JP6490461B2 - Anomaly detection terminal and program - Google Patents

Description

  The present invention relates to an anomaly detection terminal capable of detecting an anomaly such as a human fall and a program used for the anomaly detection terminal.

  Conventionally, there are systems that detect and report accidents such as health problems and falls of the elderly. The following Patent Document 1 discloses a detection unit including a piezoelectric sensor that detects body movement and respiration or pulse of a human body, a determination unit that determines normality or abnormality of health based on detection data from the detection unit, and the determination It is equipped with a wearable device that is integrated with a communication means for reporting that the means has determined abnormality, and is attached to the subject, and the detection signal from the piezoelectric sensor is inverted and amplified through a low-pass filter. The presence / absence of breathing is determined based on the difference (LFBM), the pulse wave and body movement are determined based on the difference between the maximum value and the minimum value (HFBM) of the inverted and amplified output through the high-pass filter, and the high-pass filter output (HPFO) There is disclosed a safe living support device for elderly living alone, characterized in that it is configured to determine body movement.

  In this device, when a resting state in which body movement is not detected continues for 60 minutes, it is determined that the body is in an abnormal state, and also when a resting state continues for 5 minutes after a fall has been detected, The emergency call is made after warning by the buzzer. However, if a predetermined body movement is detected after a fall is detected, it is determined that there is no need for notification because the user has acted, and the 5-minute timer is reset (fall detection is canceled). .

JP 2002-360522 A

  However, there are various modes of falls, and for example, when the user falls from a state of sitting on a chair, the impact at the time of the fall may not be noticeable. Further, even when the user loses his consciousness from the standing posture and falls down, the impact may be reduced if the head or body collides with the ground first.

  Even in such a case, if there is a life monitoring function (detection of a physical abnormal state due to the continuation of a resting state) as described in Patent Document 1, an abnormality report can be made if there is no user movement over a predetermined monitoring time. However, since this monitoring time is relatively long, it takes a long time from the occurrence of a fall accident to report, and there is a risk that emergency treatment will be delayed.

  In view of the circumstances as described above, an object of the present invention is to provide an abnormality detection terminal and a program capable of detecting an occurrence of an abnormal situation due to a user's falling even when a shock due to the falling is relatively low. .

  In order to achieve the above object, an abnormality detection terminal according to an aspect of the present invention is a terminal that is carried by a user and detects the user's fall, and includes a motion sensor, a fall detection unit, and a body motion detection unit. And an abnormality determining means. The motion sensor detects the motion of the terminal and outputs motion data. The fall detection means detects a fall when the motion data output by the motion sensor satisfies a first condition relating to the user's fall possibility, and the motion data includes the first condition relating to the fall possibility. When a second condition different from the above is satisfied, a quasi-fall is detected as an event corresponding to the fall. The body motion detection means detects the user's body motion based on the motion data output by the motion sensor. The abnormality determining means determines that an abnormality has occurred in the user when the fall is detected. In addition, when the quasi-fall is detected, the abnormality determination unit monitors the detection of the body motion by the body motion detection unit for a first period after the detection, and body motion of a predetermined condition is detected during the monitoring. When it is not detected, it is determined that the abnormality has occurred in the user.

  Thereby, the abnormality detection terminal detects the user's semi-falling in distinction from the fall, and monitors the subsequent progress (body movement), so that the occurrence of an abnormal situation due to the user's fall is relatively low in impact due to the fall. Even if it can be detected. Here, the first period is, for example, 60 minutes, but is not limited thereto.

  The abnormality determination unit may determine that the abnormal state has occurred in the user when a state in which the body movement is not detected by the body movement detection unit continues for a second period in the first period. . Here, the second period is shorter than the first period, for example, 15 minutes, but is not limited thereto. The second condition is preferably a relaxed condition that is easier to detect as a fall event than the first condition.

  The abnormality determination unit may determine that the abnormality has occurred in the user when the state in which the body movement is not detected by the body movement detection unit continues for a third period, and the second period It may be set shorter than the third period. Here, the third period is, for example, 60 minutes, but is not limited thereto.

  As a result, the abnormality detection terminal detects the body movement detection monitoring period after the quasi-fall detection (the third period) rather than the period (third period) in which an abnormal state is determined in a normal time (when the tumbling is not detected) (a non-falling) By setting a short period (second period) for determining an abnormal state in the first period), for example, when the user falls down from a sitting state on a chair and cannot move, the abnormal state is quickly determined. can do. In other words, if body movement is not detected after quasi-fall detection, it is considered that the user is more likely to have some abnormality than in non-falling. Even if it is less than the determination period, it is determined as an abnormal state.

  The fall detection means detects the fall when the fall index value acquired from the motion data is equal to or greater than a first threshold, and the fall index value is equal to or greater than a second threshold that is smaller than the first threshold. However, the quasi-falling may be detected when it is less than the first threshold value.

  Here, the fall index value is, for example, an acceleration change amount or a peak value as motion data, or an evaluation value indicating a fall possibility obtained from the data, but is not limited thereto.

  The fall detection means detects the fall when a comprehensive evaluation of a plurality of different fall index values acquired by the motion data satisfies a predetermined fall criterion, and the comprehensive evaluation of the plurality of different fall index values is the The quasi-fall may be detected when a predetermined fall criterion is not satisfied and a part of the plurality of different fall index values satisfies a predetermined criterion.

  Here, the plurality of different fall index values are, for example, an acceleration change amount and a peak value as motion data, an evaluation value indicating a fall possibility based on an impact, an evaluation value indicating a fall possibility based on a drop characteristic, or the like. However, it is not limited to these.

  The abnormality determination unit may end the monitoring when the quasi-falling is detected again in the first period.

  As a result, the abnormality detection terminal determines that there is a movement (for example, exercise) in daily life that causes a relatively strong impact on the user when quasi-falling is detected a plurality of times, and it is not necessary because it is not an abnormal situation. Monitoring can be terminated.

  An abnormality detection terminal according to another embodiment of the present invention is a terminal that is carried by a user and detects the user's fall, and includes a motion sensor, a fall detection unit, a body motion detection unit, and an abnormality determination unit. . The motion sensor detects the motion of the terminal and outputs motion data. The fall detection means detects a fall when the motion data output by the motion sensor satisfies a first condition relating to the user's fall possibility, and the motion data includes the first condition relating to the fall possibility. When a second condition different from the above is satisfied, a quasi-fall is detected as an event corresponding to the fall. The body motion detection means detects the user's body motion based on the motion data output by the motion sensor. The abnormality determination unit determines that an abnormality has occurred in the user when the state in which the body movement is not detected by the body movement detection unit continues for a first period and when the fall is detected. In addition, when the quasi-falling is detected, the abnormality determination unit shortens the first period to a second period shorter than the first period.

A program according to still another aspect of the present invention provides an abnormality detection terminal having a motion sensor that detects movement of an abnormality detection terminal carried by a user and outputs movement data.
Detecting a fall when the motion data output by the motion sensor satisfies a first condition relating to the possibility of the user falling;
Detecting a quasi-falling event as an event in accordance with the falling when the motion data satisfies a second condition different from the first condition relating to the possibility of falling;
Detecting the user's body movement based on the movement data output by the movement sensor;
Determining that an abnormality has occurred in the user when the fall is detected;
When the semi-falling is detected, the detection of the body movement is monitored for a predetermined period after the detection, and it is determined that the abnormality has occurred in the user when the body movement under the predetermined condition is not detected during the monitoring. To execute the steps and.

  As described above, according to the present invention, it is possible to detect the occurrence of an abnormal situation due to a user's falling even when the impact due to the falling is relatively low.

It is a figure which shows the structure of the abnormality detection terminal which concerns on one Embodiment of this invention. It is the flowchart which showed the flow of the life monitoring process by the said abnormality detection terminal. It is the figure which showed the change of the acceleration accompanying a user's fall. It is a figure for demonstrating the detection method of an acceleration peak. It is the figure which compared and showed the change of the acceleration accompanying a user's fall, and the change of the acceleration when a user swings a bat. It is the figure which compared and showed the change of the acceleration when the said detection is canceled after a fall detection, and the change of the acceleration when the said detection is not canceled after a fall detection. It is the figure which compared and showed the change of the acceleration when a user's fall is detected, and the change of the acceleration at the time of the low impact (quasi-fall) which does not lead to fall detection. It is the flowchart which showed the flow of the fall monitoring process by the said abnormality detection terminal. FIG. 9 is a flowchart showing a flow of progress monitoring processing after detection of quasi-falling in the falling monitoring processing shown in FIG. 8.

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

[Abnormality detection terminal configuration]
FIG. 1 is a diagram illustrating a configuration of an abnormality detection terminal according to an embodiment of the present invention.

  The abnormality detection terminal 100 according to the present embodiment is, for example, a wristband type (watch type) wearable terminal, and is worn on, for example, the wrist or arm of a user such as an elderly person. In addition to the wristband type, for example, a pendant type that can be hung from the neck, a head mount type that can be attached to the head on an ear or the like, or a waist-mounted type such as a belt shape or a belt loop hanging shape can be used.

  As shown in the figure, the abnormality detection terminal 100 includes an acceleration sensor 11, a mounting sensor 12, a battery 13, a wireless communication unit 14, a notification unit 15, an operation display unit 16, an emergency button 17, and a control unit 18.

  The acceleration sensor 11 is composed of, for example, a triaxial sensor, and detects the movement of the abnormality detection terminal 100 (the movement of the user wearing the abnormality detection terminal 100). The acceleration sensor 11 outputs acceleration data detected at a predetermined sampling period. Instead of the acceleration sensor 11, other motion sensors such as an angular velocity sensor may be used.

  The wearing sensor 12 is provided, for example, in the wristband portion of the abnormality detection terminal 100, and detects the wearing state of the abnormality detection terminal 100 on the human body (arm) of the user. As a detection method of the mounting sensor 12, for example, a capacitance type is adopted, but other types of sensors such as an induction type, an ultrasonic type, a photoelectric type, and a magnetic type may be used. When the abnormality detection terminal 100 has a band shape and is connected to both ends of the anomaly detection terminal 100 and attached to the user's wrist, the connection between the ends is detected by a contact switch or an energized state. Wearing can also be detected.

  The battery 13 supplies power to each part of the abnormality detection terminal 100. As the battery 13, for example, a rechargeable battery such as a lithium polymer battery, a lithium ion battery, or a nickel metal hydride battery is used.

  The wireless communication unit 14 wirelessly communicates with a remote monitoring center C via a mobile communication network such as 3G or LTE (Long Term Evolution), for example, and an abnormality of the user detected by the abnormality detection terminal 100 (control unit 18). To the monitoring center C. The wireless communication unit 14 can also wirelessly communicate with a home security terminal connected to the monitoring center C by, for example, BLE (Bluetooth (registered trademark) Low Energy) or specific low power wireless.

  A supervisor is stationed at the monitoring center C, and when an abnormality report is received from the abnormality detection terminal 100 (or via the security terminal in the house), necessary measures such as dispatch of emergency personnel to the user's site are taken. Taken.

  The notification unit 15 is configured as a vibration device, for example, and notifies the user of an abnormality or operation acceptance by a stimulus by vibration. In addition to vibration, notification may be made by voice.

  The operation display unit 16 is configured as a touch panel display, for example, and performs various display processes such as abnormality notification together with the notification by the notification unit 15 or instead of the notification. Further, the operation display unit 16 may function as an input unit for a user cancel operation for abnormality detection (reporting). A separate cancel button may be provided as an input unit for the cancel operation.

  The emergency button 17 is an emergency operation unit for the user to make an emergency report to the monitoring center C in an emergency.

  The control unit 18 comprehensively controls each unit of the abnormality detection terminal 100 and executes various calculations in accordance with inputs from the sensors and the operation unit. The control unit 18 performs monitoring processing to determine whether a fall abnormal state, an emergency abnormal state, and a life abnormal state have occurred in the user by the monitoring processing units of the life monitoring unit 181, the emergency monitoring unit 182 and the fall monitoring unit 183. Execute.

  The life monitoring unit 181, the emergency monitoring unit 182, and the fall monitoring unit 183 may each be configured as a dedicated hardware circuit or may be configured as software (program). When the life monitoring unit 181, the emergency monitoring unit 182 and the fall monitoring unit 183 are configured as software, they are stored in a storage device such as a RAM (Random Access Memory) or a flash memory (not shown) included in the abnormality detection terminal 100. Hereinafter, these three types of monitoring processes will be described.

(Life monitoring)
Life monitoring means that the life monitoring unit 181 of the control unit 18 monitors whether the user (the wearer of the abnormality detection terminal 100) is moving to the extent that it occurs in daily life based on the output of the acceleration sensor 11, and the user If it is detected that a state where the patient cannot move due to a sudden illness or the like is detected, this is a function for reporting this.

  Specifically, the life monitoring unit 181 detects a movement (body movement) from the output of the acceleration sensor 11 in a state where the user wears the abnormality detection terminal 100 (detected by the wearing sensor 12), and performs daily life. When the level of body movement does not continuously occur for a certain period (for example, 1 hour), it is determined that a life abnormality has occurred in the user.

The determination of the presence or absence of body movement is performed by observing the peak of acceleration or the amount of change. For example, 1) When a peak value greater than or equal to a certain value (for example, 12 m / s ² ) is detected, or 2) When the amount of acceleration change per unit time exceeds a certain value, the life monitoring unit 181 Judge that there is body movement.

  Then, the life monitoring unit 181 notifies the user that it has been determined that there is a life abnormality through the vibration or sound of the notification unit 15 or the screen of the operation display unit 16, and after the notification, for a predetermined cancellation time (for example, 10 seconds). When a body movement is detected or a predetermined cancel operation is input from the user, the life abnormality is canceled (no abnormality is reported to the monitoring center C).

(Emergency monitoring)
The emergency monitoring is a function in which the emergency monitoring unit 182 of the control unit 18 notifies the monitoring center C when the user presses the emergency button 17 for the emergency response by the monitoring center C.

  Specifically, when the user presses and holds the emergency button 17 (for example, for 2 seconds), the emergency monitoring unit 182 determines that an emergency abnormality has occurred in the user. The emergency monitoring unit 182 notifies the monitoring center C of the emergency abnormality of the user and notifies the user that the abnormality has been confirmed via the notification unit 15 or the operation display unit 16.

  Even when life abnormality and fall abnormality occur in the user, when the emergency abnormality occurs, the emergency abnormality is given priority.

(Fall monitoring)
The fall monitoring is a function in which the fall monitoring unit 183 of the control unit 18 automatically detects the fall of the user (the wearer of the abnormality detection terminal 100) based on the output of the acceleration sensor 11 and reports the occurrence of the fall accident. is there.

  Specifically, the fall monitoring unit 183 detects a fall event having a characteristic like a fall from the output of the acceleration sensor 11 while the user is wearing the abnormality detection terminal 100, and evaluates the detection result. To detect falls.

  If a body movement such as a user standing up and walking is detected within a predetermined time (for example, 15 seconds) after the fall is detected, the fall monitoring unit 183 automatically cancels the detection of the fall event without confirming the fall abnormality. .

  When the fall monitoring unit 183 determines that the fall is abnormal, the fall monitoring unit 183 notifies the user of the vibration or sound of the notification unit 15 or the screen of the operation display unit 16, and after the notification, for a predetermined cancellation time (for example, 20 seconds). When a predetermined cancel operation is input from the user, the fall abnormality is canceled (no abnormality is reported to the monitoring center C). Although the detection method of a user's body movement is mentioned later, it is good also as the detection method in a life monitoring.

  In addition, although details will be described later, in the present embodiment, the fall monitoring unit 183 can distinguish and detect a normal fall and a low-impact fall (quasi-fall) based on the output of the acceleration sensor 11. The quasi-falling cannot be judged from the output of the acceleration sensor 11 to the falling, but corresponds to a case where there is a possibility of falling. Then, when a quasi-fall is detected, the fall monitoring unit 183 executes a progress monitoring process for monitoring the user's body movement for a predetermined period thereafter, so that an abnormal situation of the user due to a low impact fall can be detected early. It is possible to report an abnormality and detect it.

[Operation of anomaly detection terminal]
Next, the monitoring operation of the abnormality detection terminal 100 configured as described above will be described. In the following description, the control unit 18 or each monitoring unit of the abnormality detection terminal 100 will be described as the main operating subject, but this operation is also performed in cooperation with a program executed under the control of the control unit 18. .

(Overview of life monitoring process)
First, an outline of the life monitoring process will be described. FIG. 2 is a flowchart showing the flow of the life monitoring process.

  As shown in the figure, the life monitoring unit 181 of the control unit 18 obtains triaxial acceleration data (ax, ay, az) from the acceleration sensor 11 (step 21).

  Subsequently, the life monitoring unit 181 scalarizes the three-axis acceleration data, which is a vector (step 22).

  Subsequently, the life monitoring unit 181 cuts high-frequency components by using a moving average (low-pass filter) for the scalarized acceleration data (step 23). Note that this processing may be omitted when body motion detection is performed based on the amount of change in acceleration.

  Subsequently, the life monitoring unit 181 calculates a change amount per unit time (for example, 1 second) with respect to the scalar value of the acceleration obtained by cutting the high frequency component (step 24). Here, the acceleration data when the user moves causes a large change in the value within a short period. Therefore, the life monitoring unit 181 obtains a difference from the value of the previous acceleration data, defines the amount of change in acceleration per unit time as the sum of absolute values of the difference over the unit time, and performs the above calculation. .

  Subsequently, the life monitoring unit 181 determines whether or not the user has a body movement based on whether or not the amount of change in acceleration per unit time is a certain value or more (step 25). However, as described above, the life monitoring unit 181 determines whether or not there is body movement by detecting whether or not the peak value of acceleration is equal to or greater than a predetermined value, not the amount of acceleration change per unit time. Good.

  When it is determined that there is no body movement (No in Step 25), the life monitoring unit 181 determines whether or not the stationary state of the user whose body movement is not detected continues for a certain period (for example, 1 hour), that is, the user It is determined whether or not there has been no body movement for a certain time from the time when the time was last moved (step 26). On the other hand, when it is determined that there is a body motion (Yes in Step 25), the life monitoring unit 181 resets the count of the duration time of the stationary state and returns to Step 21 described above.

  When the life monitoring unit 181 determines that the stationary state has continued for a certain period of time (Yes in step 26), the life monitoring unit 181 determines that a life abnormality has occurred in the user, and notifies the notification unit 15 or the operation display unit 16 to that effect. To the user (step 27). On the other hand, if it is determined that the stationary time has not continued for a certain period (No in Step 26), the life monitoring unit 181 returns to Step 21 described above.

  Subsequently, the life monitoring unit 181 determines whether the user's body movement is detected within a predetermined cancellation time (for example, 10 seconds) or a predetermined cancellation operation by the user is input (step 28).

  If it is determined that body motion detection or cancellation operation input has been made within the cancellation time (Yes), the life monitoring unit 181 cancels the detection of the above life abnormality (step 31).

  On the other hand, if the body movement is not detected within the predetermined cancellation time and the cancellation time has passed without input of the cancellation operation (Yes in step 29), the life monitoring unit 181 reports a life abnormality to the monitoring center C. (Step 30).

  The life monitoring unit 181 repeatedly executes the steps described above as long as the user wears the abnormality detection terminal 100.

(Basic fall detection logic)
Next, in describing the fall monitoring process by the fall monitoring unit 183, basic logic for detecting a user's fall will be described.

  The fall monitoring unit 183 detects a fall of the user by analyzing a change in acceleration and extracting a fall event in which a characteristic that is likely to fall appears.

  For example, the fall monitoring unit 183 detects the fall by evaluating the characteristics of the impact applied to the abnormality detection terminal 100 during the fall. However, the fall monitoring unit 183 can detect the fall by evaluating the fall characteristics that occur in the process of the user falling, and comprehensively evaluate the fall likelihood based on the impact and the fall likelihood based on the fall characteristic. Thus, the final fall determination can be performed. Below, it demonstrates focusing on the example which evaluates an impact.

―Falling judgment by impact characteristics―
Similar to the life monitoring unit 181, the fall monitoring unit 183 scalarizes acceleration data output for each of the three axes and performs a moving average process (low-pass filter) as pre-processing for fall determination. This preprocessing can be shared with the life monitoring unit 181. In the following description, when calculating the acceleration change amount, the change amount may be calculated using scalarized acceleration data before the moving average process is performed.

  Since the fall involves the collision of the user's body with the ground or the like, the acceleration changes greatly due to the impact applied to the terminal. Therefore, an event that seems to fall is detected based on the acceleration change amount at the acceleration peak.

FIG. 3 is a diagram showing a change in acceleration accompanying a user's falling. As shown in the figure, the fall monitoring unit 183 first determines, from the input acceleration data, an acceleration peak (for example, (1) in the figure) that is a threshold Th _p (for example, 40 m / s ² ) or more to be processed for fall determination. ) Is detected.

  Subsequently, the fall monitoring unit 183 obtains the total acceleration change amount (total change amount H) for a predetermined period including the detected peak time (for example, 100 ms, 50 ms before and after). Specifically, a difference from the value of the immediately preceding acceleration data is obtained, and the sum of the absolute values of the differences over the predetermined time is defined as the total acceleration change amount H. Note that the timing at which the total change in acceleration is maximized does not necessarily coincide with the acceleration peak timing. Therefore, the fall monitoring unit 183 obtains the total change amount of the acceleration for the predetermined period for each acceleration sampling period, and calculates the maximum total change amount in the period near the detected peak time (for example, 50 ms before and after). The total change amount H of acceleration corresponding to the acceleration peak may be used.

Then, the fall monitoring unit 183 determines that a fall event has occurred when the calculated total change amount H is equal to or greater than a predetermined fall reference value (H ≧ Th _a ).

  FIG. 4 is a partially simplified view of FIG. 3 for explaining a method of detecting an acceleration peak when the user falls.

As shown in the figure, the fall monitoring unit 183 obtains a point Pa that falls from the threshold value Th _p . Similarly, the fall monitoring unit 183 obtains a point Pb that rises from the threshold value Th _p . A point P (peak time) at which the acceleration value is maximum between the point Pa and the point Pb is detected as an acceleration peak, and the acceleration value at the peak time is obtained as the peak value P.

―Characteristics of acceleration change during a fall―
FIG. 5 is a diagram comparing the change in acceleration accompanying the user's falling (FIG. A) and the change in acceleration when the user swings the bat (FIG. B).

  As shown in the figure, when an impact occurs such as a fall, the acceleration sensor 11 is strongly shaken, so that the output of each axis is an acceleration generated in daily life (for example, a relatively large operation such as swinging a bat). The acceleration changes more drastically than the acceleration of the acceleration. In other words, the amount of change in acceleration per unit time increases during a fall.

  Therefore, the fall monitoring unit 183 calculates the amount of change in acceleration by accumulating the change in acceleration value in a certain period before and after as the amount of change in acceleration at a certain point in the scalarized acceleration data.

  Specifically, the fall monitoring unit 183 obtains a difference (absolute value) from the immediately preceding acceleration for each sampling data of acceleration within a predetermined period, and integrates them over the predetermined period.

-Fall detection modification 1 (peak value)-
Instead of the above, the fall monitoring unit 183 may determine that a fall event has occurred by detecting an acceleration peak value P equal to or greater than a threshold value.

  Further, the fall monitoring unit 183 may obtain a fall evaluation value based on the acceleration peak value P and a fall evaluation value based on the total change amount H described above, and determine a fall event based on the total evaluation value.

  Further, the fall monitoring unit 183 may determine the fall event based on the ratio of the total change amount H to the acceleration peak value P (ratio = H / P).

-Fall detection modification 2 (drop characteristics)-
Since the fall is a phenomenon that causes a fall toward the ground, the acceleration is lower than the gravitational acceleration (9.8 m / s ² ) during the fall. Therefore, the fall monitoring unit 183 may analyze the acceleration data immediately before the peak and evaluate the free fall characteristics.

  Specifically, as shown in FIG. 3, the fall monitoring unit 183 traces the time from the peak (1) in the figure, obtains the intersection (2) with the gravitational acceleration, and from that time, for a certain period (for example, 500 ms = Acceleration average value A until the fall determination period) is calculated. (Period (2)-(3))

Then, the fall monitoring unit 183 determines that the event is a fall event when the average acceleration value A is less than a predetermined fall reference value (A <Th _b , for example, Th _b = 8.5 m / s ² ).

-Modification 3 for fall detection (combined use of impact and drop features)-
The fall monitoring unit 183 may use both the fall determination process using the impact feature and the fall determination process using the fall feature.

First, the fall monitoring unit 183 obtains an impact evaluation value Es based on the acceleration peak value P as follows.
Es = P × c (c: adjustment coefficient for comprehensive evaluation)

Subsequently, the fall monitoring unit 183 obtains a drop evaluation value Ef based on the acceleration average value A as follows.
Ef = (a−A) × b (a: fall standard constant (for example, 8.5 m / s ² ), b: adjustment coefficient)

Then, the fall monitoring unit 183 _calculates the total evaluation value Et by adding the impact evaluation value Es and the drop evaluation value Ef, and determines that the event is a fall event when the total evaluation value Et is equal to or more than the fall reference value Thc.

(Automatic cancellation determination (body motion detection))
Next, a process for automatically canceling the detection after a fall is detected as described above will be described.

  When the fall monitoring unit 183 detects a fall, if the user subsequently moves, the fall monitoring unit 183 automatically cancels the fall detection because the user is considered not in a dangerous state.

  That is, the fall monitoring unit 183 observes the presence or absence of a user's movement (body movement) during the cancellation determination time (for example, 15 seconds) from the time when the fall is detected (or peak time), and observes the acceleration peak or change amount. Judgment by. In this case, the fall monitoring unit 183 ignores the user's movement during a predetermined delay time (for example, 3 seconds) immediately after the fall is detected as a movement accompanying the fall.

Specifically, the fall monitoring unit 183 cancels the fall detection in the following cases 1) to 4), for example. In the automatic cancellation determination, any one of the following cancellation conditions may be used, and the fall detection may be canceled when any one of the conditions is satisfied using a plurality of conditions.
1) When a peak value above a certain level (for example, 12m / s ² ) is detected a predetermined number of times 2) When the sum of the above peak values exceeds a predetermined value 3) The total value obtained by accumulating the acceleration change amount exceeds a certain value 4) When the amount of change in acceleration per unit time exceeds a certain value (similar to life monitoring above)

  On the other hand, if the user's body movement is not detected before the cancellation determination time elapses after the fall detection, the fall monitoring unit 183 determines the fall abnormality.

  FIG. 6 shows a comparison between a change in acceleration when the detection is canceled after the fall detection (FIG. A) and a change in acceleration when the detection is not canceled after the fall detection (FIG. B). FIG.

  As shown in FIG. 5A, when it is canceled after the fall is detected, a change in acceleration of a certain level or more can be seen even after the acceleration peak value. In other words, in this case, it is assumed that the user is living in the same way as before the fall after the fall, so it may be possible to cancel the fall detection.

  On the other hand, as shown in Fig. B, if the detection is not canceled after the fall is detected (abnormal report), an acceleration change occurs before the acceleration peak value after the acceleration peak value. You can see that it is not. That is, in this case, it is estimated that the user cannot move as it is after the fall, so it is considered that the user should report the abnormality without canceling the fall detection.

(Low impact fall event (quasi-fall) detection and progress monitoring process)
Next, detection of a low-impact fall event (quasi-fall) and subsequent progress monitoring processing will be described.

  FIG. 7 shows a comparison between the change in acceleration when the user's fall is detected by the above-described logic (A in the same figure) and the change in acceleration at the time of occurrence of a low impact that does not lead to the fall detection (B in the same figure). FIG. As shown in the figure, for example, when a fall is detected based on the peak value of acceleration, even if a low impact as shown in FIG. B occurs, the peak value of acceleration does not reach the threshold Th, and a fall is detected. Will not be. This is because, for example, even when the user falls from a sitting state on a chair or when the user falls from a standing state, the user's body collides with the ground before the abnormality detection terminal 100, thereby reducing the impact. It can happen in such cases.

  Therefore, the fall monitoring unit 183 detects the fall due to the low impact as a quasi-fall, as distinguished from the fall, and after the quasi-fall is detected, whether or not body movement at the daily life level occurs at a predetermined frequency for the user. By executing the monitoring process, an event that does not result in the fall detection but leaves the possibility of a fall is remedied. This progress monitoring process is time-limited monitoring for a fixed time (for example, 1 hour) after canceling the fall detection.

  That is, the fall monitoring unit 183 detects a fall event when the acceleration data satisfies the first condition related to the user's fall possibility, and when the acceleration data satisfies the second criterion related to the user's fall possibility. Detect quasi-fall events. This second criterion is preferably a detection criterion that is more relaxed than the first criterion.

Examples of the trigger condition for the progress monitoring process after the detection of the quasi-falling (when the second condition is satisfied) include the following 1) to 3).
1) Although the index value for the fall determination (the total change amount H of acceleration described above per predetermined time, the peak value P, or the total evaluation value Et by the combined use of the impact feature and the drop feature) is less than the fall reference value, the fall reference value When it is greater than or equal to the smaller relaxation standard value (semi-tumbling standard value).
For example, the total change amount H <Th1, and H ≧ Th2 (Th2 <Th1)
2) When an index value (for example, the amount of change H) used for the fall determination does not satisfy the fall standard, and another index value (for example, the peak value P) not used for the fall determination satisfies the fall standard.
3) When a fall is detected by a composite evaluation such as the above-mentioned total evaluation value Et, the total evaluation value is less than the fall standard, but some (individual) evaluation values / index values are above a certain standard (For example, in the above-described example, the total evaluation value Et does not satisfy the fall standard, but only one of the impact evaluation value Es based on the acceleration peak value P or the fall evaluation value Ef based on the acceleration average value A satisfies the predetermined fall standard) .

  Hereinafter, a specific flow of the fall monitoring process including the progress monitoring process after the semi-fall is detected will be described. In the following, a case where 1) of 1) to 3) described above is used as an example of the quasi-falling detection method will be described as an example, but of course, 2) and 3) may be used as the detection method.

  FIG. 8 is a flowchart showing the flow of the fall monitoring process by the fall monitoring unit 183.

  As shown in the figure, as described above, the fall monitoring unit 183 evaluates the acceleration data for each acquisition period of the acceleration data output from the acceleration sensor 11, and sets the index value to the fall reference value or the quasi-standard value. It is compared with the fall reference value (step 81).

  If the fall monitoring unit 183 determines that the index value is equal to or greater than the fall reference value (Yes in Step 82), the fall monitoring unit 183 sets the fall flag to ON (Step 83).

  Subsequently, as described above, the fall monitoring unit 183 determines whether or not the user's body movement is detected within a cancellation time (for example, 15 seconds) after the fall is detected (step 84). In this case, the user's movement for a predetermined delay time (for example, 3 seconds) immediately after the fall detection is ignored as the movement accompanying the fall.

  The fall monitoring unit 183 confirms the fall abnormality and reports the abnormality to the monitoring center C (step 86) when the cancellation time has elapsed without detecting the body movement within the cancellation time (Yes in step 85). Although not shown in the figure, as described above, the fall monitoring unit 183 notifies the user that the fall abnormality has been detected after the above-described step 85 by the notification unit 15 or the operation display unit 16, and is predetermined. If the cancel operation is input from the user within the cancel operation time (for example, 10 seconds or 20 seconds), the overturn abnormality is canceled, and if the cancel operation is not input within the cancel time, the overturn abnormality is determined.

  If the fall monitoring unit 183 determines that body movement is detected within the cancellation time (Yes in Step 84), the fall detection is canceled (Step 89), and the process returns to the fall monitoring process in Step 81 and thereafter.

  On the other hand, if it is determined in step 82 that the index value is less than the fall reference value (No), the fall monitoring unit 183 determines whether the index value is equal to or greater than the quasi-fall standard value (step 87). ).

  If the fall monitoring unit 183 determines that the index value is equal to or greater than the quasi-falling reference value (Yes), the fall monitoring unit 183 performs a progress monitoring process on whether or not a daily level of body movement is detected by the user (step 88).

  The detection method of the body movement is the same as the detection method by the life monitoring unit 181. However, the time for determining non-detection of body movement (the duration of the stationary state) is a time (for example, 15 minutes) shorter than the determination time (for example, 1 hour) in the life monitoring. In other words, this progress monitoring process executes a life monitoring process in which the determination time concerning the continuation of non-detection of body movement is shortened. Note that the body movement detection method in the progress monitoring process is different from the life monitoring process and may be the same method as the body movement detection in the automatic cancellation determination of the fall monitoring unit 183 described above.

  FIG. 9 is a flowchart showing the flow of the progress monitoring process after the quasi-fall detection shown in step 88 of FIG.

  As shown in the figure, the fall monitoring unit 183 detects the user's body movement based on the acceleration data when the semi-fall is detected (step 91).

  Subsequently, the fall monitoring unit 183 determines whether body movement is detected (step 92).

  If it is determined that body movement is detected (Yes), the fall monitoring unit 183 determines whether or not an elapsed monitoring time (for example, 1 hour) has elapsed (step 95).

  The fall monitoring unit 183 ends the progress monitoring process when the elapsed monitoring time has elapsed with the body motion detected (Yes in step 95), and the normal fall monitoring process (after step 81 in FIG. 8). Return to.

  On the other hand, when it is determined that the state (stationary time) in which no body movement is detected continues for a predetermined determination time (for example, 15 minutes) (Yes in Step 93), the fall monitoring unit 183 determines the fall abnormality and the monitoring center C Anomaly is reported to (step 94).

  Here, the fall abnormality determined after the semi-fall detection may be output separately from the case where the fall abnormality is confirmed after the fall detection.

[Summary]
As described above, according to the present embodiment, the fall monitoring unit 183 of the abnormality detection terminal 100 detects the user's semi-falling in distinction from the fall, and even when the semi-falling is detected, the user's body is kept for a predetermined period. The detection of movement is monitored, and when a stationary time body movement shorter than the stationary time in the life monitoring process is not detected, it is determined that the fall is abnormal. For this reason, the occurrence of an abnormal situation due to the user's falling can be detected even when the impact due to the falling is relatively low.

[Modification]
The present technology is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present technology.

  In the above-described embodiment, the fall monitoring unit 183 performs the progress monitoring process after the quasi-fall detection until the elapsed monitoring time (for example, 1 hour) elapses. However, if a firm body movement (for example, acceleration at the walking level of the user) is detected within a predetermined time (for example, 30 seconds) after the quasi-fall is detected, the progress monitoring process may not be executed.

  Further, when the fall monitoring unit 183 detects the semi-falling again during the progress monitoring process after the semi-falling detection, the fall monitoring unit 183 may end the progress monitoring process and return to the normal monitoring. In such a case, it is possible to determine that there is a movement (for example, exercise) in daily life that causes a relatively strong impact on the user and that it is not an abnormal situation, and thus it is possible to continue unnecessary progress monitoring processing. Can be prevented.

  The above embodiment has been described on the assumption that the abnormality detection terminal 100 has a life monitoring function by the life monitoring unit 181. However, the abnormality detection terminal 100 may not have the life monitoring function.

  In the above-described embodiment, the life monitoring process by the life monitoring unit 181 and the progress monitoring process by the fall monitoring unit 183 have been described as being executed in parallel. However, the life monitoring unit 181 executes the progress monitoring process. It is good. In this case, the life monitoring unit 181 normally makes an abnormality determination when the user's stationary time continues for a predetermined time (for example, 1 hour). However, if a quasi-fall is detected that does not satisfy the fall detection criterion, the life monitoring unit 181 The life monitoring process is executed by shortening the time (for example, 15 minutes). Further, the life monitoring unit 181 and the fall monitoring unit 183 may be configured as one processing unit.

  In the above-described embodiment, the abnormality detection terminal 100 is configured to include various sensors, a notification unit, an operation display unit, a wireless communication unit, and the like as a wearable terminal. However, at least a part of these units of the abnormality detection terminal 100 is used. May be provided in another device (for example, a mobile terminal such as a smartphone carried by the user), and the same processing as in the above-described embodiment may be executed by cooperation processing with the other device.

11 ... Acceleration sensor 12 ... Mounting sensor 12
14 ... Wireless communication unit 14
DESCRIPTION OF SYMBOLS 18 ... Control part 181 ... Life monitoring part 182 ... Emergency monitoring part 183 ... Fall monitoring part 100 ... Abnormality detection terminal C ... Monitoring center

Claims (6)

In an abnormality detection terminal that is carried by a user and detects the user's fall,
A motion sensor that detects the motion of the abnormality detection terminal and outputs motion data;
When the motion data satisfies a first condition related to the user's possibility of falling, a fall is detected, and when the motion data satisfies a second condition different from the first condition related to the possibility of falling A fall detection means for detecting a quasi-fall as an event similar to a fall;
Body movement detection means for detecting the body movement of the user based on the movement data;
When the fall is detected, it is determined that an abnormality has occurred in the user. When the semi-fall is detected, the body movement detection unit detects the body movement for the first period after the detection. An abnormality determination means for determining that the abnormality has occurred in the user when body movement under a predetermined condition is not detected during the monitoring ,
In the first period, the abnormality determination unit is in a state where the body movement detection unit does not detect the body movement for a second period, and the body movement detection unit does not detect the body movement. An abnormality detection terminal that determines that the abnormality has occurred in the user when the operation continues for a third period longer than the second period . In an abnormality detection terminal that is carried by a user and detects the user's fall,
A motion sensor that detects the motion of the abnormality detection terminal and outputs motion data;
When the motion data satisfies a first condition related to the user's possibility of falling, a fall is detected, and when the motion data satisfies a second condition different from the first condition related to the possibility of falling A fall detection means for detecting a quasi-fall as an event similar to a fall;
Body movement detection means for detecting the body movement of the user based on the movement data;
When the fall is detected, it is determined that an abnormality has occurred in the user. When the semi-fall is detected, the body movement detection unit detects the body movement for the first period after the detection. And when the body motion of a predetermined condition is not detected during the monitoring, it is determined that the abnormality has occurred in the user, and the monitoring is performed when the quasi-falling is detected again in the first period. An abnormality detection terminal comprising: an abnormality determination unit that terminates the operation . The abnormality detection terminal according to claim 1 or 2 ,
The fall detection means detects the fall when a fall index value acquired by the motion data is equal to or greater than a first threshold, and the fall index value is equal to or greater than a second threshold that is smaller than the first threshold. An abnormality detection terminal that detects the quasi-falling when it is less than the first threshold. The abnormality detection terminal according to any one of claims 1 to 3,
The fall detection means detects the fall when a comprehensive evaluation of a plurality of different fall index values acquired by the motion data satisfies a predetermined fall criterion, and the comprehensive evaluation of the plurality of different fall index values is the An abnormality detection terminal that detects the quasi-fall when the predetermined fall criterion is not satisfied and a part of the plurality of different fall index values satisfies the predetermined criterion. In the abnormality detection terminal having a motion sensor that detects the movement of the abnormality detection terminal carried by the user and outputs movement data,
Detecting a fall when the motion data satisfies a first condition regarding the user's fall possibility;
A step of detecting a quasi-fall as an event in accordance with the fall when the motion data satisfies a second condition different from the first condition relating to the possibility of the fall;
Detecting the user's body movement based on the movement data;
Determining that an abnormality has occurred in the user when the fall is detected;
When the quasi-falling is detected, the detection of the body movement is monitored for the first period after the detection, and the abnormality occurs in the user when the body movement under a predetermined condition is not detected during the monitoring. A program for executing the step of determining that
The step of determining includes a state in which the body movement detection unit does not detect the body movement in the first period, and a state in which the body movement detection unit does not detect the body movement. Determines that the abnormality has occurred in the user when the user continues for a third period longer than the second period.
Program . In the abnormality detection terminal having a motion sensor that detects the movement of the abnormality detection terminal carried by the user and outputs movement data,
Detecting a fall when the motion data satisfies a first condition regarding the user's fall possibility;
A step of detecting a quasi-fall as an event in accordance with the fall when the motion data satisfies a second condition different from the first condition relating to the possibility of the fall;
Detecting the user's body movement based on the movement data;
Determining that an abnormality has occurred in the user when the fall is detected;
When the quasi-fall is detected, the detection of the body movement is monitored for a predetermined period after the detection, and the abnormality occurs in the user when the body movement under a predetermined condition is not detected during the monitoring. A determining step ;
A program for executing the step of ending the monitoring when the quasi-falling is detected again during the predetermined period .

Images