JP6901942B2 - Equipment abnormality cause estimation system - Google Patents

Description

The present invention relates to an abnormality cause estimation system that estimates whether or not a device abnormality is affected by an abnormal operation of a person.

Automatic doors, ticket gates, ATMs, elevators, and the like are devices that interact with people, and are set to respond normally when a person performs an expected operation.

For example, in the case of an automatic door, when a person approaches from a predetermined direction at a predetermined speed, the door is opened before the person arrives and the door is closed after the person passes through, which is a normal operation. However, if you approach the door while running, the door may not open before the person arrives, and the person may collide with the door. In addition, the automatic door detection of people often assumes that a person will go straight and enter, and if a person enters from the side, the response to open the door will be delayed, and there is a possibility that the person will come into contact with the door. ..

In this way, in a device that interacts with a person, the sensing area and sensing method are determined by assuming the movement pattern of the person, and the device responds based on the result, so that the person performs an unexpected operation. In that case, the device may not be able to perform the desired operation.

In such a case, an abnormal load may be applied to the device. For example, in the case of an automatic door, when a person comes into contact with it, the contact force becomes resistance and the work required to open and close the door may increase, or the automatic door may be damaged due to the influence of contact with a person and the work required for opening and closing the door thereafter may occur. May increase. Especially in the latter case, it may be necessary for the device manager to adjust the device or replace parts, so it is necessary to contact the device manager immediately.

On the other hand, other causes of increased resistance when opening and closing the automatic door are the possibility that the resistance of the moving part is increased by small foreign substances such as dust and dust, and that foreign substances such as cloth are caught in the opening and closing part of the door. Therefore, there is a possibility that the resistance of the moving part is increasing. If these are the causes, the obstacle can be resolved by cleaning the equipment or removing foreign matter. Therefore, it may be sufficient to urge the building manager or the like to clean the equipment without contacting the equipment manager.

In this way, even if there is an abnormality in the operation of the device, the measures for recovery differ depending on the cause. In other words, if a malfunction occurs due to human movement, the equipment may be damaged, so highly specialized measures such as adjustment and parts replacement are required, whereas foreign matter is simple. Cleaning may be all that is needed.

Therefore, it can be said that the response to recovery from the same equipment abnormality event can be estimated to some extent by determining whether or not a person has affected the abnormality. As a method for determining the presence or absence of the influence of a person on an equipment abnormality, Patent Document 1 discloses a plant monitoring device that excludes the influence of normal operation and sets a monitoring standard.

Japanese Unexamined Patent Publication No. 11-95833

In the plant monitoring device of Patent Document 1, the presence or absence of the influence of human operation on the equipment abnormality is determined based on the past operation data input via the input device or the like. However, there are cases where the influence of humans cannot be identified by the magnitude or pattern of fluctuations in the data of the accessory sensor of the device.

In particular, in equipment such as automatic doors and ticket gates in the city, unlike plants, the human motion detection sensors installed in the equipment are limited, and the presence or absence of human influence on equipment abnormalities based only on the information inside the equipment. Is difficult to evaluate.

In this way, existing devices are equipped with the sensors necessary to achieve the designed operation, but the environment in which the devices are installed is diverse and not sufficient for sensing the surrounding environment. There is a possibility. In particular, human movements are diverse, and there is a possibility that abnormal human movements cannot be detected within the range that can be grasped by the device alone.

Therefore, in the present invention, by estimating the cause of the abnormality of the device after grasping the movement of the person around the device, it is possible to easily and accurately grasp the presence or absence of the influence of the person on the device abnormality. The purpose is to provide an estimation system.

In order to solve the above problems, the abnormality cause estimation system of the present invention estimates the cause of the abnormality that has occurred in the device, and detects the abnormal operation based on the output of the motion detection sensor that detects the motion of a person. Based on the output of the abnormal operation detection unit and the device accessory sensor that detects the status of the device, the device abnormality detection unit that detects the device abnormality, the output of the abnormal operation detection unit, and the output of the device abnormality detection unit Based on this, a person-induced abnormality determination unit for determining the presence or absence of the influence of the abnormal operation on the device abnormality is provided , and the person-induced abnormality determination unit includes the device abnormality occurrence time detected by the device abnormality detection unit and the abnormal operation. comparing the abnormal operation occurrence time detected by the detecting unit, the abnormal operation occurrence time when preceding the apparatus abnormality occurrence time, and a shall be determined that the effect of the device abnormally malfunction.

According to the present invention, since the cause of the abnormality of the device is estimated after grasping the movement of the person around the device, it is possible to easily and accurately grasp the presence or absence of the influence of the person on the device abnormality. As a result, it becomes easy to take appropriate measures according to the cause of the device abnormality, and it is possible to shorten the recovery time of the device.

It is a figure which shows the outline of the abnormality cause estimation system of Example 1. FIG. It is a figure which shows the structure of the analyzer of Example 1. FIG. It is a figure which shows the structure of the determination device of Example 1. FIG. It is a figure which shows the other structure of the determination apparatus of Example 1. FIG. It is a figure which shows the outline of the abnormality cause estimation system of Example 2.

Hereinafter, examples will be described with reference to the drawings.

FIG. 1 is a schematic view of an abnormality cause estimation system 100 according to a first embodiment of the present invention. This abnormality cause estimation system 100 targets the automatic door as a monitoring target, and when an abnormality occurs in the automatic door, it estimates whether or not the cause is influenced by a person, and the estimation result or the like is used as a building manager or the like. Is to notify.

Although not shown, the abnormality cause estimation system 100 of this embodiment includes hardware such as a computing device such as a CPU, a main storage device such as a semiconductor memory, an auxiliary storage device such as an HDD or SSD, and a communication device. And they are interconnected. Further, a program corresponding to the analyzer 6 and the like described later is loaded in the main storage device, and the abnormal operation database 8 and the like described later are recorded in the auxiliary storage device. With this configuration, each function of the abnormality cause estimation system 100, which will be described later, is realized by executing the program while referring to the database by the arithmetic unit.

First, the general operation of the automatic door of FIG. 1 will be described. This automatic door is an automatic door whose center is opened by sliding the door panel 1a to the left and the door panel 1b to the right, and the sensors 2a and 2b for detecting the approach of a person and the door panels 1a and 1b are provided. It includes a driver 4 that opens and closes, and a controller 3 that controls the drive 4. With this configuration, when the sensors 2a and 2b provided on the upper part of the closed end detect the approach of a person, the controller 3 provided in the automatic door controls the operation of the drive 4 to press the door panels 1a and 1b. Slide left and right to open the automatic door.

The surveillance cameras 5a and 5b capture the movements of people around the automatic door, and the video signals output by them are input to the surveillance terminal 7. The monitoring terminal 7 executes a monitoring task such as displaying an image on the monitor of the building manager's room based on the input video signal. Since the surveillance cameras 5a and 5b and the surveillance terminal 7 have already been used in many buildings and the like, detailed description thereof will be omitted.

The video signals output by the surveillance cameras 5a and 5b are also input to the analyzer 6. The analyzer 6 detects an abnormal operation of a person based on the input video signal and the abnormal operation model acquired from the abnormal operation database 8.

The determination device 9A is represented by an output signal (observed value expressed by an analog signal such as a current value or a voltage value, or a digital signal) related to the sensors 2a and 2b and the driver 4 input from the controller 3. The abnormality of the automatic door is detected based on the command value, etc.) and the device diagnosis model acquired from the device diagnosis database 10. For the determination device 9A, these output signals are signals for detecting the status of the automatic door, so that the sensors 2a and 2b, the controller 3, the driver 4, and the like are also incidental sensors for the automatic door.

Further, the determination device 9B determines from the output signals of the analyzer 6 and the determination device 9A whether or not the cause of the abnormality of the automatic door is an abnormal operation of a person.

Specifically, when an abnormality in door opening / closing is detected by the determination device 9A, the determination device 9B determines whether or not there is an abnormal operation of a person near the time when the door abnormality is detected based on the output signal of the analyzer 6. To do. If there is no abnormal operation of the person near the time when the abnormal door is detected, the building manager is notified as a door abnormality that is not caused by the abnormal operation of the person, and the person is notified at the time when the abnormal door is detected. If there is an abnormal operation, the possibility that the abnormal operation of a person is the cause is given and the door abnormality is notified to the building manager or the like. Upon receiving this, the building manager or the like can recover from the door abnormality at an early stage by performing the recovery work according to the content of the notification.

Next, the details of the processing contents in the analyzer 6 will be described with reference to FIG. As described above, the analyzer 6 detects the abnormal operation of a person based on the input video signal and the abnormal operation model acquired from the abnormal operation database 8. In order to realize this, the analyzer 6 includes a human detection module 6a, an operation trace module 6b, and an abnormal operation determination module 6c.

When the video signal is input from the surveillance cameras 5a and 5b, the person detection module 6a detects a person in the video by comparing the unmanned video with the captured video. The motion trace module 6b sequentially traces the detected movement of the person and generates data of the movement of the person (hereinafter, referred to as "trace data"). The abnormal operation determination module 6c compares the abnormal operation model and the trace data stored in the abnormal operation database 8 in advance, and if they match, the abnormal operation together with the matched abnormal operation pattern (for example, rushing, pinching, leaning). Output the occurrence time. The abnormal operation pattern and the abnormal operation occurrence time output here are stored in the auxiliary storage device each time the abnormal operation is detected, and can be referred to from the arithmetic unit at any time.

Here, the trace data is the position, velocity, acceleration, and jerk of a person in the image in a plane. When a distance image is used, the position, velocity, acceleration, and jerk may be set in three dimensions.

In this way, in the analyzer 6, by comparing the abnormal operation model acquired from the abnormal operation database 8 with the position and speed of the person in the trace data, the position of the person and the position of the person can be determined from the various movements of the person included in the video. Abnormal speed can be extracted and desired abnormal operation can be detected with high accuracy.

Next, the details of the processing contents in the determination device 9 will be described with reference to FIG. The determination device 9 is a combination of the determination device 9A and the determination device 9B shown in FIG. 1 for convenience.

As described above, the determination device 9 detects the door abnormality based on the output signal of the analyzer 6, the output signals of the sensors 2a and 2b and the driver 4, and the device diagnosis model acquired from the device diagnosis database 10. , It is for determining whether or not the cause of the door abnormality is an abnormal operation of a person. In order to realize this, the determination device 9 includes a signal value determination module 9a, a state determination module 9b, and an abnormal operation exclusion determination module 9c.

First, the signal value determination module 9a determines whether or not the control signal input from the controller 3 and the input signals such as the sensor signal and the driver signal are within a predetermined range. If these input signals are within the predetermined range, it is determined that no device abnormality has occurred. On the other hand, if the input signal deviates from the predetermined range, it is determined that a device abnormality may have occurred.

When there is a possibility that an equipment abnormality has occurred, the state determination module 9b compares the input signal with the drive device signal or the like (equipment diagnosis model) at the time of equipment abnormality stored in the equipment diagnosis database 10 in advance, and confirms whether the equipment condition is normal. If it is abnormal or abnormal, the device abnormality pattern (for example, damage to the door panel, damage to the door sensor, damage to the drive) is determined. Then, when the device state is determined to be abnormal, the state determination module 9b outputs the device error occurrence time together with the device error pattern.

In this way, in the state determination module 9b, by comparing the device diagnosis model acquired from the device diagnosis database 10 with the output signals of the signal value determination module 9a, desired device abnormalities can be obtained with high accuracy from various output signals. Can be detected.

After that, the abnormal operation exclusion determination module 9c compares the abnormal operation occurrence time output by the analyzer 6 with the device abnormality occurrence time output by the state determination module 9b, and confirms whether or not there is a relationship between the device abnormality and the abnormal operation. Specifically, if the time difference between the device abnormality and the abnormal operation is smaller than a predetermined value, or if the abnormal operation precedes the device abnormality, it is determined that the cause of the device abnormality is a human abnormal operation. On the other hand, if the above criteria are not met, it is determined that the cause of the device abnormality is not the abnormal operation of a person.

At this time, the type of the analyzer 6 to be referred to and the time range regarded as being in the vicinity differ depending on the pattern of the equipment abnormality. For example, if the output of the drive 4 increases momentarily and then returns to the original value immediately after that, it is possible that a person touches the automatic door only at the moment when the output increases, so that the time is the same as the increased time. Determines whether or not a person stays near the automatic door. Also, if the sensor constantly responds, it is possible that a person or object has been placed in the sensor detection range for a long time, so people may stay in the sensor detection range of the automatic door. , Judges whether or not there is an operation for a person to place an object at a time before the device abnormality occurrence time.

Further, as shown in FIG. 4, the cause estimation module 9d, the correspondence estimation module 9e, the abnormality cause database 10a, and the abnormality response database 10b may be added to the configuration of FIG. Here, the abnormality cause database 10a is a database that records the cause of the abnormality having a causal relationship with the combination of the abnormal operation pattern and the device abnormality pattern, and the abnormality response database 10b records the recovery work to be taken for the device abnormality pattern. It is a database that has been created.

The cause estimation module 9d acquires the abnormality cause corresponding to the combination of the device abnormality pattern output by the state determination module 9b and the abnormality operation pattern output by the abnormality operation determination module 6c from the abnormality cause database 10a. After that, the response estimation module 9e acquires the recovery work from the device abnormality corresponding to the device abnormality pattern output by the status determination module 9b from the abnormality response database 10b, and issues it to the building manager or the like as a specific work instruction. It is to inform.

In the configuration of FIG. 3, the building manager had to interpret the meaning of the alarm of the judgment device 9 and judge the appropriate response by himself, but according to the configuration of FIG. 4, unskilled building management Even a person can take an appropriate response to a device abnormality only by following the specific work instruction indicated by the determination device 9.

In this embodiment, in order to detect an abnormal movement of a person, a surveillance camera is used and based on the movement in the image, but other methods may be used as long as the abnormal movement can be detected. For example, a motion sensor may be used to determine the presence or absence of abnormal movement of a person based on the response, or thermography may be used to determine the presence or absence of abnormal movement of a person based on fluctuations in the amount of heat. ..

According to the present embodiment described above, it is possible to distinguish between those caused by abnormal human operation and those not caused by abnormal device operation, so that it is easy to take appropriate measures according to the cause of the device abnormality. , It is possible to shorten the recovery time from equipment abnormality.

FIG. 5 is a schematic view of the abnormality cause estimation system 101 according to the second embodiment of the present invention. The abnormality cause estimation system 101 monitors the ticket gate, and when an abnormality occurs in the ticket gate, it estimates whether or not the cause is an abnormal operation of a person. It should be noted that the common points with the first embodiment will be omitted.

In this embodiment, surveillance cameras 5a and 5b are present for the ticket gate 11. The operation signal of the ticket gate 11 is input to the determination device 9A via the controller 3, and the presence or absence of a device abnormality is determined based on the device diagnosis database 10. Further, the signals of the surveillance cameras 5a and 5b are input to the analyzer 6 and analyze the presence or absence of abnormalities in people and human flow based on the abnormal operation database 8. The output signals of the analyzer 6 and the judgment device 9A are input to the judgment device 9B, and it is estimated whether the generated abnormality is due to the device itself or the interaction with a person, and the estimation result is sent to the station staff or the like. Report.

As described above, examples of automatic doors and ticket gates have been described, but the present invention can be applied to other devices as long as they are devices that interact with humans. For example, in the case of an ATM, an abnormal operation of a person may be detected by using both a surveillance camera installed in a building and a camera provided in the ATM. In this case, not only the operation during the operation of the ATM but also the abnormality of the behavior before and after going to the ATM is detected, it is determined whether or not the person who detected the abnormal operation operated the operation, and the information is given and the alarm is issued. To do.

In addition, when applied to an elevator, surveillance cameras installed in the building, security gate opening / closing information and passerby attribute information, and building door opening / closing status can also be used to detect abnormal human movements. For example, if there are many people who pass through the security gate with temporary permission, which is different from normal users, there is a possibility that some event may be used by people who are not familiar with the elevator. As a result, the amount of elevator boarding may change.

In addition, although there are few people with temporary permission, if the same person passes through the gate many times, or if the door of the luggage entrance is opened and closed before and after that, moving or carrying in luggage is being done. There is a possibility, and as a result, an event such as hitting a luggage against an elevator may occur, and an abnormality may be detected.

In the above embodiment, the example of the present invention in a specific device is shown, but the present invention can be applied to a device in which a person exists around the device and interaction with the surrounding people occurs. Is. Depending on the type of device, the range of people around the device may be taken into consideration when performing sensing. For example, in the case of the automatic door in the above embodiment, the walking speed and walking pattern change depending on the attributes of the passersby (businessman, carrier, cleaning staff, elderly, children), so the ID card is used for authentication in advance. If it is possible to predict what kind of attribute a person will pass through, the abnormal operation database may be classified accordingly and a judgment may be made based on the result. For example, if the cleaning staff is moving erratically around the door, it is unlikely to come into contact with the door, but if the child is moving erratically around the door, it is likely to come into contact with the door. Therefore, the processing in the determination device may be switched according to the attribute.

100, 101 Abnormal cause estimation system 1a, 1b Door panel,
2a, 2b sensor,
3 controller,
4 drive,
5a, 5b surveillance cameras,
6 analyzer,
6a Human detection module,
6b motion trace module,
6c Abnormal operation judgment module,
7 Monitoring terminal,
8 Abnormal operation database,
9, 9A, 9B Judgment device,
9a Signal value judgment module,
9b Status judgment module,
9c Abnormal operation exclusion judgment module,
9d cause estimation module,
9e compatible estimation module,
10 Device diagnostic database,
10a Abnormal cause database,
10b Anomaly response database,
11 Ticket gate

Claims (5)

An abnormality cause estimation system that estimates the cause of an abnormality that has occurred in a device.
An abnormal motion detection unit that detects abnormal motion based on the output of the motion detection sensor that detects human motion,
Based on the output of the device accessory sensor that detects the status of the device, the device abnormality detection unit that detects the device abnormality and the device abnormality detection unit
It is provided with an output of the abnormal operation detection unit and a person-induced abnormality determination unit that determines the presence or absence of the influence of the abnormal operation on the device abnormality based on the output of the device abnormality detection unit.
In the person-induced abnormality determination unit,
The device abnormality occurrence time detected by the device abnormality detection unit and
The abnormal operation occurrence time detected by the abnormal operation detection unit and
Comparing, when these time difference is smaller than a predetermined value, abnormal cause estimation system that determining means determines that there is the influence of the device abnormally malfunction. An abnormality cause estimation system that estimates the cause of an abnormality that has occurred in a device.
An abnormal motion detection unit that detects abnormal motion based on the output of the motion detection sensor that detects human motion,
Based on the output of the device accessory sensor that detects the status of the device, the device abnormality detection unit that detects the device abnormality and the device abnormality detection unit
It is provided with an output of the abnormal operation detection unit and a person-induced abnormality determination unit that determines the presence or absence of the influence of the abnormal operation on the device abnormality based on the output of the device abnormality detection unit.
In the person-induced abnormality determination unit,
The device abnormality occurrence time detected by the device abnormality detection unit and
The abnormal operation occurrence time detected by the abnormal operation detection unit and
Comparing the abnormal operation occurrence time when preceding the apparatus abnormality occurrence time, abnormal cause estimation system that determining means determines that there is the influence of the device abnormally malfunction. An abnormality cause estimation system that estimates the cause of an abnormality that has occurred in a device.
An abnormal motion detection unit that detects abnormal motion based on the output of the motion detection sensor that detects human motion,
Based on the output of the device accessory sensor that detects the status of the device, the device abnormality detection unit that detects the device abnormality and the device abnormality detection unit
It is provided with an output of the abnormal operation detection unit and a person-induced abnormality determination unit that determines the presence or absence of the influence of the abnormal operation on the device abnormality based on the output of the device abnormality detection unit.
In addition, it has an abnormality cause database that records the causes of abnormalities that are causally related to the combination of abnormal operation patterns and equipment abnormality patterns.
The person-induced abnormality determination unit acquires an abnormality cause corresponding to a combination of the equipment abnormality pattern output by the equipment abnormality detection unit and the abnormality operation pattern output by the abnormality operation detection unit from the abnormality cause database. Te, abnormal cause estimation system that is characterized in that alarm. In addition, it has an anomaly response database that records the recovery work to be taken for equipment anomaly patterns.
The person caused the abnormality determining unit, the restoring work corresponding to the device failure pattern in which the apparatus abnormality detection unit has output, is obtained from the anomaly database, abnormality of claim 3, characterized in that the alarm Cause estimation system. It said motion detection sensor, surveillance cameras, motion sensor, abnormality cause estimation system according to any one of claims 1 to 4, characterized in that any one of thermography.

Images