Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device, an information processing method and an information processing program that determine reliability of walking data for use in a walking test.SOLUTION: An arithmetic unit 10 is caused to learn walking teacher data indicating a walking state of a pedestrian, and a machine learning model with reliability of the walking teacher data for a prescribed walking test, and acquires walking data indicating a walking state of an object person 200 detected using sensors 220, 230 attached to the object person 200, via a portable device 210 possessed by the object person 200, in order to estimate reliability of the acquired walking data for a prescribed walking test, according to the learnt machine learning model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program that determine the reliability of walking data indicating a subject's gait when the walking data is subjected to a predetermined walking test that determines the soundness of the subject's walking. [Background technology]

[0002] In clinical practice, a 10-meter walking test is conducted to determine the health of a subject's walking ability, measuring the subject's walking speed and number of steps over a 10-meter distance. The 10-meter walking test assesses the subject's condition when walking a 10-meter straight course, but because it requires preparation such as determining the section the subject will walk, it is difficult for individuals to perform the test on a daily basis for the purpose of health management.

[0003] Attempts have been made to determine the soundness of a subject's walking based on their walking in everyday life, rather than when they walk a specific course, as in the 10-meter walking test. Patent Document 1 discloses an invention that recognizes the walking steps that indicate the individual steps taken by a subject when walking from information detected by an acceleration sensor or the like provided in a mobile device such as a smartphone, and further identifies the type of walking step, such as going up a step, going down a step, or going up a slope. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6457346 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the invention described in Patent Document 1 uses data detected by a sensor to recognize and classify the walking steps of a subject without judging whether the data is good or bad, which results in poor accuracy in classifying the walking step types. In everyday walking, a subject may find it difficult to walk at their own pace in a crowded area, and in such cases, the subject is unable to walk according to their natural walking ability. Also, there may be times when the subject starts running for some reason, and in such cases, the subject is unable to walk according to their natural walking ability.

[0006] In consideration of the above, an object of the present invention is to provide an information processing device, an information processing method, and an information processing program for determining the reliability of walking data used in a walking test. [Means for solving the problem]

[0007] To achieve the above objectives According to the first aspect The information processing device includes an acquisition unit that acquires, via a portable device carried by the subject, walking data that indicates the gait of the subject detected by a sensor worn by the subject, and a calculation unit that estimates the reliability of the walking data acquired by the acquisition unit in a predetermined walking test using walking instructor data that indicates the gait of the walker and a machine learning model that has been trained using the reliability of the walking instructor data in the predetermined walking test. The portable device detects acceleration and angular velocity and is capable of estimating a current position by satellite positioning, the sensor includes an arm sensor that detects acceleration and angular velocity of the subject's arm, and a foot sensor that detects acceleration and angular velocity of the subject's foot and pressure on the soles of the feet, and the calculation unit applies gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current position determined by the satellite positioning, to a trained machine learning model to estimate the reliability of the walking data in the predetermined walking test. .

[0008] According to the first aspect According to the information processing device, the reliability of the walking data to be used in the walking test can be estimated, and only the highly reliable walking data can be used in the walking test, thereby improving the accuracy of analysis in the walking test. In addition, walking data can be acquired using arm sensors, foot sensors, and portable devices that the subject can wear comfortably on a daily basis. Furthermore, the reliability of the walking data can be estimated based on gait analysis data that clearly indicates the walking of the subject 200, such as the walking speed, stride length, and balance of the subject, calculated from the walking data including the acceleration and angular velocity detected by the sensors, the pressure on the soles of the feet, and the current position obtained by the satellite positioning.

[0013] To achieve the above objectives According to the second aspect The information processing method includes training a machine learning model using walking instructor data indicating a pedestrian's gait and the reliability of the walking instructor data in a predetermined walking test. No. 1 and acquiring, via a portable device carried by the subject, walking data indicating the gait of the subject detected by a sensor attached to the subject. No. 2and estimating the reliability of the acquired walking data in the predetermined walking test using a trained machine learning model. Third and An information processing method, wherein the portable device is capable of detecting acceleration and angular velocity and estimating a current location by satellite positioning, and the sensor includes an arm sensor that detects the acceleration and angular velocity of the subject's arm, and a foot sensor that detects the acceleration and angular velocity of the subject's foot and the pressure on the soles of the feet, and the third step applies gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current location by the satellite positioning, to a trained machine learning model to estimate the reliability of the walking data in the specified walking test.

[0014] According to the second aspect According to the information processing method, the reliability of the walking data to be used in the walking test is estimated, and only the highly reliable walking data is used in the walking test, thereby improving the accuracy of analysis in the walking test. In addition, walking data can be acquired using arm sensors, foot sensors, and portable devices that the subject can wear comfortably on a daily basis. Furthermore, the reliability of the walking data can be estimated based on gait analysis data that clearly indicates the walking of the subject 200, such as the walking speed, stride length, and balance of the subject, calculated from the walking data including the acceleration and angular velocity detected by the sensors, the pressure on the soles of the feet, and the current position obtained by the satellite positioning.

[0015] To achieve the above objectives According to the third aspect The information processing program causes a computer to function as a learning unit that trains a machine learning model using walking instructor data that indicates a pedestrian's gait and the reliability of the walking instructor data in a predetermined walking test, an acquisition unit that acquires walking data that indicates the subject's gait detected by a sensor attached to the subject via a portable device carried by the subject, and a calculation unit that estimates the reliability of the walking data acquired by the acquisition unit in the predetermined walking test using the trained machine learning model. An information processing program, wherein the portable device detects acceleration and angular velocity and is capable of estimating a current position by satellite positioning, the sensor includes an arm sensor that detects the acceleration and angular velocity of the subject's arm, and a foot sensor that detects the acceleration and angular velocity of the subject's foot and the pressure on the soles of the feet, and the calculation unit applies gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current position determined by the satellite positioning, to a trained machine learning model to estimate the reliability of the walking data in the specified walking test.

[0016] According to the third aspect According to the information processing program, the reliability of the walking data to be used in the walking test can be estimated, and only highly reliable walking data can be used in the walking test, thereby improving the accuracy of analysis in the walking test. In addition, walking data can be acquired using arm sensors, foot sensors, and portable devices that the subject can wear comfortably on a daily basis. Furthermore, the reliability of the walking data can be estimated based on gait analysis data that clearly indicates the walking of the subject 200, such as the walking speed, stride length, and balance of the subject, calculated from the walking data including the acceleration and angular velocity detected by the sensors, the pressure on the soles of the feet, and the current position obtained by the satellite positioning. [Effects of the Invention]

[0017] As described above, the information processing device, information processing method, and information processing program according to the present invention make it possible to determine the reliability of walking data used in a walking test. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a schematic diagram illustrating a configuration of an information processing device according to an embodiment of the present invention. [Figure 2]FIG. 2 is a block diagram showing an example of a specific configuration of a calculation device according to the present embodiment. [Figure 3] FIG. 2 is a functional block diagram of mobile devices and a computing device of the information processing device according to the embodiment. [Figure 4] (A) is a schematic diagram showing an example of walking data collected by a walking data collection unit of a portable device, (B) is a schematic diagram showing an example of walking data collected by a motion estimation data collection unit of a portable device, and (C) is a schematic diagram showing an example of surrounding situation data collected by a surrounding situation data collection unit of a portable device. [Figure 5] (A) is a schematic diagram showing an example of gait analysis data indicating walking conditions, (B) is a schematic diagram showing an example of movement estimation data, and (C) is a schematic diagram showing an example of a judgment rank. [Figure 6] 10 is a flowchart showing an example of a walking data determination process of the information processing device according to the present embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0019] An information processing device 100 according to this embodiment will be described below with reference to Fig. 1. The information processing device 100 shown in Fig. 1 includes a communication device 110 that acquires data from a plurality of subjects 200 carrying portable devices 210 capable of wireless communication, such as smartphones, that have a constant connection function to a network, a data storage 120 that accumulates the data received by the communication device 110, and a calculation device 10 that determines the reliability of walking data indicating the gait of the subjects 200 acquired from the data storage 120 in a predetermined walking test. The portable devices 210 are capable of detecting ambient sounds, and are equipped with an altimeter that detects altitude based on changes in air pressure, etc., and are capable of positioning using a GNSS (Global Navigation Satellite System) function. They are also equipped with an IMU (Inertial Measurement Unit) that includes an acceleration sensor that detects acceleration in three orthogonal axial directions, i.e., X, Y, and Z, and a gyro sensor that detects angular velocity in three orthogonal axial directions, i.e., X, Y, and Z. Furthermore, the portable device 210 is configured to receive data detected by the arm sensor 220 and the foot sensor 230 worn by the subject 200, and to transmit the received data to the data storage 120 via the communication device 110. If the altitude detection accuracy of the GNSS function of the portable device 210 is good, the altitude may be detected by the GNSS function instead of the altimeter.

[0020] The arm sensor 220 is equipped with an IMU capable of detecting acceleration and angular velocity in three orthogonal axial directions of X, Y, and Z, and the foot sensor 230 is equipped with an IMU capable of detecting acceleration and angular velocity in three orthogonal axial directions of X, Y, and Z, and a foot pressure sensor that detects foot pressure, which is pressure on the soles of the feet of the subject 200. The foot pressure sensor is, for example, a sensor that uses a piezoelectric element or the like to detect foot pressure when the subject 200 lands while walking.

[0021] As will be described later, the data storage 120 is a data server equipped with a database, and the computing device 10 is a computer capable of executing advanced computing processes at high speed. Each of the data storage 120 and the computing device 10 may be a standalone server, or may be a cloud that can distribute the processing load. The data storage 120 and the computing device 10 may be the same server.

[0022] 2 is a block diagram showing an example of a specific configuration of the arithmetic device 10 according to an embodiment of the present invention. The arithmetic device 10 is configured to include a computer 40. The computer 40 includes a CPU (Central Processing Unit) 42, a ROM (Read Only Memory) 44, a RAM (Random Access Memory) 46, and an input / output port 48. As an example, the computer 40 is preferably a model capable of executing advanced arithmetic processing at high speed.

[0023] In the computer 40, the CPU 42, ROM 44, RAM 46, and input / output port 48 are connected to one another via various buses such as an address bus, a data bus, and a control bus. The input / output port 48 is connected to various input / output devices, such as a display 50, a mouse 52, a keyboard 54, a hard disk (HDD) 56, and a disk drive 60 that reads information from various disks (e.g., CD-ROM, DVD, etc.) 58.

[0024] Furthermore, a network 62 is connected to the input / output port 48, enabling information to be exchanged with various devices connected to the network 62. In this embodiment, a data storage 120, which is a data server connected to a database (DB) 122, is connected to the network 62, enabling information to be exchanged with the DB 122.

[0025] The DB 122 stores walking data of a plurality of subjects 200 acquired via the communication device 110. Data may be stored in the DB 122 not only via the communication device 110 but also by the computer 40 or other devices connected to the network 62.

[0026] In this embodiment, it is described that walking data etc. of multiple subjects 200 are stored in DB122 connected to data storage 120, but the information of DB122 may also be stored in an external storage device such as HDD56 built into computer 40 or an external hard disk.

[0027] A program related to machine learning using a neural network is installed in the HDD 56 of the computer 40. In this embodiment, the CPU 42 executes the program to start machine learning and construct a trained machine learning model based on the machine learning. Furthermore, the constructed trained machine learning model is used to determine the reliability of the walking data acquired from the subject 200. The CPU 42 also displays the processing results of the program on the display 50.

[0028] There are several ways to install the machine learning program of this embodiment into the computer 40. For example, the program can be stored on a CD-ROM, DVD, or the like together with a setup program, and the disk can be inserted into the disk drive 60, and the setup program can be executed by the CPU 42 to install the program into the HDD 46. Alternatively, the program can be installed into the HDD 46 by communicating with another information processing device connected to the computer 40 via a public telephone line or the network 62.

[0029] Next, various functions realized by the CPU 42 of the arithmetic device 10 executing a machine learning program will be described. The machine learning program has a learning function that constructs a trained machine learning model by training an AI mathematical model using previously prepared walking teacher data, an evaluation function that evaluates the performance of the trained machine learning model using walking teacher data different from that used during training, and a selection function that selects a machine learning model with excellent performance. When the CPU 42 executes the machine learning program having these functions, the CPU 42 functions as a learning unit, an evaluation unit, and a selection unit. In this embodiment, as an example of the AI ​​mathematical model used to construct the trained machine learning model, a neural network, such as an RNN (Recurrent Neural Network), in which processing units that linearly transform inputs are connected in a network form is used. The walking instructor data used in the learning function and the selection function is, for example, data clearly indicating the walking of the subject 200, such as the walking speed, stride length, and balance of the subject 200, calculated from data including the position, acceleration, angular velocity, inclination, foot pressure, etc. of the portable devices 210 indicated by latitude and longitude detected by each of the portable devices 210, the arm sensor 220, and the foot sensor 230, and an evaluation of the reliability of the data in a predetermined walking test. The predetermined walking test is, for example, a test that measures the walking speed, number of steps, etc. of the subject 200 within a predetermined distance, such as a 10-meter walking test.

[0030] The CPU 42, in which the trained machine learning model has been constructed, is equipped with a data selection function that selects data related to the walking of the subject 200 from data transmitted from the portable devices 210 carried by the subject 200, a walking data analysis function that extracts and analyzes from the selected data each of the data on acceleration, angular velocity, and foot pressure detected by the foot sensor 230, each of the data on acceleration, angular velocity, ambient sound, and altitude detected by the portable devices 210, each of the data on acceleration and angular velocity detected by the arm sensor 220, and the position information of the portable devices 210 detected by the GNSS function of the portable devices 210, a movement estimation function that estimates the walking speed, stride length, etc. of the subject 200 from each of the extracted data, a surrounding situation estimation function that estimates the situation around the subject 200, etc., and a walking data judgment function that judges the reliability of the walking data using the trained machine learning model. By executing the program having each of these functions, the CPU 42 functions as a data selection unit, a walking data analysis unit, a surrounding situation estimation unit, a movement estimation unit, and a determination unit.

[0031] 3 is a functional block diagram of the mobile devices 210 and the arithmetic device 10 of the information processing device 100 according to this embodiment. As shown in FIG. 3, the mobile devices 210 include a walking data collection unit 212, a movement estimation data collection unit 214, and a surrounding situation data collection unit 216.

[0032] 4(A) is a schematic diagram showing an example of walking data collected by the walking data collection unit 212 of the portable device 210. The walking data collection unit 212 collects data from the foot sensor 230 worn by the subject 200. The data detected by the foot sensor 230 includes acceleration indicated by acceleration (X, Y, Z), angular velocity indicated by gyro (X, Y, Z), and foot pressure.

[0033] As described above, the foot sensor 230 can transmit detected data to the portable device 210, and the walking data collection unit 212 of the portable device 210 transmits the data received from the foot sensor 230 to the communication device 110.

[0034] 4(B) is a schematic diagram showing an example of walking data collected by the movement estimation data collection unit 214 of the portable device 210. The movement estimation data collection unit 214 collects data detected by the portable device 210 and data from the arm sensor 220 worn by the subject 200. The data detected by the portable device 210 (smartphone) is acceleration indicated by acceleration (X, Y, Z) and angular velocity indicated by gyro (X, Y, Z). The data detected by the arm sensor 220 is acceleration indicated by acceleration (X, Y, Z) and angular velocity indicated by gyro (X, Y, Z). The arm sensor 220 can transmit the detected data to the portable device 210, and the movement estimation data collection unit 214 of the portable device 210 transmits the data received from the arm sensor 220 to the communication device 110.

[0035] FIG. 4(C) is a schematic diagram showing an example of surrounding situation data collected by the surrounding situation data collection unit 216 of the portable device 210. The surrounding situation data collection unit 216 collects information on latitude and longitude detected by the GNSS function of the portable device 210, surrounding audio detected by the portable device 210, changes in altitude detected by an altimeter or the like of the portable device 210, and changes in angular velocity detected by an IMU of the portable device 210, including the latitude and longitude of the current location, the degree of road inclination or step, and the volume of surrounding audio. The road inclination is estimated, for example, based on time-series changes in the altitude detected by an altimeter or the like of the portable device 210 and the angular velocity detected by the IMU of the portable device 210. If the altitude and angular velocity change linearly, it can be estimated that the subject 200 is walking on a sloped road surface rather than a step. If the altitude and angular velocity change rapidly and nonlinearly, it can be estimated that the subject 200 has crossed a step. Furthermore, the slope or degree of unevenness of the road surface can be estimated from the quantitative changes in the altitude and angular velocity over time.

[0036] As will be described later, the surrounding situation data includes information such as whether there is a crowd, whether there is construction work or an accident, which is part of the estimation result by the operation estimation unit 24 of the arithmetic device 10.

[0037] 4(A) is transmitted from the walking data collection unit 212 of the portable device 210 to the calculation device 10 and stored in the walking data storage unit 12 of the calculation device 10. The walking data stored in the walking data storage unit 12 is used by the walking data analysis unit 14 to calculate the walking status of the subject 200, such as the walking speed. The walking status data calculated by the walking data analysis unit 14 is stored in the gait analysis data storage unit 16.

[0038] FIG. 5(A) is a schematic diagram showing an example of gait analysis data indicating walking status. The gait analysis data is a time-series change in values ​​indicating the walking speed, stride length, and balance of the subject 200 while walking. The walking speed is calculated, for example, by integrating the acceleration detected by the foot sensor 230 over time. The stride length is, for example, the distance obtained by integrating the walking speed over time between the timings at which the foot sensor 230 detects the foot pressure of the subject 200. If the foot sensor 230 is attached to either the left or right foot of the subject 200, the stride length is half the value of the above distance. The balance is estimated based on the magnitude of the foot pressure detected by the foot sensor 230. For example, the value indicating balance is estimated to be small if the foot pressure is smaller than a predetermined reference value for the subject 200, and is estimated to be large if the foot pressure is larger than the reference value.

[0039] The movement estimation data shown in FIG. 4B is transmitted from the movement estimation data collection unit 214 of the portable device 210 to the arithmetic device 10 and stored in the movement estimation data storage unit 18 of the arithmetic device 10.

[0040] The surrounding situation data shown in Fig. 4(C) is transmitted from the surrounding situation data collection unit 216 of the portable device 210 to the computing device 10 and stored in the surrounding situation data storage unit 20 of the computing device 10. The surrounding situation data stored in the surrounding situation data storage unit 20 is used by the surrounding situation estimation unit 22 to estimate the impact on walking of road surface conditions, road congestion, etc.

[0041] The walking condition data stored in the gait analysis data storage unit 16, the movement estimation data stored in the movement estimation data storage unit 18, and the information indicating the effect on walking estimated by the surrounding situation estimation unit 22 are each input to the movement estimation unit 24.

[0042] The movement estimation unit 24 generates movement estimation data that estimates the user's movement based on the data accumulated in the movement estimation data accumulation unit 18, the data accumulated in the gait analysis data accumulation unit 16, and the effects on walking estimated by the surrounding situation estimation unit 22.

[0043] Fig. 5(B) is a schematic diagram showing an example of motion estimation data. The motion estimation data is a time-series change in latitude and longitude, and a time-series change in estimated motion of the subject (user) 200. Fig. 5(B) shows the estimated motions of the subject 200 as "crowd (following)," "crowd (overtaking)," and "using smartphone while using a device."

[0044] For example, when the ambient sound detected by the portable device 210 is at a volume that indicates a crowd and the walking speed of the subject 200 is slower than a reference walking speed predetermined for the subject 200, the movement estimation unit 24 determines that the subject 200 is walking by following a crowd, like a crowd (following).

[0045] For example, when the ambient sound detected by the portable device 210 is at a volume that indicates a crowd and the walking speed of the subject 200 is faster than a reference walking speed predetermined for the subject 200, the movement estimation unit 24 determines that the subject 200 is walking, overtaking the crowd, as if passing a crowd (overtaking).

[0046] When the latitude and longitude are changing in a time series, the movement estimation unit 24 determines that the subject person 200 is operating the mobile device 210 as a case of using a smartphone while doing other things.

[0047] Information such as whether there are crowds, whether there are construction works, and whether there are any accidents included in the estimation results from the movement estimation unit 24 is input to the surrounding situation data accumulation unit 20, and the input information, together with information obtained from the surrounding situation data collection unit 216, is used to calculate the impact on walking in the surrounding situation estimation unit 22.

[0048] The movement estimation data calculated by the movement estimation unit 24 is stored in a movement estimation data storage unit 26 .

[0049] The walking data determination unit 28 determines the validity of the walking data based on the data stored in the movement estimation data storage unit 26 and the data stored in the walking analysis data storage unit 16, and ranks the data.

[0050] 5(C) is a schematic diagram showing an example of the judgment rank. In this embodiment, the validity of walking data is judged for each time step from one timestamp to the next timestamp, and the data is judged as A, B, C, etc. Then, the judgment rank accumulation unit 30 accumulates the ranked data.

[0051] The processing by the gait data analysis unit 14, the surrounding situation estimation unit 22, and the movement estimation unit 24 described above is executed by the CPU 42 of the calculation device 10. Furthermore, each of the gait data accumulation unit 12, the gait analysis data accumulation unit 16, the movement estimation data accumulation unit 18, the surrounding situation data accumulation unit 20, the movement estimation data accumulation unit 26, and the judgment rank accumulation unit 30 corresponds to a storage device such as an HDD 56, for example.

[0052] 6 is a flowchart showing an example of walking data determination processing by the information processing device 100 according to this embodiment. In step S100, various data are collected by the portable device 210. Then, in step S102, the collected data is transmitted to the arithmetic unit via the communication device 110.

[0053] In step S104, the walking data analysis unit 14 calculates walking conditions such as walking speed.

[0054] In step S106, the surrounding situation data storage unit 20 collects information from each user, and updates the data using the movement estimation result from the movement estimation unit 24.

[0055] In step S108, the surrounding conditions estimation unit 22 estimates the influence on walking from the road surface conditions, road congestion, and the like.

[0056] In step S110, the data detected by the portable devices 210 and the arm sensor 220 are stored in the motion estimation data storage unit 18.

[0057] In step S112, the movement estimation unit 24 estimates the user's movement based on the data from the movement estimation data accumulation unit 18, the calculated walking state, and the effect on walking.

[0058] In step S114, the walking data determination unit 28 determines the validity of the data from the results of the movement estimation and ranks them.

[0059] In step S116, the ranked data is stored in the judgment rank storage unit 30, and the process ends.

[0060] As described above, this embodiment can estimate the reliability of predetermined walking data used in a predetermined walking test, such as a 10-m walking test. The predetermined walking test requires providing walking data that indicates the subject's 200 normal walking state. However, because the subject's 200 daily behavior may be affected by the surrounding environment, there is a risk that inappropriate walking data may be used in the predetermined walking test. For example, the subject may walk quickly to avoid crowds, or may walk slower than the subject's natural walking speed to follow the flow of the crowd. In this embodiment, the reliability of the walking data is estimated and only reliable walking data is used in the predetermined walking test, thereby improving the analysis accuracy of the predetermined walking test.

[0061] Furthermore, in this embodiment, walking data can be acquired using the arm sensor 220, foot sensor 230, and portable devices 210 that can be worn by the subject 200 on a daily basis without feeling uncomfortable. [Explanation of symbols]

[0062] 10 Arithmetic unit 14. Gait Data Analysis Department 22 Surrounding situation estimation unit 24 Motion estimation section 28 Walking data determination unit 40 Computer 42 CPU 44 ROM 46 RAM 100 Information processing device 200 Target Audience 210 Mobile Devices 220 Arm Sensor 230 Foot Sensor

Claims

1. an acquisition unit that acquires, via a portable device carried by the subject, gait data indicating the gait of the subject detected by a sensor worn by the subject; a calculation unit that estimates the reliability of the walking data acquired by the acquisition unit in a predetermined walking test using walking instructor data indicating the gait of a pedestrian and a machine learning model that has been trained using the reliability of the walking instructor data in the predetermined walking test; Including, The mobile device is capable of detecting acceleration and angular velocity and estimating its current location by satellite positioning; the sensors include an arm sensor that detects the acceleration and angular velocity of the subject's arm, and a foot sensor that detects the acceleration and angular velocity of the subject's foot and the pressure of the sole of the foot; The calculation unit is an information processing device that applies gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current position determined by the satellite positioning to a trained machine learning model, and estimates the reliability of the walking data in the specified walking test.

2. A first step of training a machine learning model using walking instructor data indicating a pedestrian's gait and the reliability of the walking instructor data in a predetermined walking test; a second step of acquiring, via a portable device carried by the subject, gait data indicating the gait of the subject detected by a sensor worn by the subject; a third step of estimating the reliability of the acquired walking data in the predetermined walking test using a trained machine learning model; An information processing method comprising: The mobile device is capable of detecting acceleration and angular velocity and estimating its current location by satellite positioning; the sensors include an arm sensor that detects the acceleration and angular velocity of the subject's arm, and a foot sensor that detects the acceleration and angular velocity of the subject's foot and the pressure of the sole of the foot; The third step is an information processing method for estimating the reliability of the walking data in the specified walking test by applying gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current position determined by the satellite positioning, to a trained machine learning model.

3. A computer, An information processing program that functions as a learning unit that trains a machine learning model using walking instructor data that indicates a pedestrian's gait and the reliability of the walking instructor data in a predetermined walking test, an acquisition unit that acquires walking data that indicates the gait of the subject detected by a sensor attached to the subject via a portable device carried by the subject, and a calculation unit that estimates the reliability of the walking data acquired by the acquisition unit in the predetermined walking test using the trained machine learning model, The mobile device is capable of detecting acceleration and angular velocity and estimating its current location by satellite positioning; the sensors include an arm sensor that detects the acceleration and angular velocity of the subject's arm, and a foot sensor that detects the acceleration and angular velocity of the subject's foot and the pressure of the sole of the foot; The calculation unit applies gait analysis data including the subject's walking speed, stride length, and balance calculated based on the walking data including the acceleration and angular velocity detected by the sensor, the pressure on the soles of the feet, and the current position determined by the satellite positioning to a trained machine learning model, and estimates the reliability of the walking data in the specified walking test.

Images