in an exercise test of the subject's physical function based on the calculated moving speed;

Images