Data generation device, gait measurement system, data generation method, and recording medium

Abstract

Provided is a data generation device comprising: an acquisition unit that acquires sensor data measured by a measurement device mounted on footwear of a user and position data corresponding to the position at which the sensor data is acquired, in order to generate gait data corrected according to the path surface condition of a walking path; a path surface determination unit that determines the path surface condition of the walking path at a position corresponding to the position data; a generation unit that generates gait data using the sensor data; a correction unit that corrects the gait data according to the path surface condition of the walking path; and an output unit that outputs the corrected gait data.

Description

Data generation device, gait measurement system, data generation method, and recording medium

[0001] The present disclosure relates to a data generating device, a gait measurement system, a data generating method, and a recording medium.

[0002] With growing interest in healthcare, attention is being focused on services that provide information tailored to the health status of individuals. For example, in order to prevent individuals from becoming frail, it is necessary to provide appropriate training advice. In everyday life, individuals walk on walkways with diverse walking environments. Walkway surface conditions can vary widely, including asphalt, grass, and sand. Different walkway surface conditions affect the gait data measured as the individual walks. Understanding the walking environment, including the surface conditions of the walkway, will enable the provision of more appropriate training advice to elderly individuals.

[0003] Patent Literature 1 discloses a technology for recognizing road surface conditions based on a user's walking state. The technology disclosed in Patent Literature 1 recognizes road surface conditions using output values ​​of a sensor attached to a portable device carried by the user.

[0004] Japanese Patent Application Laid-Open No. 2020-140743

[0005] According to the method of Patent Document 1, the presence or absence of steps on a walkway can be determined using the output value of a sensor attached to a portable device. In the method of Patent Document 1, factors other than walking are added to the output value of the sensor attached to the portable device. Therefore, it was difficult to accurately determine the road surface condition of the walkway using the method of Patent Document 1. Furthermore, even if the method of Patent Document 1 could determine the presence or absence of steps on the walkway, it could not determine the road surface condition of the walkway. In order to provide an appropriate training menu that is suited to the walking environment, gait data corrected according to the road surface condition of the walkway is required.

[0006] An object of the present disclosure is to provide a data generation device, a gait measurement system, a data generation method, and a recording medium that are capable of generating gait data corrected in accordance with the surface conditions of a walkway.

[0007] A data generation device according to one aspect of the present disclosure includes an acquisition unit that acquires sensor data measured by a measurement device mounted on a user's footwear and position data corresponding to the position at which the sensor data was acquired, a road surface determination unit that determines the road surface condition of the walkway at the position corresponding to the position data, a generation unit that generates gait data using the sensor data, a correction unit that corrects the gait data to match the road surface condition of the walkway, and an output unit that outputs the corrected gait data.

[0008] In one aspect of the data generation method of the present disclosure, sensor data measured by a measuring device mounted on a user's footwear and position data corresponding to the position where the sensor data was acquired are acquired, the road surface condition of the walkway at the position corresponding to the position data is determined, gait data is generated using the sensor data, the gait data is corrected to match the road surface condition of the walkway, and the corrected gait data is output.

[0009] A recording medium according to one embodiment of the present disclosure has recorded thereon a program that causes a computer to execute the following processes: acquiring sensor data measured by a measuring device mounted on a user's footwear and position data corresponding to the position where the sensor data was acquired; determining the surface condition of the walkway at the position corresponding to the position data; generating gait data using the sensor data; correcting the gait data to match the surface condition of the walkway; and outputting the corrected gait data.

[0010] According to the present disclosure, it is possible to provide a data generation device, a gait measurement system, a data generation method, and a recording medium that are capable of generating gait data corrected in accordance with the surface conditions of a walkway.

[0011] 1 is a block diagram showing an example of the configuration of a gait measurement system according to the present disclosure. FIG. 1 is a block diagram showing an example of the configuration of a measurement device according to the present disclosure. FIG. 2 is a conceptual diagram showing an example of the arrangement of a measurement device according to the present disclosure. FIG. 3 is a conceptual diagram for explaining a coordinate system set in the measurement device according to the present disclosure. FIG. 4 is a conceptual diagram for explaining an example of a human body surface. FIG. 5 is a conceptual diagram for explaining an example of a gait cycle. FIG. 6 is a block diagram showing an example of the configuration of a data generation device according to the present disclosure. FIG. 7 is a conceptual diagram for explaining an example of the flow of sensor data and position data according to the present disclosure. FIG. 8 is a graph showing an example in which features according to the road surface category of a walkway appear in a gait waveform. FIG. 9 is a graph showing an example in which features according to the road surface category of a walkway appear in a gait waveform. FIG. 10 is a conceptual diagram for explaining an example of training of a road surface discrimination model according to the present disclosure. FIG. 11 is a conceptual diagram for explaining an example of training of a road surface discrimination model according to the present disclosure. FIG. 12 is a conceptual diagram for explaining an example of estimating a road surface category using a road surface discrimination model according to the present disclosure. FIG. 13 is a conceptual diagram for explaining an example of training of a gait data correction model according to the present disclosure. FIG. 14 is a conceptual diagram for explaining an example of correction of gait data by the gait data correction model according to the present disclosure. FIG. 1 is a conceptual diagram for describing an example of learning of a gait data correction model according to the present disclosure. FIG. 2 is a conceptual diagram for describing an example of correction of gait data using a gait data correction model according to the present disclosure. FIG. 3 is a conceptual diagram showing an example of data flow between components constituting a data generating device according to the present disclosure. FIG. 4 is a flowchart for describing an example of operation of a data generating device according to the present disclosure. FIG. 5 is a conceptual diagram showing an example of display of information generated using gait data corrected by a data generating device according to the present disclosure. FIG. 6 is a conceptual diagram showing an example of display of information generated using gait data corrected by a data generating device according to the present disclosure. FIG. 7 is a block diagram showing an example of the configuration of a gait measurement system according to the present disclosure. FIG. 8 is a block diagram showing an example of the configuration of a data generating device according to the present disclosure. FIG. 9 is a conceptual diagram for describing the accuracy range of position data.1 is a conceptual diagram for explaining an example in which a data generating device according to the present disclosure calculates corrected gait data using gait data corrected for each road surface category. FIG. 1 is a conceptual diagram showing an example of the flow of data between components constituting a data generating device according to the present disclosure. FIG. 2 is a flowchart for explaining an example of the operation of a data generating device according to the present disclosure. FIG. 3 is a conceptual diagram showing an example of a display of information generated using gait data corrected by a data generating device according to the present disclosure. FIG. 4 is a conceptual diagram showing an example of a map displayed on the screen of a mobile device used by a user. FIG. 5 is a conceptual diagram showing an example of a planned walking route input on the screen of a mobile device used by a user. FIG. 6 is a conceptual diagram showing an example of a recommended route according to a road surface category displayed on the screen of a mobile device used by a user. FIG. 7 is a conceptual diagram showing an example of a recommended route including a recommended route according to a road surface category displayed on the screen of a mobile device used by a user. FIG. 8 is a conceptual diagram showing an example of a recommended route including the influence of weather displayed on the screen of a mobile device used by a user. FIG. 9 is a conceptual diagram showing an example of a recommended route including the influence of weather displayed on the screen of a mobile device used by a user. FIG. 2 is a block diagram illustrating an example of a configuration of hardware that executes control and processing according to the present disclosure.

[0012] Hereinafter, embodiments for carrying out the present disclosure will be described with reference to the drawings. In this disclosure, the drawings used in describing each embodiment relate to one or more embodiments. Furthermore, elements included in each drawing may apply to one or more embodiments. The embodiments described below are limited in a manner that is technically preferable for carrying out the present disclosure, but the scope of the disclosure is not limited to the following. In all drawings used in describing the following embodiments, similar parts are designated by the same reference numerals unless otherwise specified. In the following embodiments, repeated description of similar configurations and operations may be omitted.

[0013] First Embodiment First, a gait measurement system according to this embodiment will be described with reference to the drawings. The gait measurement system of this embodiment acquires sensor data related to foot movements measured in accordance with the walking of a user who is the subject of gait data measurement. The gait measurement system of this embodiment uses the acquired sensor data to estimate the road surface conditions of the walkway along which the subject is walking. The gait measurement system of this embodiment corrects gait data generated using the sensor data measured in accordance with the walking of the subject, in accordance with the estimated road surface conditions.

[0014] (Configuration) FIG. 1 is a block diagram showing an example of the configuration of a gait measurement system according to the present disclosure. The gait measurement system 1 includes a measurement device 10 and a data generating device 12. For example, the measurement device 10 is installed in the footwear of a subject whose gait is to be measured. For example, the data generating device 12 is installed on a server or a cloud. The server or cloud is connected to a mobile device carried by the subject via a network such as the Internet. For example, the functions of the data generating device 12 may be implemented in the mobile device carried by the subject. Below, the configurations of the measurement device 10 and the data generating device 12 will be described individually.

[0015] [Measurement Device] Fig. 2 is a block diagram showing an example of the configuration of the measurement device 10. The measurement device 10 has a sensor 110, a control unit 113, a communication unit 115, and a power supply 117. The sensor 110 has an acceleration sensor 111 and an angular velocity sensor 112. The sensor 110 may include sensors other than the acceleration sensor 111 and the angular velocity sensor 112. Description of sensors other than the acceleration sensor 111 and the angular velocity sensor 112 that may be included in the sensor 110 will be omitted.

[0016] The acceleration sensor 111 is a sensor that measures acceleration in three axial directions. The acceleration in three axial directions is also called spatial acceleration. The acceleration sensor 111 measures acceleration as a physical quantity related to foot movement. The acceleration sensor 111 outputs the measured acceleration to the control unit 113. There are no limitations on the sensor used as the acceleration sensor 111 as long as it can measure acceleration. For example, the acceleration sensor 111 may be a piezoelectric, piezo-resistive, or capacitive sensor.

[0017] The angular velocity sensor 112 is a sensor that measures angular velocity around three axes. The angular velocity around three axes is also called spatial angular velocity. The angular velocity sensor 112 measures angular velocity as a physical quantity related to foot movement. The angular velocity sensor 112 outputs the measured angular velocity to the control unit 113. There are no limitations on the sensor used as the angular velocity sensor 112 as long as it can measure angular velocity. For example, a vibration type, a capacitance type, or other type of sensor can be used as the angular velocity sensor 112.

[0018] The sensor 110 is realized, for example, by an inertial measurement unit (IMU) that measures acceleration and angular velocity. An example of an IMU is an inertial measurement unit (IMU). The IMU includes an acceleration sensor that measures acceleration in three axes and an angular velocity sensor that measures angular velocity around three axes. The sensor 110 may be realized by an inertial measurement unit such as a vertical gyro (VG) or an attitude heading reference system (AHRS). The sensor 110 may also be realized by a global positioning system (GPS) / inertial navigation system (INS). The sensor 110 may be realized by a device other than an inertial measurement unit as long as it can measure physical quantities related to foot movement. For example, the sensor 110 may include a pressure sensor that measures pressure applied by the sole of the foot.

[0019] FIG. 3 is a conceptual diagram showing an example in which the measurement device 10 is placed in the shoes 100 of both feet. In the example of FIG. 3 , the measurement device 10 is placed at a position corresponding to the back of the arch of the foot. For example, the measurement device 10 is placed in an insole inserted into the shoe 100. For example, the measurement device 10 may be placed on the bottom of the shoe 100. For example, the measurement device 10 may be embedded in the body of the shoe 100. The measurement device 10 may be detachable from the shoe 100 or may not be detachable from the shoe 100. The measurement device 10 may be placed at a position other than the back of the arch of the foot as long as it can measure sensor data related to foot movement. The measurement device 10 may also be attached to socks worn by the subject or to an accessory such as an anklet worn by the subject. The measurement device 10 may also be attached directly to the foot or embedded in the foot. The measurement device 10 may also be placed in one of the shoes 100 as long as it can measure sensor data from which gait data can be generated.

[0020] In the example of FIG. 3 , a local coordinate system is set with the measurement device 10 (sensor 110) as the reference, and includes an x-axis in the left-right direction, a y-axis in the front-back direction, and a z-axis in the up-down direction. In the present disclosure, the x-axis is positive to the left, the y-axis is positive to the rear, and the z-axis is positive to the up. FIG. 3 shows an example in which the same coordinate system is set for the left foot and the right foot. For example, if sensors 110 manufactured to the same specifications are placed in left and right shoes 100, the up-down orientations (Z-axis directions) of the sensors 110 placed in the left and right shoes 100 are the same. In this case, the three axes of the local coordinate system set for the sensor data originating from the left foot and the three axes of the local coordinate system set for the sensor data originating from the right foot are the same for the left and right.

[0021] FIG. 4 is a conceptual diagram illustrating the relationship between a local coordinate system (x-axis, y-axis, z-axis) set in the measurement device 10 (sensor 110) installed on the back of the arch of the foot and a world coordinate system (x-axis, y-axis, z-axis) set relative to the ground. FIG. 4 shows the coordinate system set for the right foot. In the world coordinate system (x-axis, y-axis, z-axis), when a subject is standing upright facing the direction of travel, the x-axis corresponds to the subject's lateral direction, the y-axis corresponds to the subject's front-to-back direction, and the z-axis corresponds to the direction of gravity. Note that the example in FIG. 4 conceptually illustrates the relationship between the local coordinate system (x-axis, y-axis, z-axis) and the world coordinate system (x-axis, y-axis, z-axis). The example in FIG. 4 does not accurately illustrate the relationship between the local coordinate system and the world coordinate system, which changes depending on the subject's walking.

[0022] FIG. 5 is a conceptual diagram illustrating planes (also called human body planes) set for the human body. The sagittal plane is a plane that divides the body into left and right halves. The coronal plane is a plane that divides the body into front and back halves. The horizontal plane is a plane that divides the body horizontally. As shown in FIG. 5 , when the user is standing upright with the center lines of the feet pointing in the direction of travel, the world coordinate system and the local coordinate system are assumed to coincide. FIG. 5 shows an example in which the same coordinate system is set for the left and right feet. In this embodiment, rotation in the sagittal plane around the X-axis (x-axis) as the rotation axis is defined as roll, rotation in the coronal plane around the Y-axis (y-axis) as the rotation axis is defined as pitch, and rotation in the horizontal plane around the Z-axis (z-axis) as the rotation axis is defined as yaw. Furthermore, the rotation angle in the sagittal plane around the X-axis (x-axis) as the rotation axis is defined as roll angle, the rotation angle in the coronal plane around the Y-axis (y-axis) as the rotation axis is defined as pitch angle, and the rotation angle in the horizontal plane around the Z-axis (z-axis) as the rotation axis is defined as yaw angle.

[0023] The control unit 113 causes the acceleration sensor 111 and the angular velocity sensor 112 to measure sensor data. For example, the control unit 113 causes the acceleration sensor 111 and the angular velocity sensor 112 to start measurement in response to detection of the subject walking. For example, the control unit 113 starts measuring sensor data starting from the point in time when it is detected that either the left or right foot has started to move in the direction of travel after both feet have been at the same vertical height for a predetermined period of time. For example, the control unit 113 may cause the acceleration sensor 111 and the angular velocity sensor 112 to start measurement at a predetermined timing. For example, the control unit 113 may be configured to cause the acceleration sensor 111 and the angular velocity sensor 112 to start measurement in response to a measurement start signal transmitted from the data generating device 12.

[0024] The control unit 113 acquires acceleration in three axial directions from the acceleration sensor 111. The control unit 113 also acquires angular velocities around three axes from the angular velocity sensor 112. For example, the control unit 113 performs analog-to-digital (AD) conversion on acquired physical quantities (analog data) such as angular velocity and acceleration. The physical quantities (analog data) measured by the acceleration sensor 111 and the angular velocity sensor 112 may be converted to digital data in each of the acceleration sensor 111 and the angular velocity sensor 112. For example, an AD conversion circuit that AD converts the physical quantities (analog data) such as angular velocity and acceleration may be provided. The control unit 113 outputs the converted digital data (also referred to as sensor data) to the communication unit 115. For example, the control unit 113 may temporarily store the sensor data in a storage unit (not shown). The sensor data includes at least acceleration data converted to digital data and angular velocity data converted to digital data. The acceleration data includes acceleration vectors in the three axial directions. The angular velocity data includes angular velocity vectors around three axes. The acceleration data and angular velocity data are associated with the time at which they were acquired. The control unit 113 may also apply corrections to the acceleration data and angular velocity data, such as correction for mounting errors, temperature correction, and linearity correction.

[0025] For example, the control unit 113 is realized by a microcomputer or microcontroller that performs overall control of the measurement device 10 and performs data processing. For example, the control unit 113 includes a central processing unit (CPU), random access memory (RAM), read-only memory (ROM), flash memory, etc. For example, the control unit 113 may calculate gait data (described later) using sensor data. In this case, the measurement device 10 outputs the calculated gait data to the data generating device 12. For example, the control unit 113 may calculate gait indices (described later) using sensor data. In this case, the measurement device 10 outputs the calculated gait indices to the data generating device 12. For example, the control unit 113 may calculate feature amounts used to estimate physical abilities (described later). In this case, the measurement device 10 outputs the calculated feature amounts to the data generating device 12.

[0026] The communication unit 115 acquires sensor data from the control unit 113. The communication unit 115 transmits the acquired sensor data to the data generating device 12. The timing of transmitting the sensor data is not particularly limited. For example, the communication unit 115 transmits the sensor data in real time in response to measurement of the sensor data. For example, the communication unit 115 transmits the sensor data at a predetermined transmission timing. For example, the communication unit 115 may store sensor data measured over a predetermined period and transmit the stored sensor data all at once at a predetermined timing. For example, the communication unit 115 may be configured to receive a measurement start signal from the data generating device 12. In this case, the communication unit 115 outputs the received measurement start signal to the control unit 113.

[0027] For example, the communication unit 115 transmits the sensor data to the data generating device 12 via wireless communication. For example, the communication unit 115 transmits the sensor data to the data generating device 12 via a wireless communication function (not shown) conforming to standards such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). The communication function of the communication unit 115 may conform to standards other than Bluetooth (registered trademark) or Wi-Fi (registered trademark). The communication unit 115 may transmit the sensor data to the data generating device 12 via a wired connection such as a cable.

[0028] The power supply 117 is a battery that supplies power for operating the measuring device 10. For example, the power supply 117 may be a thin battery, such as a coin or button type. For example, the power supply 117 may be a primary battery, such as a lithium primary battery, a silver oxide battery, an alkaline button battery, or a zinc-air battery. When the power supply 117 is a primary battery, it is preferable that the power supply 117 be a long-life battery. The power supply 117 may also be a rechargeable secondary battery. When the power supply 117 is a secondary battery, the power supply 117 may be a battery that can be charged via a wired connection or a battery that can be powered wirelessly. If the power supply 117 is capable of wireless power supply, a wireless power supply device may be placed in a location where footwear is kept, such as an entrance or a shoe locker. In this case, by placing footwear equipped with the measuring device 10 on the wireless power supply device, the measuring device 10 can be appropriately charged when not in use.

[0029] Here, a gait cycle will be described with reference to the drawings. FIG. 6 is a conceptual diagram for explaining a step cycle based on the right foot. A step cycle based on the left foot is similar to that of the right foot. The horizontal axis of FIG. 6 indicates one step cycle of the right foot, starting from the point when the heel of the right foot hits the ground and ending from the point when the heel of the right foot hits the ground again. The horizontal axis of FIG. 6 is normalized with the step cycle set to 100%. Normalizing one gait cycle to 100% is called first normalization. One gait cycle of one leg is broadly divided into a stance phase, in which at least a portion of the sole of the foot is in contact with the ground, and a swing phase, in which the sole of the foot is off the ground. The stance phase is the period during which at least a portion of the sole of the foot is in contact with the ground. The stance phase is further divided into an early stance phase T1, a mid-stance phase T2, a final stance phase T3, and an early swing phase T4. The swing phase is the period during which the sole of the foot is off the ground. The swing phase is further divided into an early swing phase T5, a mid-swing phase T6, and a final swing phase T7. The horizontal axis in Figure 6 is normalized so that the stance phase is 60% and the swing phase is 40%. Normalizing gait waveform data so that the stance phase is 60% and the swing phase is 40% is called second normalization. Note that the periods shown in Figure 6 are merely examples and do not limit the periods that make up a gait cycle or the names of those periods.

[0030] As shown in FIG. 6 , multiple events occur during walking. Multiple events that occur during walking are also called walking events. P1 represents heel strike (HS). Heel strike is an event in which the heel of the right foot touches the ground. P2 represents opposite toe off (OTO). Opposite toe off is an event in which the toe of the left foot leaves the ground while the sole of the right foot is in contact with the ground. P3 represents heel rise (HR). Heel rise is an event in which the heel of the right foot lifts while the sole of the right foot is in contact with the ground. P4 represents opposite heel strike (OHS). Opposite heel strike is an event in which the heel of the left foot touches the ground. P5 represents toe off (TO). Toe-off is an event in which the toe of the right foot leaves the ground while the sole of the left foot is in contact with the ground. P6 represents foot adjacent (FA). Foot crossing is an event in which the left and right feet cross while the sole of the left foot is in contact with the ground. P7 represents tibia vertical (TV). Tibia vertical is an event in which the tibia of the right foot becomes approximately perpendicular to the ground while the sole of the left foot is in contact with the ground. P8 represents heel strike. P8 corresponds to the end point of the gait cycle that begins with P1 and the start point of the next gait cycle. Note that the gait events shown in FIG. 6 are merely examples and do not limit the events that occur during walking or the names of these events.

[0031] The timing of heel strike is the timing of the minimum peak immediately after the maximum peak that appears in the time series data of forward acceleration (Y-direction acceleration). The maximum peak that marks the heel strike timing corresponds to the maximum peak of the walking waveform data for one step cycle. The section between consecutive heel strikes corresponds to one step cycle. The timing of toe lift is the timing of the rise of the maximum peak that appears after the stance phase period in which no fluctuations appear in the time series data of forward acceleration (Y-direction acceleration). The timing midway between the timing of the minimum roll angle and the timing of the maximum roll angle corresponds to the mid-stance phase.

[0032] [Data Generating Device] Fig. 7 is a block diagram showing an example of the configuration of the data generating device 12. The data generating device 12 has an acquiring unit 121, a road surface determining unit 122, a generating unit 123, a correcting unit 125, and an output unit 127. For example, the data generating device 12 is constructed on a server or a cloud. For example, the functions of the data generating device 12 may be implemented in a mobile terminal carried by the subject. For example, the functions of the data generating device 12 may be implemented in a terminal device used by an administrator who manages gait data measured for the subject.

[0033] Data generating device 12 also has a road surface discrimination model 132 and a gait data correction model 135. For example, road surface discrimination model 132 and gait data correction model 135 are stored in storage unit 13. Road surface discrimination model 132 and gait data correction model 135 may be stored in a storage device (not shown) accessible from data generating device 12. In this case, data generating device 12 uses road surface discrimination model 132 and gait data correction model 135 via an interface (not shown) such as an API (Application Programming Interface) connected over a network.

[0034] The acquisition unit 121 acquires sensor data from the measurement device 10 mounted on the footwear of the subject. The acquisition unit 121 receives the sensor data from the measurement device 10 via wireless communication. The sensor data includes location information of the mobile device that is the sender of the sensor data. For example, the location information is measured using a global positioning system (GPS) function mounted on the mobile device and added to the sensor data.

[0035] For example, the acquisition unit 121 receives sensor data from the measurement device 10 via a wireless communication function (not shown) conforming to standards such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). As long as the acquisition unit 121 can communicate with the measurement device 10, the communication function of the acquisition unit 121 may conform to standards other than Bluetooth (registered trademark) or Wi-Fi (registered trademark). The acquisition unit 121 may receive sensor data from the measurement device 10 via a wired connection such as a cable. For example, the acquisition unit 121 may acquire gait data, gait indices, and feature amounts calculated by the measurement device 10.

[0036] FIG. 8 is a conceptual diagram for explaining an example of the flow of sensor data and position data in the present disclosure. A measurement device 10 is placed in a shoe 100 worn by a subject. In the example of FIG. 8, the measurement device 10 is placed at a position that touches the sole of the subject's arch. Sensor data measured by the measurement device 10 according to the subject's walking is transmitted to a mobile terminal 170 carried by the subject. The mobile terminal 170 receives the sensor data transmitted from the measurement device 10. The mobile terminal 170 adds position data to the received sensor data. The position data includes longitude information and latitude information corresponding to the position where the sensor data was measured. The mobile terminal 170 transmits the sensor data with the added position data to the data generating device 12.

[0037] The road surface determination unit 122 obtains the position data from the obtaining unit 121. The road surface determination unit 122 determines the road surface condition of the walkway on which the subject is walking using the obtained position data. The road surface determination unit 122 uses the road surface determination model 132 to determine the road surface condition of the walkway on which the subject is walking.

[0038] The road surface discrimination model 132 is a machine learning model. For example, the road surface discrimination model 132 is a model trained using a data set in which position data is an explanatory variable and a road surface category is a target variable as training data. For example, the road surface discrimination model 132 is a learning model trained using a convolutional neural network (CNN) technique. For example, the road surface discrimination model 132 is a model trained using a principal component analysis (PCA) technique. For example, the road surface discrimination model 132 is a learning model trained using a variational autoencoder (VAE). For example, the road surface discrimination model 132 is a learning model trained using a conditional generative adversarial network (GAN) technique. The above techniques are merely examples and do not limit the techniques for training the road surface discrimination model 132.

[0039] The road surface discrimination model 132 outputs a road surface category corresponding to input position data. The road surface category includes a reference road surface. The reference road surface is a road surface that serves as a reference for the road surface condition of a walkway. For example, the reference road surface is a walkway with a vinyl-coated surface that is easy to walk on, such as an indoor corridor. For example, road surface categories are classified according to road surfaces that exhibit walking characteristics, such as asphalt, grass, and sand. For example, road surface categories may be classified according to the slope, unevenness, and steps of the road surface. Furthermore, road surface categories may be classified according to the slipperiness of the road surface, such as the degree of wetness or dryness of the road surface. Depending on the road surface category of the walkway, the characteristics of walking on that walkway are expressed in the gait waveform. The road surface discrimination model 132 may be stored in an external storage device constructed in a cloud, a server, or the like. In this case, the road surface discrimination unit 122 uses the road surface discrimination model 132 via an interface (not shown) connected to the storage device.

[0040] 9 and 10 are graphs showing an example of gait waveforms that reflect characteristics according to the surface category of a walkway. In FIGS. 9 and 10, the reference surface is a vinyl-coated floor. FIGS. 9 and 10 show gait waveforms that are graphs of the average of two 10-meter walking tests conducted by nine subjects. FIG. 9 shows a gait waveform (dashed line) measured while walking on a walkway with a reference surface and a gait waveform (solid line) measured while walking on a grass walkway, superimposed on each other. FIG. 10 shows a gait waveform (dashed line) measured while walking on a walkway with a reference surface and a gait waveform (solid line) measured while walking on a sand walkway, superimposed on each other. By comparing the gait waveforms with the reference surface, as shown within the dashed-dotted circles in FIGS. 9 and 10, it is possible to understand that characteristics according to the walkway are present. The surface condition of a walkway can be estimated using a machine learning model that has learned characteristics according to the walkway.

[0041] Fig. 11 is a conceptual diagram for explaining an example of learning a road surface discrimination model according to the present disclosure. In the example of Fig. 11, a data set having position data as an explanatory variable and a road surface category as a response variable is used as training data to train the road surface discrimination model 132. The position data includes latitude and longitude measured by a position measurement means such as a GPS. A description of a learning device (not shown) that trains the road surface discrimination model 132 will be omitted. The road surface condition of the walkway at the position represented by position data A is road surface category a. The road surface condition of the walkway at the position represented by position data B is road surface category b. The road surface condition of the walkway at the position represented by position data N is road surface category n.

[0042] FIG. 12 is a conceptual diagram illustrating an example of learning a road surface discrimination model according to the present disclosure. FIG. 12 shows a road surface discrimination device 140 that discriminates road surfaces using image data such as satellite photographs, aerial photographs, and land parcel maps. A position in the image data is linked to position data (latitude and longitude). The road surface discrimination device 140 discriminates the road surface category included in the image data using a road surface discrimination model 145. The road surface discrimination model 145 is a machine learning model that has been trained in advance by machine learning. The road surface discrimination model 145 discriminates the road surface category based on features included in the image data in response to input image data. The road surface discrimination model 145 discriminates the road surface category for each position in the image data.

[0043] 13 is a conceptual diagram showing an example of estimation of a road surface category using the road surface identification model according to the present disclosure. The road surface identification model 145 outputs a road surface category in response to input of position data.

[0044] The road surface identification device 140 outputs a data set in which position data and road surface categories are linked. For example, the road surface identification device 140 outputs a data set in which position data A and road surface category a are linked. For example, the road surface identification device 140 outputs a data set in which position data B and road surface category b are linked. For example, the road surface identification device 140 outputs a data set in which position data N and road surface category n are linked.

[0045] 12 , as in the example of FIG. 11 , a data set in which position data is an explanatory variable and road surface category is a response variable is used as training data to train the road surface discrimination model 132. The position data includes latitude and longitude measured by a position measurement means such as a GPS. A learning device (not shown) that trains the road surface discrimination model 132 will not be described here.

[0046] The road surface discrimination model 132 trained according to the procedures of Figures 11 and 12 outputs a road surface category corresponding to input position data in response to the input position data. For example, in response to input position data A, the road surface discrimination model 132 outputs a road surface category a corresponding to the position data. For example, in response to input position data B, the road surface discrimination model 132 outputs a road surface category b corresponding to the position data. For example, in response to input position data N, the road surface discrimination model 132 outputs a road surface category n corresponding to the position data.

[0047] For example, the road surface discrimination model 132 may be generated by learning using a linear regression algorithm. For example, the road surface discrimination model 132 may be generated by learning using a support vector machine (SVM) algorithm. For example, the road surface discrimination model 132 may be generated by learning using a Gaussian process regression (GPR) algorithm. For example, the road surface discrimination model 132 may be generated by learning using a random forest (RF) algorithm. The algorithm used to train the road surface discrimination model 132 is not limited to the examples given here.

[0048] The generation unit 123 acquires sensor data from the acquisition unit 121. The generation unit 123 extracts time series data for one walking cycle from the time series data of acceleration in three axial directions and angular velocity around three axes included in the sensor data. The time series data for one walking cycle is also called walking waveform data. The generation unit 123 extracts walking waveform data based on the timing of walking events detected from the time series data of the sensor data. For example, the generation unit 123 extracts walking waveform data that starts at the timing of a heel strike and ends at the timing of the next heel strike.

[0049] The generation unit 123 normalizes (first normalization) the time of the extracted walking waveform data for one step cycle to a walking cycle of 0 to 100% (percent). The timing of 1%, 10%, etc. included in the 0 to 100% walking cycle is also called a walking phase. Furthermore, the generation unit 123 normalizes (second normalization) the first-normalized walking waveform data for one step cycle so that the stance phase is 60% and the swing phase is 40%. By subjecting the walking waveform data to second normalization, it is possible to reduce the discrepancy in the walking phase from which feature values ​​are extracted.

[0050] For example, the generation unit 123 extracts gait waveform data for one step gait cycle using the traveling direction acceleration (Y-direction acceleration). The generation unit 123 extracts gait waveform data for one step gait cycle with respect to accelerations / angular velocities / angles other than the traveling direction acceleration (Y-direction acceleration) in accordance with the gait cycle of the traveling direction acceleration (Y-direction acceleration). The generation unit 123 extracts gait waveform data related to accelerations in three axial directions, gait waveform data related to angular velocities around three axes, and gait waveform data related to angles around three axes. The generation unit 123 normalizes the extracted gait waveform data for one step gait cycle. The generation unit 123 may generate time series data of angles around three axes by integrating time series data of angular velocities around three axes.

[0051] The generation unit 123 may extract gait waveform data for one step gait cycle using acceleration / angular velocity other than forward acceleration (Y-direction acceleration). For example, the generation unit 123 may detect heel strike and toe lift from time series data of vertical acceleration (Z-direction acceleration). The timing of heel strike is the timing of a steep minimum peak that appears in the time series data of vertical acceleration (Z-direction acceleration). At the timing of the steep minimum peak, the value of vertical acceleration (Z-direction acceleration) becomes approximately 0. The minimum peak that marks the timing of heel strike corresponds to the minimum peak of the gait waveform data for one step gait cycle. The section between consecutive heel strikes is a gait cycle. The timing of toe lift is the timing of an inflection point in the time series data of vertical acceleration (Z-direction acceleration) that gradually increases after passing through a section of small fluctuation following the maximum peak immediately after heel strike. The generator 123 may extract gait waveform data for one walking cycle using both the forward acceleration (Y-direction acceleration) and the vertical acceleration (Z-direction acceleration).The generator 123 may also extract gait waveform data for one walking cycle using acceleration, angular velocity, angle, etc. other than the forward acceleration (Y-direction acceleration) and the vertical acceleration (Z-direction acceleration).

[0052] For example, the generation unit 123 calculates a gait index using normalized walking waveform data. For example, the gait index is used to estimate physical ability, etc. There are no particular limitations on the gait index calculated by the generation unit 123. For example, the generation unit 123 calculates gait indices related to distance, height, angle, speed, time, CPEI (Center of Pressure Exclusion Index), frailty level, etc. Representative gait indices are listed below. Specific methods for calculating the following gait indices will be omitted.

[0053] For example, the generation unit 123 calculates indices related to distance and height as gait indices. For example, the generation unit 123 calculates a stride length, a turning distance, a foot lift height, FTC (Foot Clearance), and MTC (Minimum Toe Clearance). The stride length indicates the distance between the front foot and the rear foot while walking. The turning distance indicates the maximum distance that the foot is separated outward in the direction of travel during the swing phase. The foot lift height indicates the maximum distance between the measurement device 10 (sensor 110) and the ground during the swing phase. The FTC indicates the maximum distance between the heel and the ground during the swing phase. The MTC indicates the minimum distance between the toe and the ground during the swing phase.

[0054] For example, the generation unit 123 calculates angle-related indices as gait indices. For example, the generation unit 123 calculates a contact angle, a takeoff angle, a toe direction, a heel-strike roll angle, a toe-off roll angle, a swing leg peak angular velocity, and a hallux angle. The contact angle indicates the maximum value of the angle between the sole of the foot and the ground at heel-strike. The takeoff angle indicates the angle between the sole of the foot and the ground during the swing phase. The toe direction indicates the average value of the orientation of the toe relative to the direction of travel during the swing phase. The heel-strike roll angle is the angle between the ankle and the ground at heel-strike, as viewed from a rear perspective. The toe-off roll angle is the angle between the ankle and the ground at push-off, as viewed from a rear perspective. The swing leg peak angular velocity is the angular velocity in the ankle dorsiflexion direction during the period from immediately after push-off until the toe comes closest to the ground. The hallux angle indicates the angle at which the big toe is tilted toward the index toe. Specifically, the hallux angle is the angle between the center line of the first metatarsal and the center line of the first proximal phalanx.

[0055] For example, the generation unit 123 calculates an index related to speed as a gait index. For example, the generation unit 123 calculates walking speed, cadence, and maximum swing speed. Walking speed indicates the walking speed. Cadence indicates the number of steps per minute. Maximum swing speed indicates the speed at which the leg is swung out during the swing phase.

[0056] For example, the generation unit 123 calculates time-related indices as gait indices. For example, the generation unit 123 calculates stance time, load time, sole contact time, push-off time, swing time, and DST (Double Support Time). Stance time indicates the time during which the foot is in contact with the ground during walking. Stance time is the sum of load time, sole contact time, and push-off time. Load time is the time during the stance phase from when the heel contacts the ground to when the toe contacts the ground. Sole contact time is the time during the stance phase when the entire sole of the foot is in contact with the ground and is horizontal to the ground. Push-off time is the time during the stance phase from when the sole is in contact with the ground to when the toe pushes off the ground. Swing time indicates the time during which the foot is off the ground during walking. DST is divided into DST1 and DST2. DST1 indicates the time during which the foot equipped with the measuring device 10 (sensor 110) is in front of the other foot during a period when both feet are in contact with the ground at the same time, and DST2 indicates the time during which the foot equipped with the measuring device 10 (sensor 110) is behind the other foot during a period when both feet are in contact with the ground at the same time.

[0057] For example, the generating unit 123 calculates a center of pressure exclusion index (CPEI) as a gait index. The CPEI indicates an estimated rate of expansion of the center of foot pressure on the ground during the stance phase.

[0058] For example, the generation unit 123 calculates a frailty level as the gait index. The frailty level is an estimated value of the frailty state according to the walking state. For example, the generation unit 123 estimates an index indicating a determination result regarding frailty as the frailty level. If there is no possibility of frailty, the generation unit 123 estimates an index indicating that the subject is not frail. If there is a possibility of frailty, the generation unit 123 estimates an index indicating that the subject is likely to be frail. Furthermore, if there is a high possibility of frailty, the generation unit 123 estimates an index indicating that there is a high possibility of frailty.

[0059] The generation unit 123 may extract feature amounts used to calculate or estimate gait indices from the walking waveform data. For example, the generation unit 123 extracts feature amounts for each walking phase cluster according to preset conditions. A walking phase cluster is a cluster that integrates temporally consecutive walking phases. A walking phase cluster includes at least one walking phase. A walking phase cluster may also include a single walking phase. The generation unit 123 may extract physical ability feature amounts used to estimate physical abilities. For example, the physical ability feature amounts are used to estimate at least one of physical abilities such as grip strength (total muscle strength of the entire body), dynamic balance, lower limb muscle strength, mobility ability, and static balance.

[0060] The generation unit 123 outputs gait data including the normalized gait waveform data, gait indices, and feature amounts to the correction unit 125. The generation unit 123 outputs gait data including at least one of the normalized gait waveform data, gait indices, and feature amounts to the correction unit 125.

[0061] The correction unit 125 acquires the road surface category from the road surface discrimination unit 122. The correction unit 125 also acquires gait data from the generation unit 123. The gait data includes normalized gait waveform data, gait indices, and feature amounts. The correction unit 125 corrects the gait data in accordance with the road surface category using the gait data correction model 135. For example, the correction unit 125 corrects the gait waveform data in accordance with the road surface category using the gait data correction model 135. For example, the correction unit 125 corrects the gait indices in accordance with the road surface category using the gait data correction model 135. For example, the correction unit 125 corrects the feature amounts in accordance with the road surface category using the gait data correction model 135.

[0062] The gait data correction model 135 is a machine learning model. For example, the gait data correction model 135 is a model trained using a data set in which gait data and road surface category are used as explanatory variables and gait data on a reference road surface is used as a target variable, as training data. For example, the gait data correction model 135 is a learning model trained using a convolutional neural network (CNN) technique. For example, the gait data correction model 135 is a model trained using a principal component analysis (PCA) technique. For example, the gait data correction model 135 is a learning model trained using a variational autoencoder (VAE). For example, the gait data correction model 135 is a learning model trained using a conditional generative adversarial network (GAN) technique. The above techniques are merely examples and do not limit the techniques for training the gait data correction model 135.

[0063] The gait data correction model 135 outputs gait data (corrected gait data) normalized to a reference road surface in response to input gait data and a road surface category. The gait data includes normalized gait waveform data, gait indicators, and feature quantities. The reference road surface is a road surface that serves as a reference for the road surface conditions of a walkway. For example, the reference road surface is a walkway with a vinyl-coated surface that is easy to walk on, such as an indoor corridor. The gait data correction model 135 corrects gait data generated using sensor data measured during walking on walkways with various road surfaces, such as asphalt, grass, and sand, to gait data corresponding to walking on a walkway with the reference road surface. The gait data correction model 135 may be stored in an external storage device (not shown) built in the cloud, a server, or the like. In this case, the correction unit 125 uses the gait data correction model 135 via an interface (not shown) connected to the storage device.

[0064] FIG. 14 is a conceptual diagram for explaining an example of training of a gait data correction model according to the present disclosure. Gait data includes gait waveform data, gait indices, and feature quantities. FIG. 14 illustrates an example in which gait data measured on walkways of various road surface categories is corrected to gait waveform data for a walkway with a reference road surface. In the example of FIG. 14 , a dataset of gait data (explanatory variables) measured for each road surface category and reference gait data (objective variable) measured on a reference road surface is used as training data. A dataset measured for each subject for multiple subjects is used as training data. In the example of FIG. 14 , a gait data correction model 135 is trained using a dataset of gait data (explanatory variables) measured for each road surface category and gait data (objective variable) measured on a reference road surface as training data.

[0065] FIG. 15 is a conceptual diagram for explaining an example of gait data correction using a gait data correction model according to the present disclosure. FIG. 15 shows an example in which the gait data correction model 135 of FIG. 14 is used. Gait data and a road surface category are input to the gait data correction model 135. The gait data correction model 135 outputs corrected gait data in response to the input gait data and road surface category. The corrected gait data is data in which gait data measured while walking on walkways of various road surface categories has been corrected to gait data on a walkway with a reference road surface. By using the gait data correction model 135, gait data that changes depending on the road surface category can be corrected to gait data on the reference road surface.

[0066] FIG. 16 is a conceptual diagram for explaining an example of learning of a gait data correction model according to the present disclosure. FIG. 16 illustrates an example of correcting gait data according to weather. Weather is specified by weather that can affect road surface conditions, such as sunny, rainy, or snowy. In the example of FIG. 16 , sunny weather is defined as the reference weather. For example, even if it is raining, the reference weather indoors is defined as sunny. Furthermore, if the road surface is wet even on a sunny day, the road surface category may be determined according to the elapsed time since the rain stopped. FIG. 16 illustrates an example of correcting gait data measured in various weather conditions to reference gait data for a walkway with a reference road surface in reference weather. In the example of FIG. 16 , a dataset of gait data (explanatory variables) measured for each road surface category in various weather conditions and gait data (objective variable) measured on a reference road surface in reference weather is used as training data. It is preferable that the training data be a dataset measured for each subject for multiple subjects. In the example of FIG. 16, gait data correction model 135-1 is trained using a data set of gait data (explanatory variables) measured for each road surface category under various weather conditions and gait data (objective variables) measured on a reference road surface under reference weather conditions as training data.

[0067] FIG. 17 is a conceptual diagram for describing an example of gait data correction using a gait data correction model according to the present disclosure. FIG. 17 illustrates an example in which the gait data correction model 135-1 of FIG. 16 is used. Gait data, weather data, and road surface category are input to the gait data correction model 135-1. For example, the weather data is acquired from an application that provides weather forecast data or weather information. The gait data correction model 135-1 outputs corrected gait data in response to the input of gait data, weather data, and road surface category. The corrected gait data is data in which gait data measured while walking in various weather conditions has been corrected to gait data for a reference weather. Using the gait data correction model 135-1, gait data measured under various road surface conditions that change depending on the weather can be corrected to gait data for a reference road surface in the reference weather.

[0068] FIG. 18 is a conceptual diagram illustrating an example of learning a gait data correction model according to the present disclosure. FIG. 18 illustrates an example of correcting gait data according to the season. A person's walking state can change depending on environmental factors such as seasonal temperature and humidity. Therefore, seasons are identified based on seasons that can affect walking state, such as spring, summer, autumn, and winter. For example, spring or autumn, which are comfortable, is used as the reference season. For example, the seasons include rainy seasons such as the rainy season and autumn rains. Furthermore, a standard season for a location identified by longitude and latitude using location data may be used. FIG. 18 illustrates an example of correcting gait data measured in various seasons to reference gait data for a walkway with a reference road surface in the reference season. In the example of FIG. 18 , training data is a dataset of gait data (explanatory variables) measured for each road surface category in various seasons and gait data (objective variable) measured on a reference road surface in the reference season. It is preferable that the training data be a dataset measured for each of multiple subjects. In the example of FIG. 18, a data set of gait data (explanatory variables) measured for each road surface category in various seasons and gait data (objective variables) measured on a reference road surface in a reference season is used as training data to train gait data correction model 135-2.

[0069] FIG. 19 is a conceptual diagram for explaining an example of gait data correction using a gait data correction model according to the present disclosure. FIG. 19 illustrates an example in which the gait data correction model 135-2 of FIG. 18 is used. Gait data, seasonal data, and road surface category are input to the gait data correction model 135-2. For example, the seasonal data is identified by the date and month on which the sensor data was measured. For example, location data may be used to identify the seasonal data. Use of location data allows for more accurate identification of the season for each location. The gait data correction model 135-2 outputs corrected gait data in response to input of gait data, seasonal data, and road surface category. The corrected gait data is data obtained by correcting gait data measured during walking in various seasons to gait data for a reference season. Using the gait data correction model 135-2, gait data measured during various walking states that change with the seasons can be corrected to gait data for a reference road surface in the reference season.

[0070] The output unit 127 outputs the corrected gait data corrected by the correction unit 125. For example, the output unit 127 outputs the corrected gait data to a terminal device or a server that uses the corrected gait data. For example, the output unit 127 outputs the corrected gait data to a mobile terminal 170 carried by the subject. For example, the output unit 127 may be configured to output the corrected gait data to an external system or the like that uses the corrected gait data.

[0071] For example, the data generating device 12 is constructed in a cloud or a server connected to a mobile terminal 170 carried by the subject via a communication network. The mobile terminal 170 is a portable communication device. For example, the mobile terminal 170 is a mobile communication device with a communication function, such as a smartphone, a smartwatch, or a mobile phone. For example, the data generating device 12 is connected to the mobile terminal 170 via wireless communication. For example, the data generating device 12 is connected to the mobile terminal 170 via a wireless communication device (not shown) conforming to standards such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). Note that the wireless communication device may conform to standards other than Bluetooth (registered trademark) or Wi-Fi (registered trademark). The corrected gait data may be used by an application installed on the mobile terminal 170. For example, the mobile terminal 170 executes processing using the corrected gait data using an application installed on the mobile terminal 170.

[0072] FIG. 20 is a conceptual diagram showing an example of data flow between components constituting a data generation device according to the present disclosure. The acquisition unit 121 acquires, from a mobile terminal 170 carried by the subject, sensor data measured in accordance with the subject's walking and position data of the mobile terminal 170 at the time the sensor data was measured. The acquisition unit 121 outputs the position data to the road surface discrimination unit 122. The road surface discrimination unit 122 uses a road surface discrimination model 132 to discriminate a road surface category according to the position data. The road surface discrimination model 132 outputs a road surface category in response to input of the position data. The road surface discrimination unit 122 outputs the discriminated road surface category to the correction unit 125. The acquisition unit 121 also outputs the sensor data to the generation unit 123. The generation unit 123 generates gait data using the sensor data. The generation unit 123 outputs the generated gait data to the correction unit 125. The correction unit 125 acquires the road surface category from the road surface discrimination unit 122. The correction unit 125 also acquires gait data from the generation unit 123. The correction unit 125 corrects the gait data of the reference road surface to the gait data of the determined road surface category using the gait data correction model 135. The gait data correction model 135 outputs corrected gait data according to the input data of the road surface category and gait data. The correction unit 125 outputs the corrected gait data.

[0073] (Operation) Next, the operation of the gait measurement system 1 will be described with reference to the drawings. The operation of the data generating device 12 included in the gait measurement system 1 will be described below. FIG. 21 is a flowchart for explaining an example of the operation of the data generating device according to the present disclosure. In describing the processing according to the flowchart of FIG. 21, the components of the data generating device 12 will be described as the subject of the operations. The subject of the operations according to the flowchart of FIG. 21 may be the data generating device 12.

[0074] 21 , first, the acquisition unit 121 acquires sensor data and position data transmitted from the mobile terminal 170 carried by the subject (step S11). The sensor data includes acceleration in three axial directions and angular velocity around three axes. The position data corresponds to the position where the sensor data was measured. Following the processing of step S11, the processing of steps S12 and S13 is executed in parallel. Note that the processing of steps S12 and S13 may be executed sequentially. For example, the processing of step S12 may be executed followed by the processing of step S13. For example, the processing of step S12 may be executed followed by the processing of step S13.

[0075] After step S11, the road surface discrimination unit 122 discriminates a road surface category using the acquired position data (step S12). The road surface discrimination unit 122 discriminates the road surface category using the road surface discrimination model 132. The road surface discrimination model 132 outputs a road surface category in response to input position data.

[0076] After step S11, the generator 123 generates gait data using the acquired sensor data (step S13). For example, the gait data includes walking waveform data, gait indices, feature amounts, and the like.

[0077] Following steps S12 and S13, the correction unit 125 corrects the gait data to match the determined road surface category (step S14). The correction unit 125 corrects the gait data using a gait data correction model 135. The gait data correction model 135 outputs corrected gait data equivalent to the gait data for the reference road surface in response to the input road surface category and gait data.

[0078] Next, the output unit 127 outputs the corrected gait data (step S15). For example, the output unit 127 outputs the corrected gait data to a terminal device or a server that uses the corrected gait data. For example, the output unit 127 outputs the corrected gait data to a mobile terminal 170 carried by the subject. For example, the output unit 127 outputs the corrected gait data to an external system or the like that uses the corrected gait data.

[0079] (Application Example) Next, an application example according to this embodiment will be described with reference to the drawings. FIGS. 22 and 23 are conceptual diagrams showing an example of displaying information generated using gait data corrected by the data generating device according to the present disclosure. FIGS. 22 and 23 show an example in which information about a user's gait is displayed on the screen of a mobile terminal 170 carried by a user walking while wearing shoes 100 in which a measurement device 10 is placed. In the example using FIGS. 22 to 23 , information about the user's gait is generated by an information processing unit (not shown) that processes information using data processed by the data generating device 12.

[0080] FIG. 22 illustrates an example in which a road surface category determined by the data generating device 12 and advice corresponding to the road surface category are displayed on the screen of the mobile device 170. In the example of FIG. 22, information such as "You are walking on sand" is displayed on the screen of the mobile device 170 in accordance with the determined road surface category. Also, in the example of FIG. 22, recommendation information such as "Be careful not to get your feet stuck in the sand" is displayed on the screen of the mobile device 170 in accordance with the determination result of the road surface category. Also, in the example of FIG. 22, recommendation information such as "There is grass nearby. Please walk on that side" is displayed on the screen of the mobile device 170 in accordance with the determination result of the road surface category related to the user's vicinity. The recommendation information is optimized according to the road surface condition of the walkway on which the user is walking. Furthermore, the recommendation information prompts the user to make a decision. After checking the information displayed on the screen of the mobile device 170, the user can avoid falling or move to a road surface that is easier to walk on by taking action in accordance with the recommendation information.

[0081] FIG. 23 illustrates an example in which information about a physical condition estimated using corrected gait data is displayed on the screen of the mobile device 170. In the example of FIG. 23 , information about the physical condition, such as "You have a tendency toward frailty," estimated using the corrected gait data, is displayed on the screen of the mobile device 170. Also, in the example of FIG. 23 , recommendation information, such as "We recommend that you receive a medical examination at a hospital," is displayed on the screen of the mobile device 170 in accordance with the estimation result of the physical condition. The recommendation information is optimized according to the user's physical condition. The recommendation information also prompts the user to make a decision. For example, links to websites and telephone numbers of hospitals that offer medical examinations may be displayed on the screen of the mobile device 170. After checking the information displayed on the screen of the mobile device 170, the user can receive a medical examination for frailty by visiting a hospital in accordance with the recommendation information. The corrected gait data has been corrected to gait data for walking on a reference road surface. Therefore, using the corrected gait data reduces the likelihood of erroneous determination of the physical condition due to temporary fluctuations depending on the road surface category.

[0082] As described above, the gait measurement system of this embodiment includes a measurement device and a data generation device. The measurement device is attached to the user's footwear. The measurement device has a sensor that measures acceleration and angular velocity. The measurement device generates sensor data using the acceleration and angular velocity measured by the sensor. The measurement device transmits the generated sensor data to the data generation device. The data generation device includes an acquisition unit, a road surface discrimination unit, a generation unit, a correction unit, and an output unit. The acquisition unit acquires sensor data measured by the measurement device attached to the user's footwear and position data corresponding to the position where the sensor data was acquired. The road surface discrimination unit discriminates the road surface condition of the walkway at the position corresponding to the position data. The generation unit generates gait data using the sensor data. The correction unit corrects the gait data to match the road surface condition of the walkway. The output unit outputs the corrected gait data.

[0083] The gait measurement system of this embodiment determines the surface conditions of the walkway in accordance with position data indicating the user's position. The gait measurement system of this embodiment corrects gait data in accordance with the determined surface conditions of the walkway. Therefore, this embodiment can generate gait data corrected in accordance with the surface conditions of the walkway.

[0084] In one aspect of the present embodiment, the road surface discrimination unit uses a road surface discrimination model to discriminate the road surface category of the walkway at the user's position. The road surface discrimination model outputs the road surface category of the walkway at the position of the position data in response to input position data. The generation unit uses a gait data correction model to correct gait data measured in response to the user's walking to gait data for walking on a reference road surface. The gait data correction model outputs gait data for walking on a reference road surface in response to input gait data and a road surface category. According to this aspect, the road surface discrimination model can be used to discriminate the road surface category of the walkway on which the user is walking. Furthermore, according to this aspect, the gait data correction model can be used to generate gait data corrected in response to the road surface condition of the walkway.

[0085] In one aspect of this embodiment, the road surface discrimination model is a machine learning model generated by machine learning using, as training data, a dataset in which position data is an explanatory variable and road surface category is an objective variable. The gait data correction model is a machine learning model trained by machine learning using, as training data, a dataset in which gait data measured for each road surface category is an explanatory variable and reference gait data measured on a reference road surface is an objective variable. For example, the gait data correction model outputs gait data for walking on a reference road surface in reference weather in response to input gait data, road surface category, and weather data. According to this aspect, gait data corrected in accordance with the road surface condition of the walkway can be generated using the road surface discrimination model and gait data correction model generated by machine learning.

[0086] In one aspect of the present embodiment, the output unit displays information corresponding to the road surface condition of the walkway on the screen of the mobile terminal used by the user, allowing the user to check the information corresponding to the road surface condition of the walkway on the spot.

[0087] In one aspect of the present embodiment, the output unit displays information optimized for the road surface conditions of the walkway along which the user is walking, and prompts the user to make a decision, on the screen of the mobile device carried by the user. According to this aspect, the information optimized for the road surface conditions of the walkway can prompt the user to make a decision.

[0088] Second Embodiment Next, a gait measurement system according to a second embodiment will be described with reference to the drawings. The gait measurement system of this embodiment differs from the gait measurement system according to the first embodiment in that it corrects gait data in accordance with the accuracy of position data.

[0089] (Configuration) FIG. 24 is a block diagram showing an example of the configuration of a gait measurement system according to the present disclosure. The gait measurement system 2 includes a measurement device 20 and a data generating device 22. The measurement device 20 has a configuration similar to that of the measurement device 10 of the first embodiment. For example, the measurement device 20 is installed in footwear of a user whose gait is to be measured. For example, the data generating device 22 is implemented on a server or a cloud. The server or cloud is connected to a mobile device carried by the user via a network such as the Internet. For example, the functions of the data generating device 22 may be implemented on a mobile device carried by the user. In the following, a description of the measurement device 20 will be omitted, and only the data generating device 22 will be described. The data generating device 22 has functions in common with the data generating device 12 of the first embodiment. In the following, descriptions of functions in common with the data generating device 12 of the first embodiment will be simplified.

[0090] [Data Generating Device] Fig. 25 is a block diagram showing an example of the configuration of the data generating device 22. The data generating device 22 has an acquiring unit 221, a road surface determining unit 222, a generating unit 223, a correcting unit, and an output unit 227. For example, the data generating device 22 is constructed on a server or in the cloud. For example, the functions of the data generating device 22 may be implemented in a mobile terminal carried by a user. For example, the functions of the data generating device 22 may be implemented in a terminal device used by an administrator who manages gait data measured for users.

[0091] The data generating device 22 also has a road surface discrimination model 232 and a category-specific gait data correction model 235. For example, the road surface discrimination model 232 and the category-specific gait data correction model 235 are stored in the storage unit 23. The road surface discrimination model 232 and the category-specific gait data correction model 235 may be stored in a storage device (not shown) accessible from the data generating device 22. In this case, the data generating device 22 uses the road surface discrimination model 232 and the category-specific gait data correction model 235 via an interface (not shown) such as an API (Application Programming Interface) connected over a network.

[0092] The acquisition unit 221 has the same configuration as the acquisition unit 121 of the first embodiment. The acquisition unit 221 acquires sensor data from a measurement device 20 mounted on the user's footwear. The acquisition unit 221 receives the sensor data from the measurement device 20 via wireless communication. The sensor data includes location information of the mobile device that is the sender of the sensor data. For example, the location information is measured by a global positioning system (GPS) function mounted on the mobile device and added to the sensor data.

[0093] The road surface discriminator 222 discriminates a road surface category corresponding to the position data, similar to the road surface discriminator 122 of the first embodiment. The road surface discriminator 222 differs from the road surface discriminator 122 in that it discriminates road surface categories at multiple positions included within a circle of an accuracy radius centered on the position of the position data. When the position data is measured by GPS, the accuracy radius corresponds to a radius according to the GPS positioning accuracy. For example, if the GPS positioning accuracy is 10 meters, the accuracy radius is set to 10 meters. Hereinafter, the range within the circle of the accuracy radius is also referred to as the accuracy range.

[0094] The road surface discrimination unit 222 acquires position data from the acquisition unit 221. The road surface discrimination unit 222 discriminates road surface categories at multiple positions included within a circle of an accuracy radius centered on the position of the acquired position data. That is, the road surface discrimination unit 222 discriminates road surface categories within an accuracy range. The road surface discrimination unit 222 uses a road surface discrimination model 232 to discriminate road surface categories at multiple positions included in the accuracy range. The granularity of positions at which road surface categories are discriminated within the accuracy range is set arbitrarily. The road surface discrimination unit 222 outputs the road surface categories discriminated for the multiple positions included in the accuracy range to the correction unit 225.

[0095] FIG. 26 is a conceptual diagram illustrating how multiple road surface categories are included in the accuracy range centered on the position of the position data. In FIG. 26, the accuracy range is an area included inside a circle G with an accuracy radius centered on the position of the position data. The accuracy range includes areas of multiple road surface categories. For example, area R A The road surface category of the area R is asphalt. G The road surface category of the area R is grass. S The road surface category of is sand. In this embodiment, the gait data is corrected according to the error in the position data.

[0096] The road surface discrimination model 232 has the same configuration as the road surface discrimination model 132 of the first embodiment. The road surface discrimination model 232 is a machine learning model. In response to input position data, the road surface discrimination model 232 outputs a road surface category corresponding to the position data. The road surface category includes a reference road surface that serves as a reference. The reference road surface is a road surface that serves as a reference for the road surface condition of a walkway. For example, the reference road surface is a walkway with a vinyl-coated surface that is easy to walk on, such as an indoor corridor. For example, road surface categories are classified into road surfaces that exhibit walking characteristics, such as asphalt, grass, and sand. Depending on the road surface category of the walkway, the characteristics of walking on that walkway appear in the walking waveform. The road surface discrimination model 232 may be stored in an external storage device constructed in a cloud, a server, or the like. In this case, the road surface discrimination unit 222 uses the road surface discrimination model 232 via an interface (not shown) connected to the storage device.

[0097] The generation unit 223 has the same configuration as the generation unit 123 in the first embodiment. The generation unit 223 acquires sensor data from the acquisition unit 221. The generation unit 223 extracts time series data for one walking cycle from the time series data of acceleration in three axial directions and angular velocity around three axes included in the sensor data. The time series data for one walking cycle is also called walking waveform data. The generation unit 223 extracts walking waveform data based on the timing of walking events detected from the time series data of the sensor data. For example, the generation unit 223 extracts walking waveform data that starts at the timing of a heel strike and ends at the timing of the next heel strike.

[0098] The generation unit 223 normalizes (first normalization) the time of the extracted gait waveform data for one step cycle to a gait cycle of 0 to 100% (percent). Furthermore, the generation unit 223 normalizes (second normalization) the first-normalized gait waveform data for one step cycle so that the stance phase is 60% and the swing phase is 40%. For example, the generation unit 223 calculates a gait index using the normalized gait waveform data. For example, the generation unit 223 extracts, from the gait waveform data, feature amounts used for calculating or estimating the gait index. The generation unit 223 outputs gait data including the normalized gait waveform data, gait index, and feature amounts to the correction unit 225. The generation unit 223 outputs gait data including at least one of the normalized gait waveform data, gait index, and feature amounts to the correction unit 225.

[0099] The correction unit 225 acquires road surface categories included in the accuracy range from the road surface determination unit 222. The correction unit 225 acquires road surface categories determined for a plurality of positions included in a circle of an accuracy radius centered on the position of the position data.

[0100] The correction unit 225 also acquires gait data from the generation unit 223. The gait data includes normalized gait waveform data, gait indices, and feature quantities. The correction unit 225 uses the category-specific gait data correction model 235 to correct the gait data according to the road surface category included in the accuracy range. The correction unit 225 uses the category-specific gait data correction model 235 to set a weight for each of multiple road surface categories according to the area ratio of the road surface category included in the accuracy range. The area ratio of the road surface category included in the accuracy range can be calculated according to the ratio of the road surface category determined at multiple positions included in the accuracy range. For example, assume that road surface categories are determined at 50 positions within the accuracy range, and that 10 of those positions are determined to be road surface category a. In this case, the area ratio (weight) of road surface category a is set to 0.2 (= 10 / 50). For example, assume that road surface category a is not determined within the accuracy range. In this case, the area ratio (weight) of road surface category a is set to 0.

[0101] The categorical gait data correction model 235 is a machine learning model. The categorical gait data correction model 235 is generated for each road surface category. For example, the categorical gait data correction model 235 is a model trained using a data set, for each road surface category, in which gait data is used as an explanatory variable and gait data on a reference road surface is used as a target variable, as training data. For example, the categorical gait data correction model 235 is a learning model trained using a convolutional neural network (CNN) technique. For example, the categorical gait data correction model 235 is a model trained using a principal component analysis (PCA) technique. For example, the categorical gait data correction model 235 is a learning model trained using a variational autoencoder (VAE). For example, the categorical gait data correction model 235 is a learning model trained using a conditional generative adversarial network (GAN) technique. The above-mentioned method is merely an example, and does not limit the method for training the category-specific gait data correction model 235.

[0102] The categorical gait data correction model 235 outputs gait data (corrected gait data) normalized to a reference road surface in response to input gait data. The gait data includes normalized gait waveform data, gait indices, and feature quantities. The reference road surface is a road surface that serves as a reference for the road surface conditions of a walkway. For example, the reference road surface is a walkway with a vinyl-coated surface that is easy to walk on, such as an indoor corridor. The categorical gait data correction model 235 corrects gait data generated using sensor data measured during walking on walkways with various road surfaces, such as asphalt, grass, and sand, to gait data corresponding to walking on a walkway with the reference road surface. The categorical gait data correction model 235 may be stored in an external storage device (not shown) constructed on a cloud, a server, or the like. In this case, the correction unit 225 uses the categorical gait data correction model 235 via an interface (not shown) connected to the storage device.

[0103] The correction unit 225 multiplies each of the corrected gait data for each road surface category included in the accuracy range by a weight corresponding to the area ratio of the road surface category. The correction unit 225 calculates the sum of the corrected gait data for each road surface category weighted according to the area ratio of the road surface category as the corrected gait data.

[0104] FIG. 27 is a conceptual diagram illustrating an example of calculating corrected gait data using gait data corrected for each road surface category. In the example of FIG. 27, category-specific gait data correction models 235-1 to 235-m are arranged for each road surface category (m is a natural number). Category-specific gait data correction model 235-1 outputs corrected gait data 1 in response to input of gait data. Corrected gait data 1 is multiplied by a weight w1 corresponding to the area ratio included in the accuracy range. Category-specific gait data correction model 235-2 outputs corrected gait data 2 in response to input of gait data. Corrected gait data 2 is multiplied by a weight w2 corresponding to the area ratio included in the accuracy range. Category-specific gait data correction model 235-m outputs corrected gait data m in response to input of gait data. Corrected gait data m is multiplied by a weight w1 corresponding to the area ratio included in the accuracy range. mFor example, the corrected gait data for a road surface category not included in the accuracy range is multiplied by a weight of 0. The sum of the weighted corrected gait data for each road surface category is calculated as the corrected gait data.

[0105] The output unit 227 has the same configuration as the output unit 127 of the first embodiment. The output unit 227 outputs corrected gait data corrected by the correction unit 225. For example, the output unit 227 outputs the corrected gait data to a terminal device or a server that uses the corrected gait data. For example, the output unit 227 outputs the corrected gait data to a mobile terminal carried by a user. For example, the output unit 227 may be configured to output the corrected gait data to an external system or the like that uses the corrected gait data.

[0106] For example, the data generating device 22 is constructed in a cloud or a server connected to a mobile device carried by a user via a communication network. For example, the data generating device 22 is connected to the mobile device via wireless communication. For example, the data generating device 22 is connected to the mobile device via a wireless communication device (not shown) conforming to standards such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). Note that the wireless communication device may conform to standards other than Bluetooth (registered trademark) or Wi-Fi (registered trademark). The corrected gait data may be used by an application installed on the mobile device. For example, the mobile device executes processing using the corrected gait data using an application installed on the mobile device.

[0107] FIG. 28 is a conceptual diagram illustrating an example of data flow between components constituting a data generation device according to the present disclosure. The acquisition unit 221 acquires, from a portable device carried by a user, sensor data measured in response to the user's walking and position data of the portable device at the time the sensor data was measured. The acquisition unit 221 outputs the position data to the road surface discrimination unit 222. The road surface discrimination unit 222 uses a road surface discrimination model 232 to discriminate road surface categories at multiple positions included in the accuracy range. The road surface discrimination model 232 outputs road surface categories in response to input of position data corresponding to multiple positions included in the accuracy range. For example, the road surface discrimination model 232 may be configured to output road surface categories corresponding to multiple positions included in the accuracy range centered on the input position data in response to input of the position data. The road surface discrimination unit 222 outputs the road surface categories determined for the multiple positions included in the accuracy range to the correction unit 225. The acquisition unit 221 also outputs sensor data to the generation unit 223. The generation unit 223 generates gait data using the sensor data. The generation unit 223 outputs the generated gait data to the correction unit 225. The correction unit 225 acquires road surface categories determined for multiple positions included in the accuracy range from the road surface determination unit 222. The correction unit 225 also acquires gait data from the generation unit 223. The correction unit 225 corrects the gait data using a category-specific gait data correction model 235 to match the gait data to the road surface category included in the accuracy range. The category-specific gait data correction model 235 outputs corrected gait data for each road surface category according to the input gait data. The correction unit 225 multiplies the corrected gait data for each road surface category by a weight for that road surface category. The correction unit 225 calculates the sum of the corrected gait data for each road surface category multiplied by the weight for that road surface category as the corrected gait data. The correction unit 225 outputs the corrected gait data.

[0108] (Operation) Next, the operation of the gait measurement system 2 will be described with reference to the drawings. The operation of the data generating device 22 included in the gait measurement system 2 will be described below. FIG. 29 is a flowchart for explaining an example of the operation of the data generating device according to the present disclosure. In describing the processing according to the flowchart of FIG. 29, the components of the data generating device 22 will be described as the subject of the operations. The subject of the operations according to the flowchart of FIG. 29 may be the data generating device 22.

[0109] 29 , first, the acquisition unit 221 acquires sensor data and position data transmitted from a mobile terminal carried by a user (step S21). The sensor data includes acceleration in three axial directions and angular velocity around three axes. The position data corresponds to the position where the sensor data was measured. Following the processing of step S21, the processing of steps S22 and S23 is executed in parallel. Note that the processing of steps S22 and S23 may be executed sequentially. For example, the processing of step S22 may be followed by the processing of step S23. For example, the processing of step S22 may be followed by the processing of step S23.

[0110] After step S21, the road surface discrimination unit 222 discriminates road surface categories at multiple positions included in the accuracy range corresponding to the acquired position data (step S22). The road surface discrimination unit 222 discriminates road surface categories at multiple positions included in the accuracy range using the road surface discrimination model 232. The road surface discrimination model 232 outputs road surface categories for each of the multiple positions in response to input of position data corresponding to the multiple positions included in the accuracy range.

[0111] After step S21, the generator 223 generates gait data using the acquired sensor data (step S23). For example, the gait data includes walking waveform data, gait indices, feature amounts, and the like.

[0112] Following steps S22 and S23, the correction unit 225 corrects the gait data for each road surface category included in the accuracy range (step S24). The correction unit 225 corrects the gait data using a category-specific gait data correction model 235. In response to input gait data, the category-specific gait data correction model 235 outputs corrected gait data corresponding to the gait data for the reference road surface for each road surface category.

[0113] Next, the corrector 225 multiplies the corrected gait data for each road surface category by a weight corresponding to the area proportion of the road surface category included in the accuracy range (step S25).

[0114] Next, the correcting unit 225 calculates the sum of the corrected gait data for each road surface category, which has been multiplied by a weight according to the area ratio, as the corrected gait data (step S26).

[0115] Next, the output unit 227 outputs the corrected gait data (step S27). For example, the output unit 227 outputs the corrected gait data to a terminal device or a server that uses the corrected gait data. For example, the output unit 227 outputs the corrected gait data to a mobile terminal carried by the user. For example, the output unit 227 outputs the corrected gait data to an external system or the like that uses the corrected gait data.

[0116] (Application Example) Next, an application example according to this embodiment will be described with reference to the drawings. FIGS. 30 to 36 are conceptual diagrams showing display examples of information generated using gait data corrected by the data generating device according to the present disclosure. FIGS. 30 to 36 are examples in which information about a user's gait is displayed on the screen of a mobile terminal 270 carried by a user walking while wearing shoes 200 in which a measurement device 20 is disposed. In the examples using FIGS. 30 to 36, information about the user's gait is generated by an information processing unit (not shown) that processes information using data processed by the data generating device 22.

[0117] FIG. 30 shows an example of a training menu displayed for a user using the gait measurement system 2 on the screen of a portable device 270 carried by the user. In the example of FIG. 30 , information including a training menu such as "Today's target number of steps is 6,000 steps" is displayed on the screen of the portable device 270. For example, the exercise load included in the training menu is set according to the user's attributes such as age, gender, and health condition. Also in the example of FIG. 30 , information prompting the user to input an operation such as "Please enter the route you plan to walk on the map" is displayed on the screen of the portable device 270. A button B1 for displaying a map is displayed on the screen of the portable device 270. In response to selection of button B1, a map of the user's surroundings is displayed on the screen of the portable device 270.

[0118] FIG. 31 shows an example of a map of the user's surroundings displayed on the screen of the mobile terminal 270. The user's house is displayed on the map. The user's house is both the start point and the finish point. Around the user's house, there is an area R whose road surface category is asphalt. A , the area R where the road surface category is grass G , area R where the road surface category is sand S , and the restricted area R N Includes:

[0119] Fig. 32 shows an example in which a planned walking route is input on the screen of the mobile terminal 270. In the example of Fig. 32, the route is input by tracing a finger on a map displayed on the screen of the mobile terminal 270. A solid line of the route is displayed at the position traced with the finger. An estimate of the number of steps from home, which is the starting point, is displayed at the position where the finger is touching. In the example of Fig. 32, the position where the finger is touching corresponds to 3,000 steps. If the user turns back to home at the position where the finger is touching, it corresponds to a total of 6,000 steps. When the user releases their finger from the screen, the route is input.

[0120] 33 is an example of a recommended route according to road surface category displayed on the screen of the mobile terminal 270. The input route (dotted line) includes a grass area R G and the area R where the road surface category is sand. SThe road surface category of a walkway with grass or sand imposes a greater load than the reference road surface. Therefore, the information processing unit sets a point 2,750 steps from the house, which is the starting point, as a recommended turning back point, which corresponds to 3,000 steps on the reference road surface, according to the road surface category included in the input route.

[0121] FIG. 34 shows an example in which information including a recommended route according to a road surface category is displayed on the screen of the mobile device 270. In the example of FIG. 34, the recommended route in the example of FIG. 33 is displayed on the screen of the mobile device 270. Also displayed on the screen of the mobile device 270 is recommendation information stating, "This route includes an area that places a burden on walking, so the following route is recommended." The recommendation information is optimized according to the road surface conditions of the path along which the user is walking. The recommendation information also prompts the user to make a decision. After checking the information displayed on the screen of the mobile device 270, the user can practice an exercise appropriate for themselves by walking the recommended route. The recommended route calculated using the corrected gait data has been corrected to be an exercise appropriate for the reference road surface. Using the corrected gait data, it is possible to recommend an exercise with an appropriate burden according to the road surface category.

[0122] 35 is an example of a recommended route that takes into account the influence of weather on the screen of the mobile terminal 270. In FIG. 35, the recommended route is displayed in the area R whose road surface category is asphalt according to the road surface category after the rain. A The input route (dotted line) includes an area R whose road surface category is grass. G and the area R where the road surface category is sand. S The walking path with the road surface category of grass or sand may be wet after rain. Therefore, the information processing unit selects the asphalt area R that is less affected by rain according to the road surface category included in the input route. A Set a recommended route that includes many of these.

[0123] FIG. 36 illustrates an example in which information including a recommended route that takes into account the influence of weather is displayed on the screen of the mobile device 270. In the example of FIG. 36, the recommended route in the example of FIG. 35 is displayed on the screen of the mobile device 270. Also, the screen of the mobile device 270 displays recommendation information according to the weather, such as, "Since it has just rained, the following route is recommended." For example, when it rains, a higher weight may be assigned to areas of asphalt that are easier to walk on compared to other road surface categories. For example, when the weather is fine, a higher weight may be assigned to areas of grass that are more comfortable to walk on compared to other road surface categories. In this way, road surface categories according to weather can be arbitrarily set to have a positive effect on the user's walking. The recommendation information is optimized according to the road surface conditions of the path the user will be walking on. The recommendation information also prompts the user to make a decision. After checking the information displayed on the screen of the mobile device 270, a user can walk along the recommended route, avoiding areas where the road surface may be wet. For example, recommended routes according to road surface categories may be provided according to the season. For example, the recommended route may be set based on the number of steps that takes into account energy consumption according to temperature and humidity, such as being shorter in summer and longer in winter.

[0124] As described above, the gait measurement system of this embodiment includes a measurement device and a data generation device. The measurement device is attached to the user's footwear. The measurement device has a sensor that measures acceleration and angular velocity. The measurement device generates sensor data using the acceleration and angular velocity measured by the sensor. The measurement device transmits the generated sensor data to the data generation device. The data generation device includes an acquisition unit, a road surface discrimination unit, a generation unit, a correction unit, and an output unit. The acquisition unit acquires sensor data measured by the measurement device attached to the user's footwear and position data corresponding to the position at which the sensor data was acquired. The road surface discrimination unit discriminates the road surface condition of the walkway at a position corresponding to the position data. The road surface discrimination unit discriminates a road surface category for each of a plurality of positions within the accuracy range of the position data. The generation unit generates gait data using the sensor data. The correction unit corrects the gait data to match the road surface condition of the walkway. The correction unit corrects the gait data according to the road surface category for each of a plurality of positions within the accuracy range. The correction unit multiplies each of the plurality of gait data corrected for each road surface category by a weight corresponding to the area proportion of the road surface category included in the accuracy range. The correction unit calculates, as corrected gait data, the sum of the plurality of gait data corrected for each road surface category multiplied by the weight corresponding to the area proportion of the road surface category. The output unit outputs the corrected gait data.

[0125] The gait measurement system of this embodiment determines the surface conditions of the walkway at multiple positions included within the accuracy range in accordance with position data indicating the user's position. The gait measurement system of this embodiment corrects the gait data in accordance with the surface conditions of the walkway determined for each of the multiple positions included within the accuracy range. Therefore, according to this aspect, it is possible to generate gait data that has been corrected by taking into account the accuracy of the position data.

[0126] Third Embodiment Next, a data generating device according to a third embodiment will be described with reference to the drawings. The data generating device according to this embodiment has a simplified configuration of the data generating device included in the gait measurement systems according to the first and second embodiments.

[0127] 37 is a block diagram showing an example of the configuration of the data generating device 32 according to the present disclosure. The data generating device 32 includes an acquiring unit 321, a road surface determining unit 322, a generating unit 323, a correcting unit 325, and an output unit 327.

[0128] The acquisition unit 321 acquires sensor data measured by a measuring device mounted on the user's footwear and position data corresponding to the position where the sensor data was acquired. The road surface determination unit 322 determines the road surface condition of the walkway at the position corresponding to the position data. The generation unit 323 generates gait data using the sensor data. The correction unit 325 corrects the gait data to match the road surface condition of the walkway. The output unit 327 outputs the corrected gait data.

[0129] (Operation) Next, the operation of the data generating device 32 will be described with reference to the drawings. Fig. 38 is a flowchart for explaining an example of the operation of the data generating device 32. In explaining the processing according to the flowchart of Fig. 38, the components of the data generating device 32 will be described as the subject of the operations. The subject of the operations according to the flowchart of Fig. 38 may be the data generating device 32.

[0130] In FIG. 38, first, the acquisition unit 321 acquires sensor data measured by a measurement device mounted on the user's footwear and position data corresponding to the position where the sensor data was acquired (step S31).

[0131] After step S31, the road surface determining unit 322 determines the road surface condition of the walkway at the position corresponding to the position data (step S32).

[0132] After step S31, the generator 323 generates gait data using the sensor data (step S33).

[0133] Following steps S32 and S33, the correction unit 325 corrects the gait data in accordance with the road surface conditions of the walkway (step S34).

[0134] Next, the output unit 327 outputs the corrected gait data (step S35).

[0135] As described above, the data generating device of this embodiment determines the surface condition of the walkway in accordance with position data indicating the user's position. The data generating device of this embodiment corrects gait data in accordance with the determined surface condition of the walkway. Therefore, this embodiment can generate gait data corrected in accordance with the surface condition of the walkway.

[0136] (Hardware) Next, a hardware configuration for executing the control and processing in the present disclosure will be described with reference to the drawings. Here, an information processing device 90 (computer) in Fig. 37 is given as an example of such a hardware configuration. The information processing device 90 in Fig. 37 is an example configuration for executing the control and processing in the present disclosure and does not limit the scope of the present disclosure.

[0137] As shown in Fig. 37 , an information processing device 90 includes a processor 91, a memory 92, an auxiliary storage device 93, an input / output interface 95, and a communication interface 96. In Fig. 37 , interface is abbreviated as I / F (Interface). The processor 91, memory 92, auxiliary storage device 93, input / output interface 95, and communication interface 96 are connected to each other via a bus 98 so as to be able to communicate data with each other. The processor 91, memory 92, auxiliary storage device 93, and input / output interface 95 are also connected to a network such as the Internet or an intranet via the communication interface 96.

[0138] The processor 91 loads a program (instructions) stored in an auxiliary storage device 93 or the like into the memory 92. For example, the program is a software program for executing the control and processing in the present disclosure. The processor 91 executes the program loaded into the memory 92. The processor 91 executes the program to execute the control and processing in the present disclosure.

[0139] The memory 92 is a storage device having an area in which a program is loaded. The processor 91 loads a program stored in an auxiliary storage device 93 or the like into the memory 92. The memory 92 is realized by a volatile memory such as a dynamic random access memory (DRAM). Alternatively, a non-volatile memory such as a magnetoresistive random access memory (MRAM) may be used as the memory 92.

[0140] The auxiliary storage device 93 stores various data such as programs. For example, the auxiliary storage device 93 is realized by a local disk such as a hard disk or flash memory. Note that it is also possible to configure the system so that various data is stored in the memory 92, thereby omitting the auxiliary storage device 93.

[0141] The input / output interface 95 is an interface for connecting the information processing device 90 to peripheral devices based on standards and specifications. The communication interface 96 is an interface for connecting to external systems and devices via a network such as the Internet or an intranet based on standards and specifications. The input / output interface 95 and the communication interface 96 may be a common interface for connecting to external devices.

[0142] Input devices such as a keyboard, mouse, and touch panel may be connected to the information processing device 90 as needed. These input devices are used to input information and settings. When a touch panel is used as the input device, a screen having the function of the touch panel serves as the interface. The processor 91 and the input devices are connected via an input / output interface 95.

[0143] The information processing device 90 may be equipped with a display device for displaying information. When the display device is equipped, the information processing device 90 is equipped with a display control device (not shown) for controlling the display of the display device. The information processing device 90 and the display device are connected via an input / output interface 95.

[0144] The information processing device 90 may be equipped with a drive device. The drive device acts as an intermediary between the processor 91 and a recording medium (program recording medium) to read data and programs stored on the recording medium and to write processing results of the information processing device 90 to the recording medium. The information processing device 90 and the drive device are connected via an input / output interface 95.

[0145] The above is an example of a hardware configuration for enabling the control and processing of the present disclosure. The hardware configuration of Figure 37 is an example of a hardware configuration for executing the control and processing of the present disclosure and does not limit the scope of the present disclosure. A program that causes a computer to execute the control and processing of the present disclosure is also included in the scope of the present disclosure.

[0146] A program recording medium on which a program for executing the processing of this embodiment is recorded is also included within the scope of the present invention. For example, the program recording medium is a computer-readable, non-transitory recording medium. The recording medium can be, for example, an optical recording medium such as a CD (Compact Disc) or a DVD (Digital Versatile Disc). The recording medium may also be a semiconductor recording medium such as a USB (Universal Serial Bus) memory or an SD (Secure Digital) card. The recording medium may also be a magnetic recording medium such as a flexible disk, or other recording medium.

[0147] The components in the present disclosure may be combined in any manner. The components in the present disclosure may be realized by software. The components in the present disclosure may be realized by circuits.

[0148] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0149] Some or all of the above embodiments may be described as, but are not limited to, the following supplementary notes. In the supplementary notes below, dependent claims in each category may also be made dependent on other categories. The statements contained in the supplementary notes below serve as grounds for amendment. (Supplementary Note 1) A data generation device comprising: an acquisition unit that acquires sensor data measured by a measurement device mounted on a user's footwear and position data corresponding to the position at which the sensor data was acquired; a road surface determination unit that determines the surface condition of a walkway at a position corresponding to the position data; a generation unit that generates gait data using the sensor data; a correction unit that corrects the gait data in accordance with the surface condition of the walkway; and an output unit that outputs the corrected gait data. (Supplementary Note 2) The data generating device according to Supplementary Note 1, wherein the road surface discrimination unit discriminates the road surface category of the walkway at the user's position using a road surface discrimination model that outputs a road surface category of the walkway at the position of the position data in response to input of the position data, and the generation unit corrects the gait data measured in response to the user walking to the gait data for walking on a reference road surface using a gait data correction model that outputs the gait data for walking on a reference road surface in response to input of the gait data and the road surface category. (Supplementary Note 3) The data generating device according to Supplementary Note 2, wherein the road surface discrimination model is a machine learning model generated by machine learning using a dataset that has the position data as an explanatory variable and the road surface category as a response variable as training data, and the gait data correction model is a machine learning model trained by machine learning using a dataset that has the gait data measured for each road surface category as an explanatory variable and reference gait data measured on the reference road surface as a response variable. (Appendix 4) The data generating device described in Appendix 2, wherein the road surface discrimination unit corrects the gait data measured in response to the user's walking to the gait data in response to walking on the reference road surface in the reference weather, using a gait data correction model that outputs the gait data in response to input of the gait data, the road surface category, and weather data.(Supplementary Note 5) The data generating device according to Supplementary Note 2, wherein the road surface determining unit determines the road surface category for each of a plurality of positions within an accuracy range of the position data, and the correcting unit corrects the gait data according to the road surface category for each of the plurality of positions within the accuracy range, multiplies each of the plurality of gait data corrected for each road surface category by a weight according to the area proportion of the road surface category included in the accuracy range, and calculates, as the corrected gait data, a sum of the plurality of gait data corrected for each road surface category multiplied by the weight according to the area proportion of the road surface category. (Supplementary Note 6) The data generating device according to Supplementary Note 1, wherein the output unit displays, on a screen of a portable terminal used by the user, information according to the road surface condition of the walkway along which the user is walking. (Supplementary Note 7) The data generating device according to Supplementary Note 6, wherein the output unit displays, on a screen of the portable terminal carried by the user, information optimized according to the road surface condition of the walkway along which the user is walking and which prompts the user to make a decision. (Supplementary Note 8) A gait measurement system comprising: the data generating device according to any one of Supplements 1 to 7; and a measurement device having a sensor for measuring acceleration and angular velocity, generating the sensor data using the acceleration and angular velocity measured by the sensor, and transmitting the generated sensor data to the data generating device. (Supplementary Note 9) A data generation method comprising: a computer acquiring sensor data measured by a measurement device mounted on a user's footwear and position data corresponding to a position at which the sensor data was acquired, determining a surface condition of a walkway at a position corresponding to the position data, generating gait data using the sensor data, correcting the gait data to match the surface condition of the walkway, and outputting the corrected gait data.(Supplementary Note 10) A data generation method as described in Supplementary Note 9, in which a computer: determines the road surface category of the walkway at the user's position using a road surface discrimination model that outputs the road surface category of the walkway at the position of the position data in response to input of the position data; and corrects the gait data measured in response to the user's walking to the gait data in response to walking on a reference road surface using a gait data correction model that outputs the gait data in response to input of the gait data and the road surface category. (Supplementary Note 11) The data generation method according to Supplementary Note 10, wherein the road surface category of the walkway at the user's position is discriminated using the road surface discrimination model, which is a machine learning model generated by machine learning using as training data a dataset in which the position data is an explanatory variable and the road surface category is a response variable, and the gait data measured in response to the user's walking is corrected to the gait data for walking on the reference road surface using the gait data correction model, which is a machine learning model trained by machine learning using as training data a dataset in which the gait data measured for each road surface category is used as an explanatory variable and reference gait data measured on the reference road surface is used as a response variable. (Supplementary Note 12) The data generation method according to Supplementary Note 10, wherein a computer corrects the gait data measured in response to the user's walking to the gait data for walking on the reference road surface in the reference weather using a gait data correction model that, in response to input of the gait data, the road surface category, and weather data, outputs the gait data for walking on the reference road surface in the reference weather. (Supplementary Note 13) A data generation method as set forth in Supplementary Note 10, in which a computer determines the road surface category for each of a plurality of positions within an accuracy range of the position data, corrects the gait data according to the road surface category for each of the plurality of positions within the accuracy range, multiplies each of the plurality of gait data corrected for each road surface category by a weight according to the area proportion of the road surface category included in the accuracy range, and calculates, as the corrected gait data, the sum of the plurality of gait data corrected for each road surface category multiplied by the weight according to the area proportion of the road surface category.(Supplementary Note 14) The data generation method according to Supplementary Note 9, wherein a computer displays information corresponding to the road surface conditions of the walkway along which the user is walking on a screen of a mobile terminal used by the user. (Supplementary Note 15) A computer-readable, non-transitory recording medium having recorded thereon a program that causes a computer to execute the following processes: acquiring sensor data measured by a measuring device mounted on the user's footwear and position data corresponding to the position at which the sensor data was acquired, determining the road surface conditions of the walkway at the position corresponding to the position data, generating gait data using the sensor data, correcting the gait data in accordance with the road surface conditions of the walkway, and outputting the corrected gait data. (Supplementary Note 16) A computer-readable non-transitory recording medium according to Supplementary Note 15, having recorded thereon a program causing a computer to execute the following steps: determining the road surface category of the walkway at the user's position using a road surface discrimination model that outputs the road surface category of the walkway at the position of the position data in response to input of the position data; and correcting the gait data measured in response to the user's walking to the gait data in response to walking on a reference road surface using a gait data correction model that outputs the gait data in response to input of the gait data and the road surface category. (Supplementary Note 17) A computer-readable, non-transitory recording medium according to Supplementary Note 16, having recorded thereon a program causing a computer to execute the following steps: discriminating the road surface category of the walkway at the user's position using the road surface discrimination model, which is a machine learning model generated by machine learning using as training data a dataset in which the position data is an explanatory variable and the road surface category is an objective variable; and correcting the gait data measured in accordance with the user's walking to the gait data for walking on the reference road surface using the gait data correction model, which is a machine learning model trained by machine learning using as training data a dataset in which the gait data measured for each road surface category is an explanatory variable and reference gait data measured on the reference road surface is an objective variable.(Supplementary Note 18) The computer-readable, non-transitory recording medium according to Supplementary Note 16, having recorded thereon a program that causes a computer to execute the following processes: correcting the gait data measured in response to the user's walking to the gait data for walking on the reference road surface in the reference weather, using a gait data correction model that outputs the gait data for walking on the reference road surface in the reference weather, in response to input of the gait data, the road surface category, and weather data. (Supplementary Note 19) The computer-readable, non-transitory recording medium according to Supplementary Note 16, having recorded thereon a program that causes a computer to execute the following processes: determining the road surface category for each of a plurality of positions within an accuracy range of the position data; correcting the gait data in accordance with the road surface category for each of the plurality of positions within the accuracy range; multiplying each of the plurality of gait data corrected for each road surface category by a weight corresponding to the area proportion of the road surface category included in the accuracy range; and calculating, as the corrected gait data, the sum of the plurality of gait data corrected for each road surface category multiplied by the weight corresponding to the area proportion of the road surface category. (Appendix 20) A computer-readable non-transitory recording medium as described in Appendix 15, having recorded thereon a program that causes a computer to execute a process of displaying information corresponding to the road surface conditions of the walkway along which the user is walking on the screen of a mobile terminal used by the user.

[0150] Furthermore, some or all of the configurations described in Supplementary Notes 2 to 8 that are dependent on Supplementary Note 1 above may also be dependent on Supplementary Notes 9 and 15 in the same dependent relationship as Supplementary Notes 2 to 8. Furthermore, not limited to Supplementary Notes 1, 9, and 15, some or all of the configurations described as Supplements may be made dependent on various hardware, software, various recording means for recording software, or systems, within the scope of each of the above-mentioned embodiments.

[0151] 1, 2 Gait measurement system 10, 20 Measurement device 12, 22, 32 Data generation device 13, 23 Memory unit 110 Sensor 111 Acceleration sensor 112 Angular velocity sensor 113 Control unit 115 Communication unit 117 Power supply 121, 221, 321 Acquisition unit 122, 222, 322 Road surface discrimination unit 123, 223, 323 Generation unit 125, 225, 325 Correction unit 127, 227, 327 Output unit 132, 232 Road surface discrimination model 135 Gait data correction model 235 Category-specific gait data correction model 140 Road surface discrimination device 145 Road surface discrimination model

Claims

1. An acquisition unit that acquires sensor data measured by a measurement device mounted on a user's footwear and position data corresponding to the position where the sensor data was acquired; a road surface determination unit that determines the road surface condition of a walking path at a position corresponding to the position data; a generation unit that generates gait data using the sensor data; a correction unit that corrects the gait data according to the road surface condition of the walking path; and an output unit that outputs the corrected gait data. A data generation device comprising the above components.

2. The road surface determination unit uses a road surface determination model that outputs the road surface category of the walking path at the position of the position data in response to the input of the position data to determine the road surface category of the walking path at the position of the user. The generation unit uses a gait data correction model that outputs the gait data in walking on a reference road surface in response to the input of the gait data and the road surface category to correct the gait data measured according to the user's walking to the gait data in walking on the reference road surface. The data generation device according to claim 1.

3. The road surface determination model is a machine learning model generated by machine learning using, as training data, a data set having the position data as an explanatory variable and the road surface category as an objective variable. The gait data correction model is a machine learning model obtained by performing machine learning using, as training data, a data set having, for each road surface category, the gait data measured as an explanatory variable and the reference gait data measured on the reference road surface as an objective variable. The data generation device according to claim 2.

4. The road surface determination unit uses a gait data correction model that outputs the gait data in walking on the reference road surface under a reference weather condition in response to the input of the gait data, the road surface category, and weather data to correct the gait data measured according to the user's walking to the gait data in walking on the reference road surface under the reference weather condition. The data generation device according to claim 2.

5. The road surface discrimination unit discriminates the road surface category for each of a plurality of positions within the accuracy range of the position data, and the correction unit corrects the gait data according to the road surface category for each of the plurality of positions within the accuracy range, multiplies each of the plurality of gait data corrected for each road surface category by a weight according to the area ratio of the road surface category included in the accuracy range, and calculates the sum of the plurality of gait data corrected for each road surface category multiplied by the weight according to the area ratio of the road surface category as the corrected gait data. The data generation device according to claim 2.

6. The output unit causes information corresponding to the road surface state of the walking path on which the user walks to be displayed on the screen of the mobile terminal used by the user. The data generation device according to claim 1.

7. The output unit causes information optimized according to the road surface state of the walking path on which the user walks and prompting a decision-making for the user to be displayed on the screen of the mobile terminal carried by the user. The data generation device according to claim 6.

8. A gait measurement system comprising the data generation device according to any one of claims 1 to 7, and a measurement device having a sensor that measures acceleration and angular velocity, generates sensor data using the acceleration and angular velocity measured by the sensor, and transmits the generated sensor data to the data generation device.

9. A data generation method in which a computer acquires sensor data measured by a measurement device mounted on a user's footwear and position data corresponding to the position where the sensor data is acquired, discriminates the road surface state of the walking path at the position corresponding to the position data, generates gait data using the sensor data, corrects the gait data according to the road surface state of the walking path, and outputs the corrected gait data.

10. The computer discriminates the road surface category of the walking path at the position of the user by using a road surface discrimination model that outputs the road surface category of the walking path at the position of the position data in response to the input of the position data, and uses a gait data correction model that outputs the gait data in walking on a reference road surface in response to the input of the gait data and the road surface category to correct the gait data measured according to the walking of the user to the gait data in walking on the reference road surface. The data generation method according to claim 9.

11. Using the road surface discrimination model, which is a machine learning model generated by machine learning with a data set having the position data as an explanatory variable and the road surface category as an objective variable as teacher data, to discriminate the road surface category of the walking path at the position of the user, and using the gait data correction model, which is a machine learning model obtained by performing machine learning with a data set having the gait data measured for each road surface category as an explanatory variable and the reference gait data measured on the reference road surface as an objective variable as teacher data, to correct the gait data measured according to the walking of the user to the gait data in walking on the reference road surface. The data generation method according to claim 10.

12. The computer uses a gait data correction model that outputs the gait data in walking on the reference road surface under a reference weather condition in response to the input of the gait data, the road surface category, and weather data to correct the gait data measured according to the walking of the user to the gait data in walking on the reference road surface under the reference weather condition. The data generation method according to claim 10.

13. The computer discriminates the road surface category for each of a plurality of positions within the accuracy range of the position data, corrects the gait data according to the road surface category for each of the plurality of positions within the accuracy range, multiplies each of the plurality of gait data corrected for each road surface category by a weight according to the area ratio of the road surface category included in the accuracy range, and calculates the sum of the plurality of gait data corrected for each road surface category multiplied by the weight according to the area ratio of the road surface category as the corrected gait data. The data generation method according to claim 10.

14. The data generation method according to claim 9, wherein a computer causes a screen of a mobile terminal used by the user to display information corresponding to a road surface condition of the walking path on which the user walks.

15. A computer-readable non-transitory recording medium having recorded thereon a program that causes a computer to execute a process of acquiring sensor data measured by a measuring device mounted on the user's footwear and position data corresponding to the position where the sensor data is acquired, a process of determining a road surface condition of a walking path at a position corresponding to the position data, a process of generating gait data using the sensor data, a process of correcting the gait data according to the road surface condition of the walking path, and a process of outputting the corrected gait data.

16. The computer-readable non-transitory recording medium according to claim 15, having recorded thereon a program that causes a computer to execute a process of determining the road surface category of the walking path at the position of the user using a road surface discrimination model that outputs the road surface category of the walking path at the position of the position data in response to the input of the position data, and a process of correcting the gait data measured according to the walking of the user to the gait data in walking on the reference road surface using a gait data correction model that outputs the gait data in walking on the reference road surface in response to the input of the gait data and the road surface category.

17. The computer-readable non-transitory recording medium according to claim 16, having recorded thereon a program that causes a computer to execute a process of determining the road surface category of the walking path at the position of the user using the road surface discrimination model, which is a machine learning model generated by machine learning using, as teacher data, a data set having the position data as an explanatory variable and the road surface category as an objective variable, and a process of correcting the gait data measured according to the walking of the user to the gait data in walking on the reference road surface using the gait data correction model, which is a machine learning model machine-learned using, as teacher data, a data set having the gait data measured for each road surface category as an explanatory variable and the reference gait data measured on the reference road surface as an objective variable.

18. A computer-readable non-transitory recording medium having recorded thereon a program for causing a computer to execute a process of correcting the gait data measured according to the user's walking to the gait data in walking on the reference road surface in the reference weather, using a gait data correction model that outputs the gait data in walking on the reference road surface in the reference weather in response to input of the gait data, the road surface category, and the weather data.

19. A process of determining the road surface category for each of a plurality of positions within the accuracy range of the position data; a process of correcting the gait data according to the road surface category for each of the plurality of positions within the accuracy range; a process of multiplying each of the plurality of gait data corrected for each road surface category by a weight according to the area ratio of the road surface category included in the accuracy range; and a process of calculating, as the corrected gait data, the sum of the plurality of gait data corrected for each road surface category by which the weight according to the area ratio of the road surface category is multiplied. A computer-readable non-transitory recording medium according to claim 16, having recorded thereon a program for causing a computer to execute the processes.

20. A computer-readable non-transitory recording medium according to claim 15, having recorded thereon a program for causing a computer to execute a process of displaying, on a screen of a mobile terminal used by the user, information according to a road surface state of the road on which the user walks.

Images