JP2023168088A - State determination device, state determination system and state determination program - Google Patents

Abstract

To provide a state determination device, state determination system, and state determination program capable of determining states of site workers at a construction site who have few opportunities to rest.SOLUTION: A state determination device for determining states of site workers at a construction site comprises: an acceleration measurement part for measuring acceleration of each site worker; an acceleration acquisition part for acquiring acceleration data from the acceleration measurement part; a pulse measurement part for measuring a pulse of the site worker; a pulse acquisition part for acquiring pulse data from the pulse measurement part; and a state determination part for determining whether or not the site worker is in an abnormal state based on sensor data including acceleration data acquired by the acceleration acquisition part and the pulse data acquired by the pulse acquisition part.SELECTED DRAWING: Figure 3

Description

The present invention relates to a condition determining device, a condition determining system, and a condition determining program for determining the condition of a construction site worker.

Conventionally, various condition determination systems for determining the condition of construction site workers have been proposed.

Patent Document 1 discloses that a threshold value for determining whether the risk of developing heat stroke has increased is generated based on biometric information of a site worker while at rest, and the biometric information during work is compared with the threshold value. A heat stroke risk management system for determining the risk of developing heat stroke among field workers has been disclosed. That is, in Patent Document 1, biological information of a site worker at rest is acquired, a threshold value is generated, and the state during work is determined based on this.

JP2019-217197A

However, most construction site workers work while moving and have few opportunities to rest. Therefore, in the state determination system such as that disclosed in Patent Document 1, the acquisition of biological information during rest may be insufficient, and a threshold value for state determination may not be generated. In such a case, the condition determination system itself does not function, and the condition of the site worker cannot be determined.

In view of the above points, an object of the present invention is to provide a condition determination device, a condition determination system, and a condition determination program that can determine the condition of construction site workers who have few chances to be in a resting state. .

Each of the following configurations is a means for solving the above problems.

<Configuration 1>
A condition determination device for determining the condition of a site worker at a construction site, the device comprising: an acceleration measurement section that measures acceleration of the site worker; an acceleration acquisition section that acquires acceleration data from the acceleration measurement section; and the construction site worker. a pulse measurement unit that measures the pulse of a person; a pulse acquisition unit that acquires pulse data from the pulse measurement unit; acceleration data acquired by the acceleration acquisition unit; and pulse data acquired by the pulse acquisition unit; A state determining device comprising: a state determining unit that determines whether the on-site worker is in an abnormal state based on sensor data including:

<Configuration 2>
The state determination device further includes a distribution center calculation unit that calculates a distribution center of the predetermined number of sensor data based on the predetermined number of sensor data acquired by the acceleration acquisition unit and the pulse rate acquisition unit; an approximate straight line calculation unit that calculates an approximate straight line of the predetermined number of sensor data based on the number of sensor data; and a separation distance from the distribution center to the sensor data acquired by the acceleration acquisition unit and the pulse rate acquisition unit. a separation distance calculation unit that calculates a separation distance calculation unit that calculates a separation distance calculation unit that calculates, when a portion of the approximate straight line extending from the distribution center is taken as a starting line, extending from the distribution center to the sensor data acquired by the acceleration acquisition unit and the pulse rate acquisition unit. a declination calculation unit that calculates a declination angle that a virtual vector makes with respect to the starting line; An abnormality range defined by an angle is used, and the state determination unit determines whether or not the on-site work is possible based on the distance and declination of the sensor data and the range of the separation distance and declination defined as the abnormal range. The state determining device according to configuration 1, wherein the device determines whether the person is in an abnormal state.

<Configuration 3>
The state determining device according to configuration 2, wherein the separation distance calculation unit calculates the separation distance as a Maharabinos distance.

<Configuration 4>
A state determination system for determining the state of a site worker at a construction site, comprising an electronic device and a server connected to the electronic device via a network, and an acceleration measurement system for measuring acceleration of the site worker. an acceleration acquisition section that acquires acceleration data from the acceleration measurement section; a pulse measurement section that measures the pulse of the on-site worker; a pulse acquisition section that acquires pulse data from the pulse measurement section; and the acceleration acquisition section. and a state determining unit that determines whether the site worker is in an abnormal state based on sensor data including acceleration data acquired by the part and pulse data acquired by the pulse rate acquiring unit. A state determination system featuring:

<Configuration 5>
A condition determination program for determining the condition of a site worker at a construction site, the program comprising: an acceleration acquisition means for acquiring measured acceleration data of the site worker; and a measured pulse rate of the site worker. The on-site worker is in an abnormal state based on sensor data including pulse rate acquisition means for acquiring data on the field worker, acceleration data acquired by the acceleration acquisition means, and pulse rate data acquired by the pulse rate acquisition means. A state determining program, comprising: a state determining means for determining whether a state exists.

According to the present invention, it is possible to determine the condition of a site worker regardless of whether he or she is in a resting state or not, so it is possible to determine the condition of a site worker at a construction site where there is little chance of being in a resting state. .

FIG. 1 is a diagram showing an example of a state determination device according to the present embodiment. 3 is a diagram showing the hardware configuration of a wearable device 5. FIG. 3 is a diagram showing a functional configuration of a wearable device 5. FIG. It is an example of the distribution diagram of the sensor data acquired predetermined number. 5 is a schematic diagram of an abnormal range defined for the sensor data distribution diagram shown in FIG. 4. FIG. 5 is a flowchart showing the flow of processing performed in the wearable device 5 to determine the condition of a site worker. 1 shows an example of a computer that executes a state determination program related to the state determination device of this embodiment. The orthogonal coordinate system shown in FIG. 5 is converted into a polar coordinate system.

[Status determination device]
Hereinafter, a state determination device according to this embodiment will be described with reference to the drawings. FIG. 1 is a diagram showing an example of a state determination device in this embodiment. In addition, in the following description, a site worker at a construction site may be simply referred to as a site worker. In this embodiment, a wearable device worn by a monitoring target (on-site worker) is used as the state determination device (electronic device), but the present invention is not limited to this, and may be a mobile terminal or the like, for example. The state determination system 1 includes a server 2, a management terminal 3, a gateway 4, and a wearable device 5. The server 2, management terminal 3, and gateway 4 are connected via network A. Gateway 4 and wearable device 5 are connected via wireless communication.

As the network A, any type of communication network can be used, such as the Internet, LAN (Local Area Network), VPN (Virtual Private Network), etc., regardless of whether it is wired or wireless. As the wireless communication used to connect the gateway 4 and the wearable device 5, in addition to wireless communication such as 5G, LTE (Long Term Evolution), and WiFi (Wireless Fidelity), short-range wireless communication such as Bluetooth (registered trademark) is used. be able to.

The server 2 is a server device that integrally manages the wearable devices 5 worn by each monitoring target. The management terminal 3 is a terminal used by a manager who monitors each site worker, and is, for example, a smartphone or a personal computer. When it is determined that the field worker is in an abnormal state, a notification that the field worker is in an abnormal state is sent from the wearable device 5 to the management terminal 3, and the administrator should monitor the field worker with special care. It becomes possible to recognize on-site workers.

The gateway 4 is a computer such as a smartphone or a relay station that is placed near a field worker wearing the wearable device 5. The gateway 4 receives from the wearable device 5 the determination result as to whether the field worker is in an abnormal state and the log data for each predetermined time, and transmits the received log data to the server 2 via the network A.

The wearable device 5 is a computer worn by a site worker, and is in the form of a wristwatch, a badge, a tag, or the like. The wearable device 5 includes an acceleration sensor (acceleration measuring section) and a pulse sensor (pulse measuring section), which will be described later, and detects the acceleration and pulse of the site worker.

(Hardware configuration)
FIG. 2 is a diagram showing the hardware configuration of the wearable device 5. As shown in FIG. Referring to FIG. 2, wearable device 5 includes a wireless section 51, an acceleration sensor 52, a pulse sensor 53, a storage 54, a memory 55, and a processor 56.

The wireless unit 51 is a communication interface for performing wireless communication with the gateway 4. Since the type of wireless communication has been described above, a description thereof will be omitted. Acceleration sensor 52 is a sensor that detects acceleration that occurs in wearable device 5 (that is, acceleration that occurs in a field worker wearing wearable device 5). The pulse sensor 53 is a sensor that detects the pulse of a site worker wearing the wearable device 5.

The storage 54 and memory 55 are storage devices that store various data and programs. The processor 56 reads a program for executing various processes to be described later from the storage 54, expands it into the memory 55, and executes the read program.

(Functional configuration)
FIG. 3 is a diagram showing the functional configuration of the wearable device 5. As shown in FIG. Referring to FIG. 3, wearable device 5 includes an acceleration acquisition section 61, a pulse acquisition section 62, a storage section 63, a calculation section 64, a state determination section 65, and a notification section 66. Note that the storage unit 63 is stored in the storage 54 or the memory 55. The acceleration acquisition section 61, the pulse acquisition section 62, the calculation section 64, the state determination section 65, and the notification section 66 are an example of an electronic circuit included in the processor 56 and an example of a process executed by the processor 56.

The acceleration acquisition unit 61 is a processing unit that acquires acceleration data detected by the acceleration sensor 52 and stores it in the sensor data DB 631 included in the storage unit 63. The sensor data DB 631 stores acceleration data detected by the acceleration sensor 52, for example, in chronological order.

The pulse acquisition unit 62 is a processing unit that acquires pulse data detected by the pulse sensor 53 and stores it in the sensor data DB 631 included in the storage unit 63. The sensor data DB 631 stores pulse data detected by the pulse sensor 53, for example, in chronological order.

In the sensor data DB 631, acceleration data acquired by the acceleration acquisition unit 61 and pulse data acquired by the pulse rate acquisition unit 62 are stored in synchronization. Hereinafter, acceleration data and pulse data may be collectively referred to as sensor data.

The calculation unit 64 includes a distribution center calculation unit 641, an approximate straight line calculation unit 642, a separation distance calculation unit 643, and an argument calculation unit 644.

When the acceleration acquisition unit 61 and the pulse acquisition unit 62 acquire a predetermined number of sensor data of the site worker, the distribution center calculation unit 641 reads out the predetermined number of sensor data from the sensor data DB 631 and calculates the sensor data of the predetermined number of sensors. Calculate the center of the data distribution. FIG. 4 is an example of a distribution diagram of a predetermined number of acquired sensor data. In FIG. 4, the horizontal axis represents acceleration and the vertical axis represents pulse rate. Referring to FIG. 4, in the present embodiment, the distribution center calculation unit 641 calculates an arithmetic average of the sum of acquired acceleration values divided by the number of data based on a predetermined number of acquired sensor data (indicated by white circles). Calculate the average value X of the acceleration, and calculate the average value Y of the pulse by the arithmetic average of the total of the acquired pulse values divided by the number of data, and if the average value of the acceleration is X and the average value of the pulse The point where is Y (indicated by a black circle) is defined as the distribution center K. However, the method of calculating the average value of acceleration and the average value of pulse rate is not limited to this. For example, the distribution center calculation unit 641 may calculate the average value of acceleration and the average value of pulse rate using a weighted average or a moving average, and use them as the distribution center K. Moreover, the distribution center K is not limited to the average value of acceleration and the average value of pulse, and may be calculated by extracting the mode or median value from the sensor data read from the sensor data DB 631.

When a predetermined number of sensor data are acquired by the acceleration acquisition unit 61 and the pulse acquisition unit 62, the approximate straight line calculation unit 642 reads out the predetermined number of sensor data from the sensor data DB 631 and calculates the distribution of the predetermined number of sensor data. An approximated straight line M is calculated. Referring again to FIG. 4, in this embodiment, the approximate straight line calculation unit 642 calculates the approximate straight line using the least squares method with acceleration and pulse rate as variables, based on a predetermined number of acquired sensor data (indicated by white circles). Calculate straight line M. However, the method of calculating the approximate straight line is not limited to this, and other known calculation methods can be used.

Note that when calculating the distribution center K in the distribution center calculation unit 641 and calculating the approximate straight line M in the approximate straight line calculation unit 642, data in which the acceleration is too large or too small, or data in which the pulse rate is too large or too small ( It is preferable to exclude outlier data). By calculating the distribution center K and the approximate straight line M while excluding outlier data, noise on the distribution center K and the approximate straight line M can be reduced, and the distribution center K and the approximate straight line M can be calculated more accurately. For example, in the sensor data distribution, the top 1% data and bottom 1% data of acceleration, and the top 1% data and bottom 1% data of pulse rate can be set as outlier data. The data range set as value data is not limited to this.

When the sensor data is acquired by the acceleration acquisition unit 61 and the pulse acquisition unit 62, the separation distance calculation unit 643 reads the sensor data from the sensor data DB 631, and calculates the sensor data from the distribution center K calculated by the distribution center calculation unit 641. The length of the virtual vector S (separation distance D from the distribution center K to the sensor data L) extending to the point L representing the data (indicated by a fine dot) is calculated (see FIG. 4). In this embodiment, the separation distance calculation unit 643 calculates the separation distance D from the distribution center K to the sensor data L as a Maharabinos distance. The reason for calculating the separation distance D as the Maharabinos distance will be described later.

In the separation distance calculation unit 643, the Maharabinos distance D is calculated using the following formula. Note that χ represents the sensor data (vector) for which the separation distance is calculated, μ represents the average value of the distribution of sensor data (i.e., acceleration value X and pulse value Y at the distribution center K), and Σ represents the sensor data Represents the covariance matrix of the distribution.

When the sensor data is acquired by the acceleration acquisition unit 61 and the pulse acquisition unit 62, the declination calculation unit 644 reads the sensor data from the sensor data DB 631, and determines the portion of the approximate straight line M extending from the distribution center K as the starting line. Then, the angle θ that the virtual vector S extending from the distribution center K calculated by the distribution center calculation unit 641 to the point L representing the sensor data makes with the starting line is the angle θ of the sensor data L. calculate. Note that depending on the calculation result of the distribution center K by the distribution center calculation unit 641 and the calculation result of the approximate straight line M by the approximate straight line calculation unit 642, the approximate straight line M may not be configured to pass through the distribution center K. In this case, by performing a correction in which the approximate straight line M is translated in parallel to a position passing through the distribution center K, it is possible to calculate the declination angle of the predetermined sensor data using the method described above.

The abnormal range DB 632 stores data on the range of separation distance D and angle of deviation θ (abnormal range) in which it is determined that the on-site worker is in an abnormal state. The method of defining the abnormality range will be described later.

When new sensor data is acquired by the acceleration acquisition unit 61 and the pulse acquisition unit 62, the state determination unit 65 reads the abnormal range data from the abnormal range DB 632, and determines the separation distance D and the declination of the point representing the new sensor data. It is determined whether θ is within the abnormal range. Then, if it is determined that the separation distance D and the declination angle θ of the points representing the new sensor data are continuously in the abnormal range for a predetermined period of time (for example, if the new sensor data is in the abnormal range, ), it is determined that "the site worker is in an abnormal state". However, the criteria for determining that "the site worker is in an abnormal state" is limited to "whether or not the separation distance D and the declination angle θ of the points representing new sensor data remain within the abnormal range for a predetermined period of time." It's not something you can do. For example, the standard may be set to ``Whether or not the number of times that the separation distance D and declination angle θ of points representing new sensor data are within the abnormal range among the latest predetermined number of new sensor data is equal to or greater than the standard number of times.'' Good too. In addition, if the separation distance D and/or the declination θ of the points representing new sensor data significantly deviate from the abnormal range (for example, the separation distance D and/or the declination θ of the points representing the new sensor data and the abnormal state (if the difference between the separation distance D and/or the range of the declination angle θ that is determined to be equal to or greater than a predetermined value), it is determined that "the site worker is in an abnormal state" regardless of the above criteria. It may be configured as follows. The condition determination unit 65 can further determine the detailed condition of the site worker according to the abnormality range in which the point representing the new sensor data exists, but this point will be described later.

When the state determination unit 65 determines that “the site worker is in an abnormal state,” the notification unit 66 transmits a notification to that effect to the management terminal 3. A manager who receives notification that a site worker is in an abnormal state can implement various measures to encourage the site worker to recover, such as calling out to the site worker or reassigning the site worker. The notification unit 66 can transmit the detailed status of the site worker to the management terminal 3 when the status determining unit 65 determines the detailed status of the site worker.
Furthermore, when the status determination unit 65 determines that “the site worker is in an abnormal state,” the notification unit 66 displays a message such as “abnormal state” on a display unit (not shown) provided in the wearable device 5. A configuration may be provided. Thereby, the field worker who has confirmed the message can understand his or her own condition and take various measures to recover the condition.

(Circumstances leading to the invention according to this embodiment)
The circumstances leading to the invention according to this embodiment will be explained in detail with reference to FIGS. 2 to 4 again. First, the state determination device according to the present embodiment includes an acceleration sensor 52 (acceleration acquisition section) that measures acceleration of a site worker, an acceleration acquisition section 61 that acquires acceleration data from the acceleration sensor 52, and a pulse rate of the site worker. A pulse sensor 53 (pulse measurement unit) that measures pulse rate, a pulse acquisition unit 62 that acquires pulse data from the pulse sensor 53, and a pulse acquisition unit 62 that acquires acceleration data acquired by the acceleration acquisition unit 61 and pulse data acquired by the pulse acquisition unit 62. A state determination unit 65 is provided that determines whether the site worker is in an abnormal state based on the included sensor data. According to this configuration, the condition of the site worker can be determined based on the acceleration and pulse rate of the site worker, regardless of whether the site worker is in a resting state or a non-resting state. Therefore, according to the state determination device according to the present embodiment, it is possible to determine the state of a construction site worker who has few chances to be in a resting state.

Here, since there are individual differences in the work content and pulse rate of each site worker, the distribution of sensor data differs for each site worker. Therefore, in order to more accurately determine the condition of the on-site worker, the present inventors have conducted extensive studies on a method for defining the abnormal range. For example, if the abnormal range is uniformly defined based on the orthogonal coordinate area of acceleration and pulse rate, even if the abnormal range is defined based on the sensor data distribution of a certain field worker, the abnormal range will not be the same as that of other field workers. However, there is a possibility that the accuracy of condition determination for on-site workers may not be sufficiently improved. Therefore, the present inventors came up with the idea of defining the abnormal range as a polar coordinate area with the approximate straight line M of the sensor data distribution as the starting line, rather than uniformly defining it using the rectangular coordinate area of acceleration and pulse rate. Specifically, the abnormal range is defined based on (1) the deviation angle θ of the sensor data distribution with respect to the approximate straight line M, and (2) the separation distance D from the distribution center K. According to this configuration, regardless of the difference in sensor data distribution, the abnormal range is determined based on the deviation angle based on the correlation direction of each sensor data distribution (the direction in which the approximate straight line M extends) and the relative distance from the distribution center K. Can be specified. Therefore, regardless of the state of the sensor data distribution of the on-site worker, it is possible to define an abnormality range suitable for each on-site worker, and the condition of the on-site worker can be determined more accurately.

Here, in the upper left hatched area A1 in the distribution diagram shown in FIG. 4, the pulse/acceleration ratio is excessively high. That is, if the sensor data of the site worker exists in the hatched area A1, it is presumed that "the site worker is in poor physical condition." On the other hand, in the lower right hatched area A2 in the distribution diagram shown in FIG. 4, the pulse/acceleration ratio is excessively low. That is, if sensor data of the site worker exists in the hatched area A2, it is presumed that "the attentiveness of the site worker has decreased."

The work carried out by construction site workers is highly dangerous, so working on-site while in poor physical condition or with reduced alertness is likely to cause injury or accidents. Therefore, it is considered to be a "state that interferes with on-site work" (that is, an abnormal state). Therefore, by determining the abnormal range to include at least part of the hatched areas A1 and A2, the condition of the site worker can be determined even more accurately.

Here, in a known system for determining the condition of field workers, in a distribution map in which the vertical and horizontal axes are set to predetermined parameters, the separation distance from the distribution center to the sensor data to be evaluated is expressed as a Euclidean distance. I was evaluating it. At this time, the boundary line of the abnormal range that the field worker determines to be in an abnormal state is defined isotropically from the center of distribution, so it has a perfect circular shape (or an arc shape as a part of a perfect circle). was. However, when evaluating the separation distance from the distribution center using the Euclidean distance, sensor data in a normal state and sensor data in an abnormal state (hatched areas A1, A2) cannot be distinguished, making it difficult to more accurately judge the condition of field workers. The possibility of improvement is reduced.

In view of this point, the present inventors first focused on the point that the hatched areas A1 and A2 exist in a direction that goes against the correlation of the distribution (a direction in which the orthogonal axis N perpendicular to the approximate straight line M extends). Then, we came up with the idea of evaluating the separation distance from the center of distribution to the sensor data to be evaluated using Mahalanobis distance instead of Euclidean distance. According to this configuration, compared to the case where the separation distance from the distribution center is evaluated using the Euclidean distance, the separation distance in the direction that goes against the correlation of the distribution can be evaluated more strictly, so the abnormal range can be determined more accurately. The condition of the field worker can be determined even more accurately.

Based on these points, FIG. 5 shows a schematic diagram of the abnormality range defined in this embodiment. FIG. 5 is a schematic diagram of the abnormality range determined for the sensor data distribution diagram shown in FIG. 4. In FIG. 5, components having the same configuration and functions as those in FIG. 4 are given the same reference numerals. However, for convenience of explanation, the distribution of sensor data indicated by white circles in FIG. 4 is omitted in FIG. 5. For convenience of further explanation, the portion of the approximate straight line M that extends in the direction in which acceleration increases from the distribution center K is assumed to be M+, and the portion that extends in the direction in which acceleration decreases from the distribution center K is assumed to be M-.

The abnormal range in FIG. 5 is defined as "a range that is within a specific declination range and is greater than or equal to a predetermined separation distance in the Maharabinos distance", and thereby the hatched areas A1 and A2 shown in FIG. The abnormal range is defined so as to include the abnormal range. In this embodiment, the abnormal range includes a first abnormal range B1 and a second abnormal range B2.

The first abnormal range B1 shown in FIG. 5 is an abnormal range that includes the hatched area A1, and is within the declination range from a position rotated by θ1 counterclockwise from M+ to a position rotated clockwise by θ2 from M-. , and the separation distance is defined in a range that is equal to or greater than r1 (Mahalanobis distance). If the new sensor data continues for a predetermined period of time and belongs to the first abnormality range B1, it is determined that "the site worker is in an abnormal state". Here, as mentioned above, since it is assumed that "the field worker is in poor physical condition" in the hatched area A1, if the new sensor data continues for a predetermined period of time and belongs to the first abnormal range B1, "field worker It is also possible to make a detailed determination that the person is in poor physical condition.

The second abnormal range B2 shown in FIG. 5 is an abnormal range that includes the hatched area A2, and is within the declination range from a position rotated θ3 clockwise from M+ to a position rotated θ4 counterclockwise from M-. , and the separation distance is defined to be equal to or greater than r2 (Mahalanobis distance). If the new sensor data continues for a predetermined period of time and belongs to the second abnormality range B2, it is presumed that "the site worker is in an abnormal state". Here, as mentioned above, it is assumed that "the attentiveness of the site worker is reduced" in the hatched area A2, so if the new sensor data continues for a predetermined period of time and belongs to the second abnormal range B2, It is also possible to determine in detail that "the on-site worker's attentiveness is decreasing."

Incidentally, the argument angles θ1 to θ4 may be set to different values, or any two or more argument angles may be set to the same value. Further, the separation distance r1 and the separation distance r2 may be set to different values or may be set to the same value.

(Flow of processing for determining the condition of field workers)
Next, the flow of the process performed in the wearable device 5 to determine the condition of the field worker will be explained. FIG. 6 is a flowchart showing the flow of processing performed in the wearable device 5 to determine the condition of the site worker.

First, the acceleration acquisition unit 61 acquires acceleration data of the site worker measured by the acceleration sensor 52 (S101), and the pulse rate acquisition unit 62 acquires data of the site worker's pulse measured by the pulse sensor 53. Acquire (S102). The acceleration data acquired by the acceleration acquisition unit 61 and the pulse data acquired by the pulse rate acquisition unit 62 are stored in the sensor data DB 631 in synchronization.

When the acquired sensor data (acceleration data and pulse rate data) reaches a predetermined number (S103 Yes), the distribution center calculation unit 641 reads out the predetermined number of sensor data from the sensor data DB 631 and calculates the sensor data based on the distribution of the read sensor data. , the distribution center K is calculated (S104), and the approximate straight line calculation unit 642 reads out the predetermined number of sensor data from the sensor data DB 631, and calculates the approximate straight line M based on the distribution of the read sensor data (S105). . The method of calculating the distribution center K and the approximate straight line M has been described above, so the explanation thereof will be omitted. On the other hand, if the acquired sensor data (acceleration data and pulse rate data) does not reach the predetermined number (S103 No), the acceleration acquisition unit 61 and the pulse rate acquisition unit 62 acquire the sensor data again (S101, S102).

Next, the acceleration acquisition unit 61 acquires new acceleration data measured by the acceleration sensor 52 (S106), and the pulse acquisition unit 62 acquires new pulse data measured by the pulse sensor 53 (S107). . The new acceleration data acquired by the acceleration acquisition unit 61 and the new pulse rate data acquired by the pulse rate acquisition unit 62 are stored in the sensor data DB 631 in synchronization.

Subsequently, the separation distance calculation unit 643 reads new sensor data from the sensor data DB 631, calculates the separation distance D from the distribution center K (calculated in S104) to the new sensor data (S108), and the declination calculation unit 644 However, when new sensor data is read from the sensor data DB 631 and the approximate straight line M extending from the distribution center K is set as the starting line, the virtual vector S extending from the distribution center K (calculated in S104) to the new sensor data is the starting line. An angle (deflection θ) made with respect to the target is calculated (S109). As described above, in this embodiment, the separation distance D is calculated as the Maharavinos distance.

The state determination unit 65 reads the abnormal range data from the abnormal range DB 632 and calculates the separation distance D of the new sensor data calculated by the separation distance calculation unit 643 (S108) and the deviation of the new sensor data calculated by the declination calculation unit 644. Based on the angle θ (S109), it is determined whether the new sensor data exists in the abnormal ranges B1 and B2 (S110). If it is determined that the new sensor data does not exist in the abnormal ranges B1 and B2 (S110 No), the acceleration acquisition unit 61 and the pulse acquisition unit 62 acquire new sensor data again (S106, S107). . On the other hand, if it is determined that the new sensor data exists in the abnormal range B1 or B2 (S110 Yes), the state determination unit 65 determines whether the new sensor data continues to exist in the abnormal range B1 or B2 for a predetermined period of time. is determined (S111).

If it is determined that the new sensor data continues for a predetermined period of time and is not within the abnormal ranges B1 and B2 (S111 No), the acceleration acquisition section 61 and the pulse acquisition section 62 acquire new sensor data again. (S106, S107). On the other hand, if it is determined that the new sensor data continues for a predetermined period of time and remains within the abnormality range B1 or B2 (S111 Yes), the state determination unit 65 determines that "the site worker is in an abnormal state" (S112). ), the notification unit 66 notifies the management terminal 3 to that effect (S113). At the time of this notification, the detailed condition of the site worker (poor physical condition/decreased attentiveness) is also notified according to the abnormality ranges B1 and B2 in which the new sensor data exists. The manager who receives the notification can take various measures to recover the condition of the site worker, thereby reducing the risk of the site worker's condition worsening and causing injury or an accident.

[Status determination program]
Each function constituting the above-mentioned state determination device can be realized by a program, and a computer program prepared in advance to realize each function is stored in an auxiliary storage device, and a control section is stored in the auxiliary storage device. The functions of each section can be operated by reading a program into the main storage device and having the control section execute the program read into the main storage device. Therefore, using FIG. 7, an example of a computer that executes a state determination program having the same functions as the above-described state determination device will be described below.

FIG. 7 shows an example of a computer that executes a state determination program related to the state determination device of this embodiment. As shown in FIG. 7, computer 1000 includes an operation section 1100, a speaker 1200, a camera 1300, a display 1400, and a communication section 1500. Further, the computer 1000 includes a CPU (control unit) 1600, an HDD (auxiliary storage device) 1700, and a RAM (main storage device) 1800. These units are connected via a bus 1900.

As shown in FIG. 7, the HDD 1700 stores in advance a state determination program that performs the same functions as the functional units shown in the above-described embodiments. This state determination program may be integrated or separated as appropriate, similar to each component of each functional unit shown in FIG. In other words, all of the data stored in the HDD 1700 does not need to be stored in the HDD 1700 at all times, and only data required for processing may be stored in the HDD 1700.

Then, the CPU 1600 reads the status determination program from the HDD 1700 and expands it into the RAM 1800. Thereby, the state determination program functions as a state determination process. In this state determination process, various types of data read from the HDD 1700 are expanded to an area allocated to itself on the RAM 1800 as appropriate, and various processes are executed based on the expanded various data. Note that the state determination process includes processing executed by each functional unit shown in FIG. 3, for example, the processing shown in FIG. 6. Furthermore, it is not necessary for all the processing units to be virtually realized on the CPU 1600 to operate on the CPU 1600 at all times, and it is sufficient that only the processing units necessary for processing are virtually realized.

Note that the above-mentioned status determination programs do not necessarily need to be stored in the HDD 1700 from the beginning, but each program may be stored in a "computer-readable recording medium" and the computer 1000 can store this "computer-readable recording medium". Alternatively, each program may be obtained from and executed. Examples of "computer-readable recording media" include optical disks such as CD-ROM, phase change optical disks such as DVD-ROM, magneto-optical disks such as MO (Magnet Optical) and MD (Mini Disk), and floppy disks (registered trademark). Examples include magnetic disks such as discs and removable hard disks, and memory cards such as CompactFlash (registered trademark), SmartMedia, SD memory cards, and memory sticks. Furthermore, hardware devices such as integrated circuits (IC chips, etc.) that are specially designed and configured for the purpose of the present invention are also included as recording media. Furthermore, each program may be stored in another computer or server device connected to the computer 1000 via a public line, the Internet, a LAN, a WAN, or the like. In that case, computer 1000 may acquire each program from another computer or server device and execute it.

(Modified example)
In the embodiment described above, the wearable device 5 includes an acceleration acquisition section 61, a pulse acquisition section 62, a storage section 63, a calculation section 64, a state determination section 65, and a notification section 66. However, the present invention is not limited to this, and the server 2 includes at least some of the acceleration acquisition section 61, the pulse acquisition section 62, the storage section 63, the calculation section 64, the state determination section 65, and the notification section 66. It may be a configuration. For example, if the server 2 is configured to include at least an acceleration acquisition section 61, a pulse acquisition section 62, and a storage section 63, the acceleration data measured by the acceleration sensor 52 included in the wearable device 5 is stored on the network A The data of the pulse rate measured by the pulse sensor 53 of the wearable device 5 is acquired by the acceleration acquisition unit 61 of the server 2 via the network A, and stored in the sensor data DB 631 by the acceleration acquisition unit 61. is acquired by the pulse acquisition unit 62 provided in the server 2, and stored in the sensor data DB 631 by the pulse acquisition unit 62.

(Modified example)
The abnormal range stored in the abnormal range DB 632 is defined using, for example, machine learning based on past sensor data of the site worker. Further, appropriate adjustments may be made to a predefined abnormality range based on new acceleration data acquired by the acceleration acquisition unit 61 and new pulse data (i.e., new sensor data) acquired by the pulse acquisition unit 62. . Further, the range of the separation distance D and the deviation angle θ defined as the abnormal range may be different for each site worker, or may be the same.

(Modified example)
The abnormal range is defined as a polar coordinate area based on (1) the deviation angle θ of the sensor data distribution with respect to the approximate straight line M, and (2) the separation distance D from the distribution center K. In the above embodiment, as shown in FIG. 5, in a rectangular coordinate system with acceleration on the horizontal axis and pulse on the vertical axis, abnormal ranges B1 and B2 are defined as polar coordinate areas, and the condition of the site worker is determined. Ta. However, the present invention is not limited to this, and as shown in FIG. 8, after converting the orthogonal coordinate system graph to a polar coordinate system graph with the approximate straight line M extending from the distribution center K as the starting line, the abnormal range can be defined as a polar coordinate area. , the condition of the site worker may be determined. FIG. 8 shows the orthogonal coordinate system shown in FIG. 5 converted into a polar coordinate system. Structures and areas corresponding to those in FIG. 5 are given the same numbers and letters. In FIG. 8, the approximate straight line M, which is the starting line, is set on the vertical axis, but the approximate straight line M may be set on the horizontal axis. By converting a Cartesian coordinate system graph to a polar coordinate system graph, the abnormal range as a polar coordinate area can be defined in the polar coordinate system with the same starting line, so compared to the Cartesian coordinate system, sensor data is less likely to fall within the abnormal range. It is possible to more easily determine whether or not there is one.

Furthermore, in FIG. 5, the separation distance from the distribution center K is calculated as the Maharabinos distance, so even if the separation distance is apparently the same, the Maharabinos distance is different. In view of this, when converting to a polar coordinate system, it is preferable to define the polar coordinate system so that when the separation distances (Maharabinos distances) from the distribution center K are equal, the separation distances are apparently the same. . With this regulation, it is possible to visually and easily determine whether the sensor data is within the abnormal range.

1 State Judgment System 5 Wearable Device 52 Acceleration Sensor 53 Pulse Sensor 61 Acceleration Acquisition Unit 62 Pulse Acquisition Unit 65 Status Judgment Unit 641 Distribution Center Calculation Unit 642 Approximate Straight Line Calculation Unit 643 Separation Distance Calculation Unit 644 Declination Calculation Unit

Claims (5)

A condition determination device for determining the condition of a site worker at a construction site,
an acceleration measurement unit that measures the acceleration of the site worker;
an acceleration acquisition unit that acquires acceleration data from the acceleration measurement unit;
a pulse measurement unit that measures the pulse of the site worker;
a pulse acquisition unit that acquires pulse data from the pulse measurement unit;
a state determination unit that determines whether the site worker is in an abnormal state based on sensor data including acceleration data acquired by the acceleration acquisition unit and pulse data acquired by the pulse rate acquisition unit;
A state determination device comprising: The state determination device further includes:
a distribution center calculation unit that calculates a distribution center of the predetermined number of sensor data based on the predetermined number of sensor data acquired by the acceleration acquisition unit and the pulse acquisition unit;
an approximate straight line calculation unit that calculates an approximate straight line of the predetermined number of sensor data based on the predetermined number of sensor data;
a distance calculation unit that calculates a separation distance from the distribution center to the sensor data acquired by the acceleration acquisition unit and the pulse rate acquisition unit;
When a part of the approximate straight line extending from the center of distribution is taken as the starting line, a virtual vector extending from the center of distribution to the sensor data acquired by the acceleration acquisition section and the pulse acquisition section is relative to the starting line. an argument calculation unit that calculates an argument made by
Equipped with
In the state determining unit, an abnormal range defined by a separation distance and a declination angle is used to determine that the site worker is in an abnormal state,
The state determination unit determines whether the site worker is in an abnormal state based on the separation distance and declination angle of the sensor data and the range of the separation distance and declination angle defined as the abnormal range. The state determining device according to claim 1, characterized in that: The state determination device according to claim 2, wherein the separation distance calculation unit calculates the separation distance as a Maharabinos distance. A condition determination system for determining the condition of a site worker at a construction site, the system comprising:
comprising an electronic device and a server connected to the electronic device via a network,
an acceleration measurement unit that measures the acceleration of the site worker;
an acceleration acquisition unit that acquires acceleration data from the acceleration measurement unit;
a pulse measurement unit that measures the pulse of the site worker;
a pulse acquisition unit that acquires pulse data from the pulse measurement unit;
a state determination unit that determines whether the site worker is in an abnormal state based on sensor data including acceleration data acquired by the acceleration acquisition unit and pulse data acquired by the pulse rate acquisition unit;
A state determination system comprising: A condition determination program for determining the condition of a site worker at a construction site,
to the computer,
an acceleration acquisition means for acquiring measured acceleration data of a site worker;
pulse acquisition means for acquiring measured pulse data of the site worker;
state determining means for determining whether the site worker is in an abnormal state based on sensor data including acceleration data acquired by the acceleration acquiring means and pulse data acquired by the pulse rate acquiring means;
A state determination program characterized by causing execution of.