JP5952907B2 - Walking style discrimination system, walking style discrimination device, and walking style discrimination method - Google Patents

Description

  The present invention relates to a walking style discrimination system.

  As background art in this technical field, there is JP-A-2005-114537 (Patent Document 1). This publication describes a walking motion detection processing device for detecting a human walking motion. This walking motion detection processing device detects the gravitational acceleration from the angular velocity vector and the acceleration vector, removes the acceleration in the vertical direction from the gravitational acceleration, extracts the residual acceleration component, performs the principal component analysis of the residual acceleration component, The principal component direction vector of the vertical direction acceleration component and the traveling direction acceleration component is calculated, and when the peak pair changing from the peak of the vertical direction acceleration component to the valley peak is detected based on the principal component direction vector, the traveling direction acceleration is detected. When a peak pair that changes from a valley peak of a component to a mountain peak is detected, a gradient of a traveling direction acceleration component is calculated, and the gradient at a detection time of a valley peak that changes from a mountain peak of a vertical direction acceleration component to a valley peak is predetermined. When it is equal to or greater than the value, it is determined that the walking motion is detected.

  Patent Document 1: Japanese Patent Application Laid-Open No. 2005-114537

  According to Patent Document 1, when the period of the Roll axis angular velocity component measured by the sensor worn by the subject is twice the walking cycle, it is determined that the subject is going up the stairs. This means that the Roll axis angular velocity frequency when going up the stairs is about half of the walking frequency.

  FIG. 6 is an explanatory diagram showing an example of a measured power spectrum of acceleration and angular velocity during walking.

  Specifically, FIGS. 6A and 6B are an x-axis acceleration power spectrum and a Roll-axis angular velocity power spectrum measured when the first subject is climbing the stairs, respectively. Here, the x axis is an axis perpendicular to the ground (that is, an axis parallel to the direction of gravitational acceleration), and the roll axis is an axis parallel to the component of the walking direction of the subject in a plane orthogonal to the x axis. It is. The peak frequency of the power spectrum of the x-axis acceleration corresponds to the walking frequency, and the peak frequency of the power spectrum of the Roll axis angular velocity (that is, the angular velocity of the rotational motion around the Roll axis) corresponds to the Roll axis angular velocity frequency.

  According to FIGS. 6A and 6B, while the walking frequency is about 2 Hz, the Roll axis angular velocity frequency is about 1 Hz, which is half of that. Since this result is consistent with the knowledge of Patent Document 1, it is considered that the technique of Patent Document 1 can detect the first subject's stair climbing.

  On the other hand, FIGS. 6C and 6D are an x-axis acceleration power spectrum and a roll-axis angular velocity power spectrum measured when the second subject is climbing the stairs, respectively. According to these figures, while the walking frequency is about 2 Hz, the Roll axis angular velocity frequency is about 1 Hz which is almost the same as that. FIGS. 6E and 6F are an x-axis acceleration power spectrum and a Roll-axis angular velocity power spectrum measured when the third subject is climbing the stairs, respectively. According to these figures, the walking frequency is about 2 Hz, while the Roll axis angular velocity frequency is about 3 Hz. Since these results do not agree with the knowledge of Patent Document 1, when the technique of Patent Document 1 is used, there is a possibility that the detection of the stair ascending / descending of the second subject and the third subject may fail.

  As described above, the conventional stair climbing detection technique is highly subject-dependent and the detection accuracy is not sufficient.

  Moreover, according to the technique of the above-mentioned Patent Document 1, stair climbing is detected based on the Roll axis angular velocity, but the Roll axis of the angular velocity sensor attached to the subject does not always coincide with the Roll axis of the subject. For this reason, in order to detect a test subject's stair ascending / descending by the technique of patent document 1, an angular velocity sensor is attached to a test subject so that a Roll axis of a test subject and a Roll axis of an angular velocity sensor may correspond, or it measures by an angular velocity sensor. It was necessary to calculate the angular velocity of the subject's Roll axis from the measured three-axis angular velocity.

In order to solve the above problems, the present invention is a walking style discrimination system including a terminal device attached to a pedestrian and a server device that acquires data from the terminal device, the terminal device having an angular velocity The server device includes a processor and a storage device connected to the processor, and the storage device is a three-dimensional orthogonal coordinate measured by the angular velocity sensor. System angular velocity data and acceleration data of a three-dimensional orthogonal coordinate system measured by the acceleration sensor, and the processor uses the coordinate data except for the coordinate axis having the smallest angle with the vertical direction based on the acceleration data. and was identified as 2-axis close to the traveling direction a plane perpendicular to the vertical direction, based on the angular velocity data of the two-axis closer to the traveling direction plane, said traveling A power spectrum of a biaxial angular velocity close to a plane is calculated, and based on the calculated power spectrum, a power spectrum intensity of a one-step cycle component and a power spectrum strength of a two-step cycle component are calculated; Based on whether or not the power spectrum intensity of the one-step cycle component and the power spectrum strength of the two-step cycle component satisfy a predetermined condition, the pedestrian walking mode is either stair walking or flat ground walking. or to determine it.

  According to one embodiment of the present invention, even if the angular velocity is strongly subject-dependent, the stairs are accurately performed without performing coordinate conversion based on the angular velocity measured by a sensor attached in an arbitrary posture. Elevation can be detected.

It is a block diagram which shows the structure of the walking style discrimination | determination system of embodiment of this invention. It is a block diagram which shows the hardware constitutions of each part which comprises the walking style discrimination | determination system of embodiment of this invention. It is a flowchart which shows the walk style discrimination | determination process which the server apparatus of embodiment of this invention performs. It is a flowchart which shows the flat stairs discrimination | determination process which the server apparatus of embodiment of this invention performs. It is a flowchart which shows the stair climbing determination process which the server apparatus of embodiment of this invention performs. It is explanatory drawing which shows the example of the power spectrum of the acceleration at the time of walking actually measured, and angular velocity. It is a top view of a pedestrian wearing a terminal device of an embodiment of the present invention. It is explanatory drawing of rotation with respect to the advancing direction plane by the exercise | movement of a pedestrian's leg utilized for the walking style discrimination | determination process of embodiment of this invention. It is explanatory drawing of the walking pitch range used in embodiment of this invention. It is explanatory drawing of the walk pitch intensity | strength of the pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the half pitch intensity | strength of the pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the half pitch intensity | strength of another pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the discrimination result interpolation process and discrimination result rejection process which the server apparatus of embodiment of this invention performs. It is explanatory drawing of the Z-axis angular velocity half pitch intensity | strength of the 1st pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the Y-axis angular velocity half pitch intensity | strength of the 1st pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the Z-axis angular velocity half pitch intensity | strength of the 2nd pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the Y-axis angular velocity half pitch intensity | strength of the 2nd pedestrian calculated by the server apparatus of embodiment of this invention. It is explanatory drawing of the relationship between the angular velocity half pitch intensity | strength for every frame in the staircase section in embodiment of this invention, and a walking style. It is explanatory drawing of the relationship between the average value of the angular velocity half pitch intensity | strength in the staircase section in embodiment of this invention, and a walking style.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a configuration of a walking style discrimination system according to an embodiment of the present invention. FIG. 2 is a block diagram showing a hardware configuration of each unit constituting the walking style discrimination system according to the embodiment of the present invention.

  The walking style discrimination system of this embodiment includes a terminal device 101, a server device 103, and a network 102 that connects them.

  The terminal device 101 is a terminal carried by a pedestrian, for example, a gyro sensor 201, an acceleration sensor 202, an orientation sensor 203, a temperature sensor 204, a GPS (Global Positioning System) 205, a memory 207, a storage medium 208, and a communication unit 209. , An input / output unit 210, and a CPU (Central Processing Unit) 206 connected thereto.

  The gyro sensor 201 measures the angular velocity of the terminal device 101. Specifically, the gyro sensor 201 measures the angular velocities around the X axis, the Y axis, and the Z axis with respect to the orientation of the terminal device 101.

  The acceleration sensor 202 measures the acceleration of the terminal device 101. Specifically, the acceleration sensor 202 measures the acceleration in the axial directions of the X axis, the Y axis, and the Z axis with respect to the orientation of the terminal device 101.

  The direction sensor 203 detects geomagnetism and measures the absolute direction based on it.

  The temperature sensor 204 measures the temperature inside or around the terminal device 101. When the temperature dependence of the offset amount of the gyro sensor 201 is clear, the angular velocity measured by the gyro sensor 201 may be corrected using the temperature measured by the temperature sensor 204.

  The GPS 205 receives a signal from a GPS satellite and measures the absolute position of the terminal device 101 based on the signal.

  For example, the terminal device 101 uses the GPS 205 and the direction sensor 203 in an environment where the GPS 205 and the direction sensor 203 can be used, and uses the gyro sensor 201 and the acceleration sensor 202 in an environment where the GPS 205 and the direction sensor 203 cannot be used (for example, indoors). You may collect the necessary data. As will be described later, the present invention relates to walking style discrimination using a gyro sensor 201 and an acceleration sensor 202. For this reason, although the terminal device 101 of this embodiment needs to be provided with the gyro sensor 201 and the acceleration sensor 202, the direction sensor 203, the temperature sensor 204, and GPS205 do not necessarily need to be provided.

  The CPU 206 controls the terminal device 101 according to a program stored in the memory 207.

  The memory 207 is a semiconductor memory, for example, and stores a program executed by the CPU 206, data referred to by the CPU 206, data acquired as a result of processing executed by the CPU 206, and the like. At least a part of the program and data stored in the storage medium 208 may be copied to the memory 207 as necessary, or the acquired data may be copied from the memory 207 to the storage medium 208 as necessary. Good.

  The storage medium 208 is a non-volatile storage medium such as a flash memory.

  The communication unit 209 is an interface that is connected to the network 102 and communicates with the server apparatus 103.

  The input / output unit 210 includes an input device that receives an input from a pedestrian wearing the terminal device 101 and an output device that outputs information to the pedestrian. For example, the input / output unit 210 may include a keyboard, a button, a pointing device, or the like as an input device, and may include an image display device or the like as an output device, or may include a so-called touch panel having functions equivalent to those. .

  For example, the pedestrian can input data acquisition start and end instructions to the input / output unit 210 at arbitrary timings. The CPU 206 controls the gyro sensor 201 and the acceleration sensor 202 from the start of data acquisition to the end thereof, for example, periodically measures the angular velocity and acceleration, and stores the results in the memory 207. To store. In the meantime, when the pedestrian walks on the flat ground and the stairs, the angular velocity and acceleration resulting from the walk are measured and recorded.

  The server apparatus 103 is a computer including a communication unit 211, a memory 213, a storage medium 214, an input / output unit 215, and a CPU 212 connected to them.

  The CPU 212 executes, for example, a walking style determination process described later according to a program stored in the memory 213.

  The memory 213 is, for example, a semiconductor memory, and stores a program executed by the CPU 212, data referred to by the CPU 212, data acquired as a result of processing executed by the CPU 212, and the like. At least a part of the program and data stored in the storage medium 214 may be copied to the memory 213 as necessary, or the acquired data may be copied from the memory 213 to the storage medium 214 as necessary. Good.

  The storage medium 214 is a non-volatile storage medium such as a hard disk or a flash memory.

  The communication unit 211 is an interface that is connected to the network 102 and communicates with the terminal device 101.

  The input / output unit 215 includes an input device that receives input from the user of the server device 103 and an output device that outputs information to the user. For example, the input / output unit 215 may include a keyboard, buttons, a pointing device, or the like as an input device, and may include an image display device or the like as an output device, or may include a so-called touch panel having functions equivalent to those. .

  The network 102 may be any network as long as data can be transmitted from the terminal device 101 to the server device 103 via the network 102. For example, the network 102 may be a local area network (LAN), a wide area network, or a combination thereof. Alternatively, the network 102 may be a data transmission path such as a USB (Universal Serial Bus).

  1 and 2 show a configuration in which the terminal device 101 and the server device 103 communicate data via the network 102 as a typical example of the walking route estimation system. Instead of communication, data can be exchanged using a storage medium. For example, data measured by the gyro sensor 201 or the like is written in a replaceable storage medium such as a memory card connected to the terminal apparatus 101, and the storage medium is transported to the installation location of the server apparatus 103 to be server apparatus. The data written may be copied to the storage medium 214 or the memory 213.

  Alternatively, instead of the server device 103, the terminal device 101 itself (more specifically, the CPU 206 of the terminal device 101 may perform walking mode determination described later according to a program stored in the memory 207). In that case, the network 102 and the server apparatus 103 are unnecessary.

  FIG. 3 is a flowchart illustrating the walking style determination process executed by the server apparatus 103 according to the embodiment of this invention.

  As described above, the terminal apparatus 101 stores the angular velocity data and acceleration data measured by the gyro sensor 201 and the acceleration sensor 202 in the memory 207 or the storage medium 208 and transmits them to the server apparatus 103 via the network 102. To do. For example, a pedestrian wearing the terminal device 101 walks in a space including stairs, and angular velocity data and acceleration data acquired during that time are stored in the storage medium 208 or the like. After the pedestrian finishes walking, the terminal device 101 transmits data stored in the storage medium 208 or the like to the server device 103. The server device 103 stores the angular velocity data and the acceleration data received from the terminal device 101 in the storage medium 214 and executes the walking style discrimination process shown in FIG. That is, at the time when the processing shown in FIG. 3 is started, the server apparatus 103 at least acquires the time-series data of the X-axis, Y-axis, and Z-axis accelerations acquired from the terminal apparatus 101, and the X-axis and Y-axis. And time-series data of angular velocities around the respective Z-axis.

  Here, the X axis, the Y axis, and the Z axis are coordinate axes of angular velocity and acceleration measured by the gyro sensor 201 and the acceleration sensor 202 included in the terminal device 101. The relationship between these three axes, the pedestrian's Yaw axis (ie, the vertical axis), the pedestrian's Roll axis (ie, the longitudinal axis), and the pedestrian's Pitch axis (ie, the horizontal axis) It depends on the direction in which the gyro sensor 201 and the acceleration sensor 202 are installed in the device 101 and the direction in which the terminal device 101 is worn by a pedestrian. Since neither direction is limited in this embodiment, the coordinate axis of the sensor does not normally coincide with a coordinate axis such as a Yaw axis based on a pedestrian (see FIG. 7).

  First, the server device 103 executes a walking discrimination parameter calculation (step 302) and a power spectrum calculation (step 303) for each frame (described later) (loop 301). The walking determination parameter calculated in step 302 is a parameter used in determining whether or not a pedestrian is walking at each time (that is, walking determination in step 306 described later). Since the walking discrimination can be executed by any known method such as an autocorrelation method, the detailed method of walking discrimination and the parameters used therefor are omitted. In step 303, the server apparatus 103 calculates a power spectrum of angular velocities around at least the X axis, the Y axis, and the Z axis. Furthermore, if necessary for walking determination or the like, the server device 103 may calculate the power spectrum of the acceleration of each axis.

  The server device may calculate the power spectrum of the angular velocities around all the X-axis, Y-axis, and Z-axis as described above. When two close axes (see FIG. 7A) are specified, only the power spectrum of angular velocity around each of these two axes (for example, the Y axis and the Z axis) may be calculated.

  Here, the frame is time-series data divided into a predetermined length (for example, a predetermined number of samples or a predetermined time). For example, the server apparatus 103 calculates the power spectrum of the angular velocity around each axis for the first frame composed of N samples at time t = 1 to N, and calculates the calculated value at each time from 1 to N. Stored as power spectrum value. Next, the server apparatus 103 performs the same calculation as described above for the second frame including N samples at times N + 1 to 2N, and holds the result as the value of the power spectrum at each time of N + 1 to 2N. . The server apparatus 103 repeatedly executes the same processing for each frame until the calculation is completed for all the samples of the acquired data (loop 301).

  In the above example, when the first frame includes N samples at time t = 1 to N, the second frame is not N samples at time N + 1 to 2N but time 1 + M to N + M. N samples may be included (M <N, for example, M = N / 2). As a result, a plurality of power spectrum values may be calculated for one time. In this case, the average value of the plurality of power spectrum values is held as the power spectrum value at that time.

  When the loop 301 ends, the server apparatus 103 next performs normalization processing on the angular velocity data and the power spectrum calculated in the loop 301 (step 304). As shown in FIG. 6, there are large individual differences in the power spectrum of acceleration and angular velocity during walking. One of the purposes of the normalization processing in the present embodiment is to reduce subject dependency (that is, pedestrian dependency) of acceleration data and angular velocity data.

  A general normalization method for this purpose includes normalization based on an average value and a variance value. Although such normalization processing may be performed in the present embodiment, here, in addition to the above-described purpose, the quartile value is used for another purpose of reducing the influence of outliers. Perform normalization. Here, an example of outliers of acceleration data and angular velocity data during walking is angular velocity data when a pedestrian crawls during walking. Since such data does not reflect the walking style, it cannot be used for discrimination of the walking style, but it is not desirable to use it for normalization because the average value varies greatly.

と定義される。Specifically, for example, when normalizing angular velocity data around the X axis, the server apparatus 103 rearranges all of the held angular velocity data around the X axis in the order of the magnitudes of the values. For example, when n pieces of angular velocity data x around the X axis are rearranged as x ₁ ≦ x ₂ ≦ ... ≦ x _n , for example, q quantiles Q _q such as quartile Q _1/4 Is

Is defined.

によって正規化される。ここで、ｘ_ｉは、各時刻における角速度の値である。Using the quartile values as above, the angular velocity is

Normalized by Here, x _i is the value of the angular velocity at each time.

によって正規化される。ここで、ｘ_ｉは、各時刻における角速度のパワースペクトルの値である。On the other hand, the power spectrum is

Normalized by Here, x _i is the value of the power spectrum of the angular velocity at each time.

  At least one of the time-series angular velocity data measured by the gyro sensor 201, the power spectrum intensity of the angular velocity, or the time-series data of the average value of the power spectrum intensity of the angular velocity in a predetermined frequency range to be described later is as described above. By normalizing the data using the quartile values, it is possible to reduce variations in data depending on pedestrians while reducing the influence of outliers.

  Next, the server apparatus 103 executes steps 306 to 308 for each sample (loop 305).

  Specifically, the server device 103 executes walking determination based on the sample value at each time (step 306), and when it is determined that the pedestrian is walking (step 307), the step determination of the flat stairs is performed. Execute (step 308). The server apparatus 103 repeats the above processing until the processing is completed for the sample values at all times (loop 305). Here, walking discrimination is discrimination of whether or not a pedestrian is walking. Since the walking determination can be executed by any known method, the description is omitted. The flat stairs discrimination is discrimination of whether a pedestrian is walking on a flat ground or a stairs, and the detailed procedure will be described with reference to FIG.

  FIG. 4 is a flowchart illustrating the flat land staircase determination process executed by the server apparatus 103 according to the embodiment of this invention.

  When it is determined that the pedestrian is walking at that time as a result of the walking determination process based on the sample value at a certain time (step 306), the server device 103 determines that time (hereinafter, in the description of FIG. “Flat ground” is set as the initial value of the walking style (described) (step 401).

  Next, the server apparatus 103 determines whether or not the sample value at the time satisfies each of a plurality of preset conditions (step 403). The server apparatus 103 repeats step 403 until it is determined that the sample value at the time satisfies any condition or the determination for all conditions is completed (loop 402).

  When it is determined in step 403 that one of the conditions is satisfied, the server apparatus 103 determines that the walking style at the time is “stairs” (that is, determines that the pedestrian was walking on the stairs at the time). (Step 404). On the other hand, if it is not determined in step 403 that any of the conditions is satisfied, the server apparatus 103 determines that the walking style at the time is “flat” as set in step 401.

  Here, details of the determination in step 403 will be described.

  FIG. 7A is a top view of a pedestrian 701 wearing the terminal device 101 according to the embodiment of this invention.

  FIG. 7B is an explanatory diagram of the rotation with respect to the traveling direction plane due to the movement of the legs of the pedestrian 701, which is used for the walking style discrimination process according to the embodiment of the present invention.

  The X axis, the Y axis, and the Z axis that are orthogonal to each other of the terminal device 101 illustrated in FIG. 7A are coordinate axes of the gyro sensor 201 and the acceleration sensor 202 of the terminal device 101. That is, the gyro sensor 201 detects angular velocities around the X axis, Y axis, and Z axis shown in FIG. 7A, and the acceleration sensor 202 accelerates in the directions of the X axis, Y axis, and Z axis shown in FIG. 7A. Is detected. In the present embodiment, since the position and posture of the terminal device 101 attached to the pedestrian 701 are not limited, these three axes usually do not coincide with any of the Roll axis, Pitch axis, and Yaw axis of the pedestrian 701 (see FIG. In the example of 7A, it seems that the Yaw axis and the X axis almost coincide with each other, but generally they also do not coincide).

  Here, a plane (that is, a plane parallel to the horizontal plane) perpendicular to the Yaw axis (the vertical direction of the pedestrian 701, in other words, the vertical coordinate axis) is referred to as a traveling direction plane. In the example of FIG. 7A, the X axis is the closest to the Yaw axis among the X axis, the Y axis, and the Z axis (that is, the X axis and the Yaw axis out of the angles formed by the X axis, the Y axis, and the Z axis with the Yaw axis). Therefore, the two axes excluding the X axis, that is, the Y axis and the Z axis are two axes close to the traveling direction plane. In this case, the angular velocities about the Y axis and the Z axis represent rotation with respect to the plane of travel. As the rotation with respect to the traveling direction plane, the rotation 702 caused by the movement of the legs accompanying walking is dominant (see FIG. 7B).

  Such a rotational motion includes a component having a cycle corresponding to one step (walking pitch) and a component having a cycle corresponding to two steps (half pitch). For this reason, the power spectrum intensity (hereinafter referred to as walking pitch intensity) of the walking pitch of the angular velocity around two axes (Y axis and Z axis) close to the traveling direction plane and the power spectrum intensity of the half pitch (hereinafter referred to as half pitch intensity and Based on the description, it is possible to capture the characteristics of the walking mode.

  Since the Yaw axis is parallel to the direction of gravitational acceleration, of the three axes of the terminal device 101, the axis with the maximum absolute value of the acceleration amplitude (X axis in the example of FIG. 7A) is closest to the Yaw axis. It can be specified that the other two axes (Y axis and Z axis in the example of FIG. 7A) are two axes close to the traveling direction plane. The server apparatus 103 executes a procedure for specifying two axes close to the traveling direction plane at any time before the determination in step 403 is performed.

  In this embodiment, the statistic of the power spectrum in the frequency range (walking pitch range) corresponding to the walking pitch is used as the walking pitch intensity, and the power spectrum statistic in the frequency range (half pitch range) corresponding to the half pitch is used. Used as half pitch strength. In the present embodiment, an average value is employed as a statistic, but other statistic such as a maximum value may be employed.

  As described above, walking discrimination can be performed by a known technique such as an autocorrelation method. The same known technique can be used for detecting the walking pitch, but such a method erroneously detects a pitch different from the original walking pitch (for example, a double pitch or a half pitch) as the walking pitch. This reduces the accuracy of walking style discrimination. For this reason, in this embodiment, the frequency range of the walking pitch range and the half pitch range is determined in advance, and the average value of the power spectrum in the range is used as the walking pitch strength and the half pitch strength.

  FIG. 8 is an explanatory diagram of the walking pitch range used in the embodiment of the present invention.

  Specifically, FIG. 8 shows an example of the result of actually measuring the pace of a pedestrian (ie, the number of steps per second) (edited by Masahiko Sato, “Japanese Dictionary”, Asakura Shoten, June 2003). . The age of pedestrians is classified into adolescents, middle-aged and elderly, and actual results for men and women of each age are shown. According to this result, the pace becomes slower (that is, the value is smaller) as the age increases, and the male pace tends to be slower than the female pace, but the overall distribution is approximately 1.5 to 2.5 steps / It can be seen that it is within the range of seconds, and the center is approximately 2.0 steps / second.

  On the basis of this, in this embodiment, the average value of the power spectrum in the range of 1.5 to 2.5 Hz is the walking pitch intensity, and the average value of the power spectrum in the range of 0.75 to 1.25 Hz is half of that. Used as pitch strength. Thereby, the walking pitch intensity and the half pitch intensity can be calculated without depending on the extraction accuracy of the walking pitch by the autocorrelation method or the like.

  Note that the walking pitch range shown in FIG. 8 is an example, and other walking pitch ranges may be used. For example, when the attributes such as the age, sex, and assigned duties of the pedestrian are limited and the walking pitch range can be limited according to the attributes, the walking pitch range limited as such can be used.

  Examples of walking pitch intensity and half pitch intensity actually calculated as described above are shown in FIGS. 9A to 9C. The determination in step 404 will be described with reference to this example.

  FIG. 9A is an explanatory diagram of the walking pitch intensity of the pedestrian calculated by the server device 103 according to the embodiment of this invention.

  The horizontal axis of the graph shown in FIG. 9A is time, and the vertical axis is Y-axis angular velocity walking pitch intensity. In this example, times 902, 904, 906, 908 and 910 correspond to the time when the pedestrian was walking on the flat ground, and times 901, 903 and 905 correspond to the time when the pedestrian was down the stairs. , 907, 909 and 911 correspond to the time when the pedestrian was climbing the stairs.

  If such an angular velocity walking pitch intensity is obtained, it cannot be determined whether the pedestrian was climbing up or down the stairs based on this, but whether the pedestrian was walking on the stairs or flat ground Can be determined. For example, the server device 103 sets the walking pitch strength value “0” as a threshold value, and when the walking pitch strength value at each time does not exceed the threshold value, the condition in step 403 is satisfied, and the walking in step 404 is performed. It can be determined that the person was walking on the stairs.

  FIG. 9B is an explanatory diagram of the pedestrian half-pitch strength calculated by the server device 103 according to the embodiment of this invention.

  The horizontal axis of the graph shown in FIG. 9B is time, and the vertical axis is the Z-axis angular velocity half pitch intensity. In this example, times 922, 924, 926, 928, and 930 correspond to the time that the pedestrian was walking on the flat ground, and times 921, 923, and 925 correspond to the time that the pedestrian was down the stairs. , 927, 929 and 931 correspond to the time when the pedestrian was climbing the stairs.

  If such an angular velocity half-pitch strength is obtained, it may be possible to determine whether the pedestrian is climbing up or down the stairs based on this. Determine if you were walking. For example, the server apparatus 103 sets the half pitch intensity values “4” and “−2” as threshold values, and the half pitch intensity value at each time exceeds “4” or falls below “−2”. In this case, it can be determined that the condition of step 403 is satisfied and the pedestrian is walking on the stairs (step 404).

  FIG. 9C is an explanatory diagram of the half pitch intensity of another pedestrian calculated by the server device 103 according to the embodiment of this invention.

  That is, FIG. 9C shows the half pitch intensity calculated based on the angular velocity data acquired from the terminal device 101 worn by a pedestrian different from FIG. 9B.

  The horizontal axis of the graph shown in FIG. 9C is time, and the vertical axis is the Z-axis angular velocity half pitch intensity. In this example, times 942, 944, 946, 948, and 950 correspond to the time that the pedestrian was walking on the flat ground, and times 941, 943, and 945 correspond to the time that the pedestrian was down the stairs. , 947, 949, and 951 correspond to the time when the pedestrian was climbing the stairs.

  When such an angular velocity half-pitch strength is obtained, for example, if the half-pitch strength value “5” is set as a threshold, the server apparatus 103 causes the pedestrian to walk the stairs at times 947, 949, and 951. Can be determined. However, in the example of FIG. 9C, the difference between the half pitch intensity when the pedestrian is walking down the stairs and the half pitch intensity when the pedestrian is walking on the flat ground is small, and therefore, in steps 941, 943 and 945, step 403 is performed. This condition may not be satisfied, and it may not be possible to determine that the pedestrian is walking on the stairs.

  Depending on the manner in which the pedestrian walks and the attitude of the terminal device 101, the pedestrian walks the stairs even though the pedestrian is walking the stairs, for example, at times 941, 943, and 945 in FIG. 9C. There are cases where the conditions set for determining that the user has done are not satisfied. However, as described above, a plurality of conditions are set, and when one of them is satisfied, it is determined that the pedestrian is walking on the stairs (loop 402), so that there is no leakage and high accuracy determination. It can be performed.

  In the above example, the threshold value used for the determination in step 403 may be manually set by the user of the walking style determination system based on the acquired walking pitch strength and half pitch strength. May be automatically set by the server apparatus 103.

  9A to 9C show only the Y-axis angular velocity pitch strength and the Z-axis angular velocity half-pitch strength, but in actuality, the Y-axis and Z-axis are two axes close to the traveling direction plane as shown in FIG. 7A. The Z-axis angular velocity pitch intensity and the Y-axis angular velocity half-pitch intensity are also calculated. If any of them satisfies a predetermined condition in step 403, the pedestrian is walking on the stairs in step 404. It is determined that

  When the loop 305 in FIG. 3 ends, the server apparatus 103 next executes a discrimination result interpolation process (step 309) and a discrimination result rejection process (step 310).

  FIG. 10 is an explanatory diagram of the discrimination result interpolation process and the discrimination result rejection process executed by the server apparatus 103 according to the embodiment of this invention.

  As shown in FIGS. 9A to 9C, the measured power spectrum intensity varies even when the pedestrian is walking on either the flat ground or the stairs. For this reason, as a result of discrimination based on the threshold value, when it is determined that a very short flat-ground walk is performed after a continuous stair walk and then a stair walk is performed again, and vice versa, It may be determined that a very short staircase walk was performed later, and then a flat ground walk was performed again.

  However, in reality, both flat ground walking and stair walking are performed continuously for a certain period of time, and the above-mentioned extremely short flat ground walking and stair walking are not the result of actual walking. This is considered to be caused by data variation. For this reason, the server apparatus 103 performs the discrimination result interpolation process (step 309) for discriminating the short-time walking on the flat ground as described above as the stair walking and the discrimination result rejection process (step 310) for discriminating the short-time stair walking as the flat ground walking. ).

  For example, as shown in FIG. 10, when the discrimination result at the time 1002 between the times 1001 and 1003 when the discrimination result is a staircase walk is a flat ground walk and the time 1002 is shorter than a predetermined threshold, The server apparatus 103 changes the determination result at time 1002 to stair walking (step 309). As a result, the discrimination result at time 1004 corresponding to times 1001 to 1003 is stair walking.

  On the other hand, when the determination result at the time 1006 sandwiched between the times 1005 and 1007 when the determination result is walking on a flat ground is the staircase walking, and the time 1006 is shorter than a predetermined threshold, the server apparatus 103 determines that the time 1006 The discrimination result in is changed to flat ground walking (step 310). As a result, the discrimination result at time 1008 corresponding to times 1005 to 1007 is walking on a flat ground.

  It is considered that the discrimination result of the walking style more accurately reflects the actual walking style by the discrimination result interpolation process and the discrimination result rejection process.

  Next, the server device 103 executes a stair climbing determination process (step 312) for each stair section (loop 311). Here, the stair section is a time determined as a staircase between the times determined to be flat, and for example, a time 1004 in FIG. 10 is a stair section. The server apparatus 103 repeatedly executes the stair climbing determination process (step 312) until the process is completed for all the stair sections (loop 311). The stair climbing determination is a determination of whether or not the pedestrian is going up or down the stairs, and the detailed procedure will be described with reference to FIG.

  FIG. 5 is a flowchart illustrating stair climbing determination processing executed by the server apparatus 103 according to the embodiment of this invention.

  First, the server apparatus 103 sets “flat ground” as the initial value of the walking style of the stair section to be processed (hereinafter referred to as the section in the description of FIG. 5). (Step 501).

  Next, the server apparatus 103 determines whether or not the half pitch intensity of the section satisfies each of a plurality of conditions set in advance in order to determine that the walking style is “up” (step 503). If it is determined in step 503 that the condition is not satisfied, the server apparatus 103 determines whether the half pitch intensity of the section satisfies each of a plurality of conditions set in advance to determine the walking style as “down”. Is determined (step 505). The server apparatus 103 repeats steps 503 and 505 until it is determined that the half pitch intensity of the section satisfies any one of the conditions or the determination for all the conditions is completed (loop 502).

  When it is determined in step 503 that any of the conditions is satisfied, the server apparatus 103 determines that the walking style in the section is “up” (that is, determines that the pedestrian has climbed the stairs in the section). (Step 504). When it is determined in step 505 that any of the conditions is satisfied, the server apparatus 103 determines that the walking style of the section is “down” (that is, determines that the pedestrian is down the stairs in the section). (Step 506). On the other hand, if it is not determined in step 503 and 505 that both conditions are satisfied, the server apparatus 103 determines the walking style of the section as “flat” as set in step 501.

  As described above, the target of the stair climbing determination process is a stair section where it is determined that the pedestrian is walking up the stairs, but a plurality of conditions for determining that the pedestrian is up the stairs, and walking If none of the multiple conditions for determining that the person was walking down the stairs was satisfied, it was determined that the determination that the pedestrian was walking on the stairs was incorrect, and the determination result of the walking style Is changed to “flat” (steps 501, 503, and 505). Note that such determination is an example. For example, when neither of the conditions is satisfied in Steps 503 and 505, the walking mode is determined to be one of “up” or “down” which is predetermined. May be.

  Here, details of the determination in steps 503 and 505 will be described. Specifically, the example of the half pitch intensity of two pedestrians shown in FIGS. 11A to 11D is referred to.

  FIG. 11A is an explanatory diagram of the Z-axis angular velocity half pitch intensity of the first pedestrian calculated by the server device 103 according to the embodiment of this invention.

  FIG. 11B is an explanatory diagram of the Y-axis angular velocity half pitch intensity of the first pedestrian calculated by the server device 103 according to the embodiment of this invention.

  Threshold values 1111 and 1112 in FIG. 11A are threshold values for determining the walking style “up” and “down”, respectively. In step 503, the server apparatus 103 determines that the condition is satisfied when the value of the Z-axis angular velocity half pitch intensity exceeds the threshold 1111. In step 505, the server apparatus 103 determines that the value of the Z-axis angular velocity half pitch intensity exceeds the threshold 1112. It is determined that the condition is satisfied.

  Such a determination can also be made for the value of the Z-axis angular velocity half pitch intensity for each time or frame in the staircase section, but in this embodiment, the average value of the half pitch intensity in the staircase section is compared with the threshold value. Is done. The same applies to FIGS. 11B to 11D described later. The reason will be described later with reference to FIGS. 12A and 12B.

  In FIG. 11A, for example, the average value of the Z-axis angular velocity half pitch intensity in the staircase section 1101 is smaller than the threshold 1111 and larger than the threshold 1112. For this reason, in this example, it is determined that neither of the conditions in steps 503 and 505 is satisfied. On the other hand, since the average value of the Z-axis angular velocity half pitch intensity in the staircase section 1102 is larger than the threshold value 1111, it is determined that the condition of step 503 is satisfied, and step 504 is executed, and the first pedestrian walking in the staircase section 1102 It is determined that the style is “up” (that is, the first pedestrian has climbed the stairs in the stair section 1102).

  FIG. 11B shows the Y-axis angular velocity half-pitch intensity of the first pedestrian in the same time zone as shown in FIG. 11A. Therefore, the staircase sections 1101 and 1102 in FIG. 11B are the same as those shown in FIG. 11A. The threshold values 1113 and 1114 are threshold values for discriminating the walking style “up” and “down”, respectively.

  In FIG. 11B, the average value of the Y-axis angular velocity half pitch intensity in the staircase section 1101 is smaller than the threshold value 1114. Therefore, in this example, it is determined that the condition of step 505 is satisfied, step 506 is executed, and the walking style of the first pedestrian in the staircase section 1101 is “down” (that is, the first staircase section 1101 has the first Pedestrians were walking down the stairs).

  As described with reference to FIG. 5, when any one of the plurality of conditions is satisfied, the walking mode is determined as “up” or “down”. In the example of FIGS. 11A and 11B, the condition that “the average value of the Z-axis angular velocity half-pitch strength is below the threshold 1112” is not satisfied for the staircase section 1101, but “the average value of the Y-axis angular velocity half-pitch strength is Since the condition “below the threshold 1114” is satisfied, it is determined that the walking style in the staircase section 1101 is “down”.

  FIG. 11C is an explanatory diagram of the Z-axis angular velocity half pitch intensity of the second pedestrian calculated by the server device 103 according to the embodiment of this invention.

  FIG. 11D is an explanatory diagram of the second pedestrian's Y-axis angular velocity half-pitch intensity calculated by the server device 103 according to the embodiment of this invention.

  The thresholds 1111 and 1112 in FIG. 11C are the same as those shown in FIG. 11A. In FIG. 11C, for example, the average value of the Z-axis angular velocity half pitch intensity in the staircase section 1103 is smaller than the threshold value 1112. Therefore, it is determined that the condition of step 505 is satisfied, step 506 is executed, and it is determined that the walking style of the second pedestrian in the staircase section 1103 is “down”. On the other hand, since the average value of the Z-axis angular velocity half pitch intensity in the staircase section 1104 is larger than the threshold value 1111, it is determined that the condition in step 503 is satisfied, and step 504 is executed, whereby the second pedestrian walking in the staircase section 1104 It is determined that the style is “up”.

  FIG. 11D shows the Y-axis angular velocity half pitch intensity of the second pedestrian in the same time zone as shown in FIG. 11C. Therefore, the staircase sections 1103 and 1104 in FIG. 11D are the same as those shown in FIG. 11C. The threshold values 1113 and 1114 are threshold values for determining the walking style “up” and “down”, respectively, and are the same as those shown in FIG. 11B.

  In FIG. 11D, the average value of the Z-axis angular velocity half pitch intensity in the staircase section 1103 is smaller than the threshold value 1113 and larger than the threshold value 1114. For this reason, in this example, it is determined that neither of the conditions in steps 503 and 505 is satisfied. Similarly, the average value of the Y-axis angular velocity half pitch intensity in the staircase section 1104 is smaller than the threshold value 1113 and larger than the threshold value 1114. For this reason, it is determined that neither of the conditions in steps 503 and 505 is satisfied for this example.

  In the example of FIGS. 11C and 11D, the condition that “the average value of the Y-axis angular velocity half-pitch strength is below the threshold value 1114” is not satisfied for the staircase section 1103, but “the average value of the Z-axis angular velocity half-pitch strength is Since the condition “below threshold 1112” is satisfied, it is determined that the walking style in the staircase section 1103 is “down”. For the staircase section 1104, the condition that “the average value of the Y-axis angular velocity half-pitch strength exceeds the threshold value 1113” was not satisfied, but the condition that “the average value of the Z-axis angular velocity half-pitch strength exceeds the threshold value 1111” was satisfied. Therefore, it is determined that the walking style in the staircase section 1104 is “up”.

  As described above, by setting multiple conditions for discriminating the walking style, if any one of the conditions is met, the walking style is discriminated as "up" or "down" Even when there are variations in measured values of angular velocity due to the way of walking and the posture of the terminal device 101 worn by a pedestrian, the walking mode can be accurately determined.

  FIG. 12A is an explanatory diagram of the relationship between the angular velocity half pitch intensity for each frame in the staircase section and the walking style in the embodiment of the present invention.

  Specifically, the graph shown in FIG. 12A is a scatter diagram with the horizontal axis representing the Y-axis angular velocity power spectrum half pitch intensity and the vertical axis representing the Z-axis angular velocity power spectrum half pitch intensity. The circular and triangular symbols plotted are the values calculated for each frame based on the angular velocity data measured when a pedestrian is actually going up and down the stairs, respectively. .

  According to FIG. 12A, although the tendency of the half pitch intensity when a pedestrian is going up and down is different, it is not clearly separated. This is considered to be because the half pitch intensity at each time in the staircase section varies as shown in FIGS. 11A to 11D, for example. Therefore, when the walking style is determined based on these values, it is inevitable that erroneous determination will occur regardless of what threshold value is set.

  FIG. 12B is an explanatory diagram of the relationship between the average value of the angular velocity half pitch intensity in the staircase section and the walking style in the embodiment of the present invention.

  Specifically, the horizontal axis and the vertical axis of the graph shown in FIG. 12B are the same as those in FIG. 12A, and the values calculated for each staircase section based on the same angular velocity data as in FIG. 12A are plotted. Similar to FIG. 12A, the circular and triangular symbols correspond to up and down stairs, respectively.

  According to FIG. 12B, the half pitch intensity when the pedestrian is climbing up and down is clearly separated. This is considered to be because the influence of the variation in the half pitch intensity at each time in the stair section is suppressed by obtaining the average value. Therefore, if an appropriate threshold value is set, the walking style can be accurately determined based on these values.

  When the stair climbing determination process (step 312) is completed for all the stair sections, the server apparatus 103 ends the walking style determination process (FIG. 3). The server device 103 stores, in the memory 213 or the storage medium 214, the walking mode determination result, that is, the determination result of whether the pedestrian was walking on the flat ground at each time, whether the stairs were up, or down the stairs. And can be output from the input / output unit 215 as necessary.

  According to the above embodiment of the present invention, since the walking style is determined based on the biaxial angular velocities in the traveling direction plane, it is not necessary to limit the posture of the terminal device 101 attached to the pedestrian, and is measured. There is no need to perform coordinate conversion of the data. Further, according to the embodiment of the present invention, it is first determined whether the walking style is “flat” or “stairs” based on the walking pitch intensity and the half pitch intensity of the power spectrum of the angular velocity. At this time, a predetermined frequency range is used as the walking pitch. The angular velocity power spectrum data is normalized using a quartile value. For sections where the walking style is determined to be “stairs”, interpolation and rejection are performed based on the length of the section, and for each stored staircase section, the walking style is determined based on the half-pitch strength of angular velocity. Whether it is “up” or “down” is determined. A plurality of conditions for determining each of “stairs”, “up”, and “down” are set as walking styles, and when one of them is satisfied, “stair”, “up” or It is determined as “down”. As a result, even when there is a variation in angular velocity data due to the manner in which each pedestrian walks, the posture of the terminal device 101, or other factors, the walking mode can be accurately determined.

  Information such as a program, a table, and a file that realize each function of the above embodiment is a storage device such as a nonvolatile semiconductor memory, a hard disk drive, an SSD (Solid State Drive), or an IC card, an SD card, a DVD, or the like It can be stored on a computer readable non-transitory data storage medium.

  The present invention is not limited to the embodiments described above, and includes various modifications. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to one having all the configurations described. Further, a part of the configuration of an embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of an embodiment. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

Claims (10)

A walking style discrimination system including a terminal device attached to a pedestrian and a server device that acquires data from the terminal device,
  The terminal device includes an angular velocity sensor and an acceleration sensor,
  The server device includes a processor and a storage device connected to the processor,
  The storage device holds the angular velocity data of the three-dimensional orthogonal coordinate system measured by the angular velocity sensor and the acceleration data of the three-dimensional orthogonal coordinate system measured by the acceleration sensor,
  The processor is
  Based on the acceleration data, the two coordinate axes excluding the coordinate axis having the smallest angle with the vertical direction are specified as two axes close to the traveling direction plane orthogonal to the vertical direction,
  Based on the biaxial angular velocity data close to the traveling direction plane, the power spectrum of the biaxial angular velocity close to the traveling direction plane is calculated,
  Based on the calculated power spectrum, calculate the power spectrum intensity of the component of the cycle of one step and the power spectrum intensity of the component of the cycle of two steps,
  Based on whether or not the power spectrum intensity of the one-step cycle component and the power spectrum strength of the two-step cycle component satisfy a predetermined condition, the pedestrian walking mode is either stair walking or flat ground walking. A walking style discriminating system characterized by discriminating whether it is or not. The walking style discrimination system according to claim 1,
  The processor, based on whether or not the power spectrum intensity of the component of the cycle of the two steps satisfies a predetermined condition for the section in which the walking mode of the pedestrian is determined to be staircase walking, A walking style discriminating system characterized by discriminating whether a walking style is ascending walking or descending staircase. The walking style discrimination system according to claim 2,
  The predetermined condition used for determining whether the walking style of the pedestrian is stair walking or flat ground walking is a plurality of conditions for determining that the walking style of the pedestrian is stair walking. Including
  Each of the plurality of conditions for determining that the walking mode of the pedestrian is stair walking is that the power spectrum intensity of the component of the cycle of one step and the two steps of each of the two axes close to the traveling direction plane A condition specified by the relationship between each of the power spectrum intensities of the components of the period and a predetermined threshold,
  The processor determines that the pedestrian's walking style is staircase walking when at least one of a plurality of conditions for determining that the pedestrian's walking style is staircase walking is satisfied,
  The predetermined condition used to determine whether the walking style of the pedestrian is uphill walking or downhill walking is a plurality of conditions for determining that the walking style of the pedestrian is uphill walking. Including a plurality of conditions and a plurality of conditions for discriminating that it is a descending walk,
  Each of the plurality of conditions for determining that the walking style of the pedestrian is an ascending walk, and each of the plurality of conditions for determining that the walking style of the pedestrian is a descending walking are on the traveling direction plane. A condition specified by the relationship between the power spectrum intensity of the component of the cycle of two steps and a predetermined threshold value for each of the two nearby axes;
  The processor is
  When at least one of a plurality of conditions for determining that the pedestrian's walking style is ascending walking is satisfied, it is determined that the pedestrian's walking style is ascending walking;
  The walking mode is characterized in that when at least one of a plurality of conditions for determining that the walking mode of the pedestrian is descending walking is satisfied, the walking mode of the pedestrian is determined to be descending walking. Discriminating system. The walking style discrimination system according to claim 2,
  The processor is based on whether the average value of the power spectrum intensity of the component of the cycle of the two steps in each section in which the walking mode of the pedestrian is determined to be stair walking satisfies a predetermined condition, A walking style discrimination system for discriminating whether the walking style of a pedestrian in each section is an ascending or descending staircase. The walking style discrimination system according to claim 1,
  The processor calculates a statistic of power spectrum intensity in a first predetermined frequency range as a power spectrum intensity of a component of the one-step cycle, and corresponds to a half frequency corresponding to a half of the first predetermined frequency range. 2. A walking style discrimination system, wherein a statistic of power spectrum intensity in a predetermined frequency range of 2 is calculated as a power spectrum intensity of a component of the cycle of two steps. The walking style discrimination system according to claim 1,
  The processor is sandwiched between two sections in which the walking style is determined to be stair walking, and when the length of the section in which the walking style is determined to be flat ground walking is shorter than a predetermined value, the walking style is The walking style of the section determined to be flat ground walking is changed to stair walking, and the walking style is sandwiched between two sections determined to be flat ground walking and the walking style is determined to be stair walking When the length of the walking is shorter than a predetermined value, the walking style discriminating system is characterized in that the walking style of the section in which the walking style is determined to be staircase walking is changed to flat ground walking. The walking style discrimination system according to claim 2,
  The processor is
  Time-series data of the biaxial angular velocity close to the traveling direction plane, the power spectrum of the biaxial angular velocity close to the traveling direction plane, the power spectrum intensity of the component of one step period calculated based on the power spectrum, and 2 Normalize at least one of the power spectrum intensities of the components of the step cycle based on the quartile;
  A walking style discrimination system that discriminates the walking style of the pedestrian based on the normalized value. A walking style discrimination device having a processor and a storage device connected to the processor,
  The storage device stores three-dimensional orthogonal coordinate system angular velocity data measured by an angular velocity sensor attached to a pedestrian and three-dimensional orthogonal coordinate system acceleration data measured by an acceleration sensor attached to the pedestrian. Hold and
  The processor is
  Based on the acceleration data, the two coordinate axes excluding the coordinate axis having the smallest angle with the vertical direction are specified as two axes close to the traveling direction plane orthogonal to the vertical direction,
  Based on the biaxial angular velocity data close to the traveling direction plane, the power spectrum of the biaxial angular velocity close to the traveling direction plane is calculated,
  Based on the calculated power spectrum, calculate the power spectrum intensity of the component of the cycle of one step and the power spectrum intensity of the component of the cycle of two steps,
  Based on whether or not the power spectrum intensity of the one-step cycle component and the power spectrum strength of the two-step cycle component satisfy a predetermined condition, the pedestrian walking mode is either stair walking or flat ground walking. A walking style discriminating device characterized by discriminating whether or not. The walking style discrimination device according to claim 8,
  The processor, based on whether or not the power spectrum intensity of the component of the cycle of the two steps satisfies a predetermined condition for the section in which the walking mode of the pedestrian is determined to be staircase walking, A walking style discriminating apparatus for discriminating whether a walking style is an ascending walk or a descending staircase. A walking style discrimination method executed by a walking style discrimination device having a processor and a storage device connected to the processor,
  The storage device stores three-dimensional orthogonal coordinate system angular velocity data measured by an angular velocity sensor attached to a pedestrian and three-dimensional orthogonal coordinate system acceleration data measured by an acceleration sensor attached to the pedestrian. Hold and
  The walking style discrimination method is:
  The processor calculates a power spectrum of angular velocities of two coordinate axes excluding the coordinate axis having the smallest angle with the vertical direction;
  The pedestrian is based on whether the power spectrum intensity of the one-step cycle component and the two-step cycle component calculated based on the power spectrum satisfies a predetermined condition. A procedure for determining whether the walking style is walking on stairs or walking on flat ground,
  For the section in which the processor determines that the walking mode of the pedestrian is stair walking, based on whether the power spectrum intensity of the component of the cycle of the two steps satisfies a predetermined condition, the processor And a procedure for discriminating whether the walking style is an ascending walk or a descending staircase.

Images