KR102454411B1 - Method for Disaster Detection Based on Artificial Intelligence Using Railbot and Building Automation System for Disaster Response and the System therefor - Google Patents

Abstract

The present invention relates to an automatic building control method using a building automation system (BAS) server (401) and a railbot server (402) capable of communicating with a railbot (100) capable of moving along a rail (200) provided in a building comprising a facility temperature detection function, a human detection function, a fire detection function, a flame detection function, a smoke detection function, a flood detection function, a facility detection function, and the like. Therefore, the present invention is capable of enabling to detect various disasters such as dangerous people, fire, smoke, flame, flooding, and the like.

Description

Method for Disaster Detection Based on Artificial Intelligence Using Railbot and Building Automation System for Disaster Response and the System therefor

The present invention relates to a technology in which robot technology, building control technology, and advanced IT technology are fused. Specifically, it relates to software and a system for automatically controlling a building according to the detection of a rail bot, which is a robot that can move along rails.

The internal and external environment of a building changes in various ways, and operating multiple facilities in a building according to these changes provides a comfortable environment for residents in residential and commercial buildings, and optimizes the production rate of major devices located in buildings in plants, etc. essential to do In addition, the necessity of sensing the environment inside and outside the building and smart control of the facilities in the building is also necessary for the enlargement and complexity of buildings, the emergence of various controllable facilities according to the development of technology, solving the problem of optimizing the energy efficiency of the building, and strengthening the security problem in the building. this is getting bigger A technology that controls a building from an energy point of view is referred to as BEMS (Building Energy Management System). In general, a technology that automatically detects and controls a building from various points other than the energy point while including some of the functions of BEMS is BAS (Building Automation System). system).

BAS is a system in which the central detection and control facility is based on the BAS server, and various facilities in the building, including sensors, are communicatively connected. Facilities in buildings include air conditioners, boilers, temperature and humidity sensors, power/peak controllers, lighting, heat exchangers, valves and fine dust sensors, etc., which are only exemplary. It has a complex operating system.

Through the manager terminal connected to the BAS server, the manager can check the current status of the building's internal and external environment, the facilities and major devices in real time, and can remotely control the facilities inside the building if necessary.

However, general BAS has the following problems.

First, the BAS can be linked with the fire detection function provided in the building, but in most cases, it is only confirmed after the fire has progressed to some extent. For example, a fire detector linked to a sprinkler detects a fire only when high-temperature air from a fire flows to the sensor position. At this time, if you try to extinguish the sprinkler by automatically operating the sprinkler or sending a warning to the manager, it may be after a certain amount of damage has occurred, and in severe cases, it will have a great adverse effect on the building itself.

Second, to solve the first problem mentioned above, autonomous driving devices such as robots can be used, but a large initial investment is required to detect the entire facility space such as a machine room in a large building. In addition, the autonomous driving device mainly performs wireless communication, and if the repeater is damaged due to a fire or the like, data transmission may be similarly difficult.

Third, the location of the accident point (fire, flame, etc.) can be grasped with high accuracy through a depth camera (3D depth camera) provided on a robot, etc., but in order to recognize it in a situation such as a fire, the accuracy of several centimeters means a lot. There is little meaning to use an expensive depth camera.

In this regard, the present applicant has applied for a deep learning-based building management system equipped with a rail robot device through Korean Patent Application Laid-Open No. 10-2022-0061312. However, the above-mentioned problems are still confirmed in the above patents, and in particular, when two or more people are detected based on deep learning, an additional problem in that recognition accuracy is lowered was also confirmed.

In addition, techniques for improving the problems of the existing BAS system have been disclosed through Korean Patent No. 10-1866861, Korean Patent No. 10-1860872, and Korean Patent No. 10-1845396 by the present applicant. One problem is still unresolved.

(Patent Document 1) KR 10-2022-0061312 A

(Patent Document 2) KR 10-1866861 B1

(Patent Document 3) KR 10-1860872 B1

(Patent Document 4) KR 10-1845396 B1

(Patent Document 5) KR 10-2096175 B1

(Patent Document 6) KR 10-2020-0074503 A

(Patent Document 7) KR 10-1543561 B1

(Patent Document 8) KR 10-2020-0061578 A

The present invention has been devised to solve the above problems.

We would like to propose a method that can be linked with the BAS server, but can detect a fire early and warn the manager as quickly as possible.

At the same time, we want to propose a method that can detect a large building space with minimum equipment and minimum energy. For example, it is intended to propose a method useful for a building with a large space such as a machine room, such as a high-rise building, a large plant, and a large number of caution facilities to be monitored. If the distance can be recognized by a certain algorithm without having to use the depth camera, the depth camera may not be used to reduce the initial cost.

In addition, when two or more situations occur at the same time, by applying the method of setting priorities, we want to propose a method for quickly identifying more urgent situations and continuously monitoring and observing them.

An embodiment of the present invention for solving the above problems is an automatic building control method using a rail bot 100 movable along a rail 200 provided in a building and a server unit 400 capable of communicating therewith, The server unit 400 includes a BAS (Building Automation System) server 401 and a railbot server 402 for controlling the railbot 100, and the railbot 100 includes a depth camera (3D depth). a camera module 170 including a real image camera 171 that is not a camera) and a VGA camera, and a thermal imager 172 that is driven by tilt and pan together with the real image camera 171; a tilt step motor 141 for controlling the tilt angle of the camera module 170; and a fan step motor 142 for controlling the fan angle of the camera module 170, wherein the server unit 400 includes: a railbot driving module 410 for controlling the driving of the railbot 100; a human detection module 430 that determines whether the object is human when the camera module 170 detects the movement of an object on the screen using artificial intelligence and calculates recognition accuracy at this time; and a warning module 490, wherein the method includes the steps of: (a1) detecting, by the railbot 100, a movement of an object on a screen captured by the camera module 170; (a2) determining, by the human detection module 430, that a human is detected when the recognition accuracy of the object in which the motion is sensed is equal to or greater than a preset value; (a3) judging, by the human detection module 430, the detected human as a dangerous person when the movement is detected within the detection time on a preset schedule; (a4) When the human detection module 430 determines that there are two or more dangerous people because there are two or more objects for which the motion is sensed, it is determined that an object with a high recognition accuracy value is a high risk person, and the recognition determining an object having a large number of pixels on a screen corresponding to the object as a high risk person when the accuracy values are the same; (a5) The railbot driving module 410 checks the center on the screen through the pixel of the highest priority in the step (a4), and the confirmed center is photographed by the camera module 170 operating at least one of the tilt step motor 141 and the pan step motor 142 to come to the center of the screen; (a6) estimating, by the human detection module 430, the distance between the railbot 100 and the center of the highest-priority risk person in a preset manner while performing the step (a5); (a7) At the same time as performing steps (a5) and (a6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving it until it is between 95 degrees and -85 degrees; (a8) the human detection module 430 includes: i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (a6), estimating the position on the plane of the risk person with the highest priority; and (a9) displaying the estimated location on a preset floor plan by the warning module 490 and outputting it through a preset web client terminal 510, wherein the step (a6) includes the human detection The module 430 identifies pixels corresponding to the head of a risk person having the highest priority among pixels of the recognized object, and uses the correlation between the number of pixels corresponding to the average head size of a person and the identified number of pixels Thus, it provides a method, including the step of estimating the distance between the rail bot 100 and the highest priority risk person.

In addition, the server unit 400 further includes a fire detection module 440 for judging a fire on the screen photographed by the camera module 170 using artificial intelligence, the method comprising: (b1) the fire determining, by the detection module 440, that a fire has occurred on the screen photographed by the camera module 170; (b3) the fire detection module 440 identifies the outline of the group of the portion determined to be a fire, checks the number of fires using the discontinuity of the identified outline, and determines that the number of fires is two or more judging a group having a larger area of a group bounded by a discontinuous outline as a high-priority fire; (b5) the railbot driving module 410 checks the center of the fire having the highest priority in the step (b3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142; (b6) estimating, by the fire detection module 440, the distance between the railbot 100 and the center of the highest priority fire in a preset manner while performing the step (b5); (b7) At the same time as performing the steps (b5) and (b6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving the railbot 100 so that the distance estimated according to step (b6) does not become less than a preset limit distance while moving it until it is between 95 degrees and -85 degrees; (b8) the fire detection module 440 includes: i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (b6), estimating the location on the plane of the highest priority fire; and (b9) the warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and a preset rule based on the estimated location on the floor plan The method further includes the step of turning off the power of one or more facilities and stopping the supply/exhaust facility and the damper according to the The pan angle and tilt angle of the camera module 170 at any one position (A) before starting to move according to step (b7), and ii) the railbot 100 according to step (b7) After starting the movement, using the pan angle and tilt angle of the camera module 170 at the one position (A) and the other position (B), the railbot 100 and the highest priority It preferably includes the step of estimating the distance between the centers of the fire.

In addition, the server unit 400 further includes a flame detection module 450 for judging the presence or absence of a flame on the screen photographed by the camera module 170 using artificial intelligence, the method comprising: (c1) the flame determining, by the detection module 450, that a flame has occurred on the screen photographed by the camera module 170; (c3) the flame detection module 450 identifies the outline of the group of the portion corresponding to the determined flame, and checks the number of flames using whether the identified outline is discontinuous, and the number of flames is two. judging a group having a larger area of a group bordered by a discontinuous outline as a flame having a higher priority when it is determined as above; (c5) the railbot driving module 410 checks the center of the flame having the highest priority in the step (c3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142; (c6) estimating, by the flame detection module 450, the distance between the railbot 100 and the center of the highest priority flame in a preset manner while performing the step (c5); (c7) At the same time as performing steps (c5) and (c6), when the railbot driving module 410 has a fan angle of the camera module 170 between -85 and 85 degrees, the railbot ( If the angle of the camera module 170 is 85 degrees to 95 degrees or -95 degrees while moving the rail bot 100 to move the rail bot 100 backward 100) and moving the rail bot 100 backward when it is between 95 degrees and -95 degrees. moving the railbot 100 so that the distance estimated according to step (c6) does not become less than a preset limit distance while moving it until it is between degrees and -85 degrees; (c8) the flame detection module 450, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) estimating the position on the plane of the flame having the highest priority by using the distance estimated in step (c6); and (c9) the warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and a preset rule based on the estimated location on the floor plan The method further includes turning off the power of one or more facilities according to the step (c6), wherein the fire detection module 440 i) the railbot 100 moves according to the step (c7). The pan angle and tilt angle of the camera module 170 at any one position (A) before starting, and ii) After the railbot 100 starts moving according to the step (c7), the Using the pan angle and tilt angle of the camera module 170 at the position (A) and the other position (B), the distance between the railbot 100 and the center of the highest priority flame is estimated It is preferable to include the step of

In addition, the server unit 400 uses artificial intelligence to determine whether there is smoke on the screen photographed by the camera module 170 and uses that smoke is moving from a darker direction to a lighter smoke on the screen. The method further includes a smoke detection module 460 that identifies the movement direction of the smoke by doing so, (d1) that the smoke detection module 460 detects that smoke has occurred on the screen captured by the camera module 170 determining and identifying the direction of movement of the smoke; (d2) identifying, by the smoke detection module 460, a smoke generating point using pixels corresponding to the determined smoke and the identified moving direction of the smoke; (d5) the railbot driving module 410, the tilt step motor 141 and the pan step so that the smoke generating point identified in step (d2) comes to the center of the screen photographed by the camera module 170 operating any one or more of the motors 142; (d6) estimating, by the smoke detection module 460, the distance between the railbot 100 and the smoke generating point in a preset manner while performing the step (d5); (d7) At the same time as performing steps (d5) and (d6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving the railbot 100 so that the distance estimated according to step (d6) does not become less than a preset limit distance while moving it until it is between 95 degrees and -85 degrees; (d8) the smoke detection module 460, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) estimating a location on a plane of a smoke generating point using the distance estimated in step (d6); and (d9) the warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and a preset rule based on the estimated location on the floor plan The method further includes the step of turning off the power of one or more equipment and stopping the supply/exhaust equipment and the damper according to the (d7) the pan angle and tilt angle of the camera module 170 at any one position (A) before starting to move according to step (d7) and ii) the railbot 100 according to step (d7) After starting the movement, using the pan angle and tilt angle of the camera module 170 at the one position (A) and the other position (B), the distance between the railbot 100 and the smoke generating point is It preferably includes the step of estimating the distance.

In addition, the server unit 400 further includes a immersion detection module 470 that determines whether there is immersion on the screen photographed by the camera module 170 using artificial intelligence, and the method includes: (e1) the immersion determining, by the detection module 470, that submergence has occurred on the screen photographed by the camera module 170; (e3) the submersion detection module 470 identifies the outline of the group of the portion corresponding to the determined submergence, checks the number of submergences using the discontinuity of the identified outline, and the number of submergence is two judging a group having a larger area of a group bordered by a discontinuous outline as a high-priority immersion if it is determined as above; (e5) the railbot driving module 410 checks the center of the immersion with the highest priority in the step (e3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142; (e6) estimating, by the submersion detection module 470, the distance between the railbot 100 and the center of the highest priority submersion in a preset manner while performing step (e5); (e8) the immersion detection module 470, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (e6), estimating the position on the plane of the submerged with the highest priority; and (e9) the warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and a preset rule based on the estimated location on the floor plan The method further includes turning off the power of one or more equipment according to the step (e6), wherein the submersion detection module 470 and the railbot 100 start moving according to the step (e7). The pan angle and tilt angle of the camera module 170 at any previous position (A) and the position (A) after the railbot 100 starts to move according to step (e7) and estimating the distance between the railbot 100 and the center of the highest priority immersion using the pan angle and the tilt angle of the camera module 170 at another position (B). It is preferable to do

In addition, the server unit 400 includes, for each of a plurality of facilities, a facility database 409 in which the location of the facility on the plane and the location of the monitoring on the plane of the facility are pre-stored; and a facility monitoring module 480 capable of identifying an abnormality in any one of the plurality of facilities using the facility database 409, the method comprising: (f1) monitoring the facility confirming, by the module 480, that there is an abnormality in any one of the plurality of facilities; (f5) the railbot driving module 410 operates any one or more of the tilt step motor 141 and the fan step motor 142 so that the equipment identified in the step (f1) comes to the center of the screen step; (f7) simultaneously performing the step (f5), the railbot driving module 410 moving the railbot 100 to the monitoring position of the equipment stored in advance for the equipment; and (f9) outputting, by the warning module 490 , the image captured by the camera module 170 of the railbot 100 through the web client terminal 510 .

In addition, in the facility database 409, for each facility, the first upper limit value that is the maximum normal operating temperature, the second upper limit value that is greater than the first upper limit value as the equipment breakage temperature, the first lower limit value that is the normal operating minimum temperature, the equipment freezing temperature a second lower limit value smaller than the first lower limit value, a first control variable for lowering the facility operation rate to control overheating of the facility, a second control variable for increasing the facility operation rate to prevent freezing of the facility, and the second control variable to prevent the freezing of the facility A third control variable having a higher facility operation rate than the second control variable is pre-stored, and the server unit 400 further includes a facility temperature sensing module 420 and a facility control module 421, (g11) the facility Step, by the temperature sensing module 420, checking the temperature of the facility on the screen photographed by the camera module 170 and identifying the facility; (g12) the facility temperature sensing module 420, when the temperature checked in step (g11) is greater than the second upper limit value stored in advance for the facility, the facility control module 421 turns off the facility, outputting a warning to a preset web client terminal 510 through the warning module 490; (g13) when the facility temperature detection module 420, the temperature checked in step (g11) is greater than the first upper limit value stored in advance for the facility and smaller than the second upper limit value, the facility control module 421 controls the facility changing the control variable of , to a first control variable stored in advance for the corresponding facility, controlling the control variable, and outputting a warning to a preset web client terminal 510 through the warning module 490; (g14) the facility temperature sensing module 420, when the temperature checked in step (g11) is smaller than the first lower limit value stored in advance for the facility and greater than the second lower limit value, the facility control module 421 controls the facility changing the control variable of , to a second control variable stored in advance for the facility, controlling the control variable, and outputting a warning to a preset web client terminal 510 through the warning module 490; and (g15) the facility temperature sensing module 420, when the temperature checked in step (g11) is less than the second lower limit value stored in advance for the facility, the facility control module 421 controls the facility's control variable. Preferably, the method further includes the step of controlling the corresponding facility by changing it to a stored third control variable, and outputting a warning to a preset web client terminal 510 through the warning module 490 .

In addition, the real image camera 171 and the thermal image camera 172 of the railbot 100 are oriented in the same direction to acquire a real image and a thermal image screen for the same screen together, and the railbot 100 ) further includes a communication module 160 for performing a wireless communication function to wirelessly transmit the acquired real image and thermal image screen to the user terminal 511 through the communication module 160, and the server unit 400 ) communicates with the Ethernet hub 311 of the repeater unit 310 through Ethernet, and the Ethernet hub 311 and the railbot 100 are wired through a copper tube located in the rail 200, or Alternatively, it is preferable that the railbot 100 transmits the acquired real image and thermal image to the web client terminal 510 through the Ethernet hub 311 through a wireless connection.

In addition, it is preferable that the rail bot 100 and the power supply unit 310 are connected by wire through the copper tube in the rail 200 .

In addition, the rail 200 is preferably installed on the lower surface of the ceiling of the space in the building adjacent to the outer wall of the space.

In addition, the rail bot 100, one or more wheels 102 rotating on the rail 200; a mainboard module 130 for communicating with the repeater unit 310 by wire or wireless connection; a housing 103 in which the main board module 130 is mounted; a power roller 110 that is in contact with the rail 200 and is wiredly connected to the power supply unit 310 to supply power; an encoder operation module 143 connected to the main board module 130 by wire to encode the image captured by the camera module 170; a motor 144 to which electric power is supplied to rotate the wheel 102; a driving gear 145 connecting the motor 144 and the wheel 102; a battery and charging module 146 to which power is supplied to charge power; a front proximity sensor 151 for detecting an obstacle in front of the railbot 100; And it is preferable to further include a rear proximity sensor 152 for detecting a rear obstacle of the rail bot (100).

In addition, the BAS server 400 is a CCMS (Central Control & Monitoring System), and the BAS server 400 is a gateway 610 and DDC (Direct Digital Control) 620 through the facilities 630 in the building. and, the facility 630 preferably includes any one or more of an air conditioner, a boiler, a temperature and humidity sensor, a power/peak controller, lighting, a heat exchanger, a valve, and a fine dust sensor.

The present invention for solving the above problems further provides a computer program recorded in a computer program for executing the above-described method, stored in a computer-readable recording medium and the recording medium.

According to the present invention, it is possible to operate a BAS capable of early detection of various disasters such as fire. By analyzing images by railbots that can move to various locations, it is possible to detect various disasters such as dangerous people, fire, smoke, sparks, and flooding.

At the same time, it can detect a large building space with minimal equipment and minimal energy. By applying the present invention to a large-sized building, by installing rails at an appropriate location and appropriately managing the movement schedule of the railbot, even a minimum number of railbots can detect the entire large space without blind spots.

In addition, if there is an abnormality in the equipment, it is possible to visually check the equipment that is known to have an abnormality without the manager moving to the space where the equipment is located.

The rail bot according to the present invention can detect first by increasing the priority of the identified human with higher accuracy when two or more human beings are detected at the same time. By supplementing artificial intelligence's unstable human detection, the above warning can be prevented. In addition, even when two or more fires, two or more flames, or two or more floods occur, resources can be efficiently distributed in preparation for a larger accident by prioritizing and detecting larger fires, flames, flooding, etc. .

According to the present invention, the distance to the dangerous person is estimated using the number of pixels on the head, and the distance to the fire, flame, smoke, and submergence point uses trigonometry. This generally considers a large movement of a person, and may increase the accuracy of distance estimation.

1 schematically shows a system in which a method according to the invention is carried out.
2 schematically shows a wired communication method between a server unit and a facility according to the present invention.
3 shows an example of a screen output from a web client terminal according to the present invention.
4 exemplarily shows a perspective view of a railbot in which the method according to the present invention is performed.
5 schematically shows the configuration of a railbot in which the method according to the present invention is performed.
6 illustrates an example of a screen on which a real image and a thermal image are simultaneously output in the web client terminal according to the present invention.
7 schematically shows the configuration of a railbot server according to the present invention.
8 is a view for explaining a method of estimating a distance to a point of fire, flame, smoke, flooding, etc. in the method according to the present invention.

Hereinafter, the present invention will be described in detail with reference to the drawings.

1. Description of the system

A system for carrying out the method according to the present invention will be described with reference to FIGS. 1 to 7 .

A system for performing the method according to the present invention includes a railbot 100 and a server unit 400 . The server unit 400 includes a BAS server 401 and a railbot server 402 .

Railbot (railbot) 100 is collectively referred to as a robot that moves along the rail (200). In one embodiment of the present invention, it moves along the rail 200 using the wheels 102, but is not limited thereto.

The BAS server 401 includes a monitoring function of facilities in a building like a conventional general BAS server. The railbot server 402 analyzes information acquired from the railbot 100 unique to the present invention, and the railbot 100 ), and further includes specific functions for controlling equipment in the building. The modules responsible for each function are shown in Fig. 7, which will be described in detail in the description of the method below.

First, the server unit 400 and related configurations will be described with reference to FIG. 1 .

The server unit 400 includes a BAS server 401 and a railbot server 402 . The BAS server 401 and the Railbot server 402 do not necessarily have to be physically separate servers, and may perform both roles of two servers on one piece of hardware, if necessary. Hereinafter, for reference, it will be described with bending.

The BAS server 401 is wired or wirelessly connected to various facilities 630 in the building like the conventional BAS server, checks various information obtained from the facility 630, and controls them (Central Control & Monitoring CCMS) system). Here, the facility 630 includes, but is not limited to, any one or more of an air conditioner, a boiler, a temperature and humidity sensor, a power / peak controller, a lighting, a heat exchanger, a valve, and a fine dust sensor. can also be connected.

The BAS server 401 communicates with the facilities 630 in the building through a gateway 610 and a direct digital control (DDC) 620 . A more detailed example is shown in FIG. 2 .

In addition, the BAS server 401 may be connected to the web client terminal 510 through Ethernet. The administrator may check the state of the facility 630 through the web client terminal 510 and control it. Examples of screens that can be checked in the web client terminal 510 are illustrated in FIG. 3 as (a) to (f).

The railbot server 402 receives information from the railbot 100 and performs a function of controlling the railbot 100 .

Meanwhile, the server unit 400 is provided with a facility database 409 including information on the facility 630 . The facility database 409 stores in advance the names and characteristics of the facilities 630 in the building.

In one embodiment of the present invention, in the facility database 409 , for each of a plurality of facilities, the location of the facility on the plane and the location of the monitoring on the plane of the facility may be stored in advance. Here, the monitoring position on the plane is a position preset as an optimal position for the rail bot 100 to monitor the corresponding facility, and is a position on the rail 200 .

In another embodiment of the present invention, in the facility database 409, for each of a plurality of facilities, a first upper limit value that is a normal operating maximum temperature, a second upper limit value that is greater than the first upper limit value as a facility breakage temperature, and a first that is a normal operating minimum temperature The lower limit value, the second lower limit value smaller than the first lower limit value as the equipment freezing temperature, the first control variable lowering the equipment operation rate to control overheating of the equipment, a second control variable increasing the equipment operation rate to prevent the equipment freezing, and freezing of the equipment A third control variable having a higher facility operation rate than in the second control variable may be pre-stored to prevent . Utilizing the facility weather detection function described later, if the temperature of the facility is particularly high or particularly low, whether a fire has occurred due to damage or overheating (more than the second upper limit value), or simply overheating due to a malfunction due to a user setting error, etc. Whether it is a stage (more than the first upper limit and less than the second upper limit), whether it is operating normally (more than the first lower limit or less than the first upper limit), whether freezing is in progress (more than the second lower limit and less than the first lower limit), whether freezing has occurred or is likely to occur It can be distinguished whether this is high (below the second lower limit). Since each facility has different characteristics, values for these are previously stored in the facility database 409 . In addition, since the control variables for increasing or decreasing the operation rate of equipment according to each situation are also different for each equipment, these values are also stored in the equipment database 409 in advance.

Next, the railbot 100 will be described with reference to FIGS. 4 to 6 . FIG. 4 shows a perspective view of an embodiment of the railbot 100 , and FIG. 5 schematically shows components of the railbot 100 .

The railbot 100 monitors the equipment 630 and the like while moving along the rail 200 according to a preset schedule.

The rail 200 is preferably installed on the lower surface of the ceiling of a space such as a machine room in which the equipment is located in the building and is installed adjacent to the outer wall of the space. It can monitor as many spaces as possible.

In addition to the function of guiding the wheel 102 of the rail bot 100, the rail 200 exposes the contact point of the communication cable connected from the Ethernet hub 311 and the contact point of the power cable connected from the power supply unit 310 to the outside, so that the power roller (110) has a function of connecting these and the railbot 100 by wire. That is, as the rail bot 100 travels on the rail 200 , the cables mounted inside the rail bot 100 are connected to connect the rail bot 100 , the repeater unit 310 , and the power supply unit 320 . Here, the cables are located in a copper tube located in the rail 200 . Therefore, when the rail 200 including the copper tube is installed, even if the space in the building to be monitored by the rail bot 100 is large, wired connection is easily made, and stable performance is exhibited by protecting the cable from fire by the copper tube. .

In addition to the wired connection between the railbot 100 and the Ethernet hub 311, they may be connected wirelessly.

Meanwhile, the railbot 100 is capable of wireless communication with the user terminal 511 by the communication module 160 . Accordingly, the administrator may directly wirelessly communicate with the rail bot 100 through the user terminal 511 to control it or check information, and to control the rail bot 100 through the above-described rail bot server 402 or receive information. You can also check

As such, the railbot 100 according to the present invention is capable of both a wired connection and a wireless connection.

On the other hand, facility monitoring of the railbot 100 is performed through the camera module 170 . The camera module 170 includes a real image camera 171 and a thermal image camera 172 . The screen or image captured by the camera module 170 is transmitted to the server unit 400 by communication, and various modules of the server unit 400 analyze it to diagnose the problem of the facility 630 in the building, and, if necessary, the facility. Directly controlling the 630 , delivering a warning to the user, or additionally driving the railbot 100 .

The server unit 400 receives data from the facility 630 through the above-described communication line, and by further using the railbot 100, it is possible to prepare for an error in a sensor provided in the facility 630, an error in communication connection, etc. have. In particular, in situations such as fire, flooding, and abnormalities in the facility 630 itself, normal data is difficult to be transmitted to the server unit 400, so the analysis of other methods due to the screen taken by the railbot 100 directly is Operational stability can be greatly improved.

Since the real image camera 171 is a depth camera (3D depth camera), the camera itself may calculate the distance from the rail bot 100 to the accident point such as a dangerous person or a fire with high accuracy, but in the present invention, it is not a depth camera Preferably, a VGA camera may be employed. In situations such as fire and smoke, high accuracy of several centimeters does not mean much, so there is little meaning to apply an expensive depth camera.

The camera module 170 including the real image camera 171 and the thermal imager 172 may have a tilt angle in the XY direction (see FIG. 8 ) adjusted by the tilt step motor 141, and the pan step motor ( 142), the fan angle in the XZ direction (refer to FIG. 8) may be adjusted. Adjustments are made manually or automatically.

Railbot 100 further includes a communication module 160 that performs a wireless communication function. The real image and thermal image screen are wirelessly transmitted to the user terminal 511 by the communication module 160 . That is, the real image and the thermal image screen are transmitted to the web client terminal 510 by wire or wirelessly through the repeater unit 310 , and are transmitted to the user terminal 511 wirelessly through the communication module 160 . As described above, it is possible to prepare for situations such as fire. 6 illustrates an output screen of a real image and a thermal image screen checked in the web client terminal 510 .

Here, the railbot 100 may communicate with the server unit 400 by wired or wireless or both wired and wireless, and either method is usually used. In the case of wireless communication, the repeater unit 310 is generally adopted to solve the distance problem.

On the other hand, the railbot 100 includes a main board module 130 that communicates through a wired or wireless connection with the repeater unit 310 , a housing 103 in which the main board module 130 is mounted, and a main board module 130 . The encoder arithmetic module 143 that is connected to a wired connection and encodes the image taken by the camera module 170, and a motor 144 that rotates the wheel 102 through the connected driving gear 145 when power is supplied 144, and power A battery and charging module 146 for charging the, a front proximity sensor 151 for detecting a front obstacle of the railbot 100, and a rear proximity sensor 152 for detecting a rear obstacle of the railbot 100 are further added. include

2. Description of the method

Hereinafter, a method according to the present invention will be described with further reference to FIGS. 7 to 8 .

2.1 Facility temperature detection function

A first embodiment of the method according to the invention relates to a method for sensing the temperature of an installation 630 .

If the facility 630 is overheated or the temperature rises in case of a fire, there is a risk of freezing if the temperature is too low. The reference temperature is different for each facility 630 . When a corresponding situation occurs, the temperature can be controlled by increasing or decreasing the operation rate of the facility 630 . The control variable for controlling the operation rate is also different for each facility 630 . Such information is previously stored in the facility database 409 .

Specifically, in the facility database 409, for each of a plurality of facilities, the first upper limit value that is the normal operating maximum temperature, the second upper limit value that is greater than the first upper limit value as the equipment breakage temperature, the first lower limit value that is the normal operating minimum temperature, and the equipment freezing temperature A second lower limit value smaller than the first lower limit value, a first control variable for lowering the facility operation rate to control overheating of the facility, a second control variable for increasing the facility operation rate to prevent freezing of the facility, and the second control variable to prevent freezing of the facility A third control variable, which increases the facility operation rate higher than the control variable, is pre-stored.

On the other hand, the facility database 409 further stores the location on the plane of the facility for each of the plurality of facilities. Therefore, when the railbot 100 takes a screen shot, the facility temperature sensing module 420 may identify the facility appearing on the screen. Since the facility temperature detection module 420 has identified the facility, it is possible to check the upper limit value, the lower limit value, and the control variable of the temperature corresponding to the identified facility. It is possible to control the facility by checking whether it is within the normal value using the temperature standard and checking the control variable of the corresponding facility.

A specific method will be described.

The facility temperature detection module 420 checks the temperature of the facility on the screen captured by the camera module 170 and identifies the facility.

Next, when the facility temperature detection module 420, the previously checked temperature is greater than the second upper limit value stored in advance for the facility, it is determined as overheating or fire, the facility control module 421 turns off the facility, and a warning module A warning is output to the preset web client terminal 510 through 490 .

If the previously checked temperature is greater than the first upper limit value stored in the facility and smaller than the second upper limit value, it is determined as simple overheating, and the facility control module 421 sets the control variable of the facility to the first control stored in the facility. It is controlled by changing it into a variable, and a warning is outputted to a preset web client terminal 510 through the warning module 490 . Here, the first control variable is a variable that lowers the operation rate.

If the previously checked temperature is smaller than the first lower limit value stored in the facility and greater than the second lower limit value, the facility control module 421 determines the level of fear of freezing and sets the control variable of the facility to the second control stored in the facility. It is controlled by changing it into a variable, and a warning is outputted to a preset web client terminal 510 through the warning module 490 . Here, the second control variable is a variable that slightly increases the operation rate.

If the previously checked temperature is smaller than the second lower limit value stored in advance for the facility, it is determined that the risk of freezing is high, and the facility control module 421 changes the control variable of the facility to a third control variable previously stored for the facility. and outputs a warning to a preset web client terminal 510 through the warning module 490 . Here, the third control variable is a variable that greatly increases the operation rate.

2.2 Human detection function

A second embodiment according to the present invention relates to a method for detecting a person in a predetermined space in a building. Here, sensing a person refers to sensing a person or a movement of a person.

The corresponding function is performed by the human detection module 430 . The human detection module 430 determines whether the object is human when the camera module 170 detects the movement of the object on the screen using any widely used human detection artificial intelligence, and calculates the recognition accuracy at this time.

A specific method will be described.

When the railbot 100 detects the movement of an object on the screen captured by the camera module 170, the human detection module 430 detects the human being detected when the recognition accuracy of the object whose movement is detected is equal to or greater than a preset value. judge to be

Next, the human detection module 430 determines that the detected human being is a dangerous person when the time at which the movement is sensed is within the detection time on the preset schedule. If it is not the detection time according to the preset schedule, for example, if it is the manager's work time or regular patrol time, this is because even if a person is detected by the human detection module 430 , it may be an authorized person. For this, the schedule can be input by the administrator.

Next, when the human detection module 430 determines that there are two or more dangerous people because there are two or more objects for which motion is detected, the object with a large value of recognition accuracy is determined as a high-risk person with high priority, and the value of recognition accuracy is In the same case, an object with a large number of pixels on the screen corresponding to the object is judged as a high-priority risk person. The priority of recognition accuracy is to operate a warning light only when it is more reliably determined as a human because the currently commercialized human detection artificial intelligence is not 100% perfect. For the same reason, if the recognition accuracy values are the same, the priority of a person located closer to the railbot 100, that is, an object having a large number of pixels is set higher.

Next, the railbot driving module 410 checks the center of the screen through the pixel of the person with the highest priority, and the tilt step motor ( 141) and the fan step motor 142 to operate any one or more to direct the camera to the person.

At the same time, the human detection module 430 estimates the distance between the rail bot 100 and the center of the highest priority risk person in a preset manner. Estimation of the distance may use the size of a human head. The human detection module 430 identifies pixels corresponding to a human head among pixels of the recognized object and checks the number of pixels. The number of pixels corresponding to the average head size of a person is different for each distance, and their correlation is stored in advance. By using the confirmed number of pixels, the distance between the railbot 100 and the human can be estimated.

At the same time, the railbot driving module 410 determines that there is a human in front of the railbot 100 when the fan angle of the camera module 170 is between -85 and 85 degrees and advances the railbot 100 and If it is between 95 degrees and -95, it is determined that there is a human behind the rail bot 100 and moves the rail bot 100 to move the rail bot 100 backward, thereby bringing the rail bot 100 closer to the human. Here, the pan angle of 0 degrees means the traveling direction of the railbot 100 .

When the rail bot 100 moves and approaches the human, the human reaches a position perpendicular to the progress direction of the rail bot 100, and the corresponding position becomes the position where the rail bot 100 is closest to the human. At this time, the fan angle of the camera module 170 is between 85 degrees to 95 degrees or -95 degrees to -85 degrees. The railbot driving module 410 stops the railbot 100 because it is closest to the person when the angle is reached while checking the fan angle in real time. Currently, the railbot 100 is communicating the screen that most closely approaches the human and looks at the human.

Now, the human detection module 430 is configured to: i) the current on-plane position of the railbot 100 , ii) the pan angle of the camera module 170 , iii) the tilt angle of the camera module 170 , and iv) the previous Using the estimated distance between the rail bot 100 and the person, the location of the person on the plane can be estimated.

The warning module 490 displays the estimated human location on a preset floor plan (that is, a plan view of the space where the rail bot 100 is located) and outputs it through the preset web client terminal 510 .

On the other hand, in estimating the distance to humans, the distance between the rail bot 100 and the person can be estimated using trigonometry, which will be described later in the fire detection function, etc., but unlike fire, flame, smoke, etc., there is a high possibility that a person will continuously move. In this case, it is more effective to use the number of pixels in the head rather than trigonometry to quickly estimate the position of a human moving in real time on the plane.

2.3 Fire detection function

A third embodiment according to the present invention relates to a method for detecting a fire in a predetermined space in a building.

The corresponding function is performed by the fire detection module 440 . The fire detection module 440 determines whether there is a fire on the screen captured by the camera module 170 using the fire detection artificial intelligence. Fire detection artificial intelligence may use any method, and since this is a prior art, detailed description thereof will be omitted.

A specific method will be described.

The fire detection module 440 determines that a fire has occurred on the screen captured by the camera module 170, but the fire detection module 440 identifies the outline of the group of the fire part, and uses whether the identified outline is discontinuous. Thus, by checking the number of fires, it is possible to determine whether the number of fires is two or more. If the number of fires is determined to be two or more, the fire may actually have occurred in two places, or the fire occurred only in one place, but when the railbot 100 looks at it, it is covered by the facilities 630, etc. may have been judged. However, whether it is the former or the latter, it is not necessary to separate them because the measures of promptly conducting disaster prevention work, stopping the operation of facilities in the space, and warning the manager are the same. However, it is important to find the ignition point through the larger side of the fire and notify the management site, so the group with a larger area bordered by the discontinuous outline is judged as a high priority fire.

Next, the railbot driving module 410 checks the center of the fire and directs the camera to face it. The method is the same as in the second embodiment.

At the same time, the fire detection module 440 estimates the distance between the railbot 100 and the center of the fire in a preset manner. In this case, trigonometry can be used. 8, i) the pan angle and tilt angle of the camera module 170 at any one position (A) before the railbot 100 starts moving toward the center of the fire, which will be described later, and ii ) By using the pan angle and tilt angle of the camera module 170 at the one position (A) and the other position (B) after the railbot 100 starts moving toward the center of the fire, which will be described later, It is possible to estimate the distance between the current railbot 100 and the center of the highest priority fire (that is, L, the distance between B and C).

At the same time, the railbot driving module 410 moves the railbot 100 toward the center of the fire. This is the same as the second embodiment. However, unlike the case of approaching people, when the rail bot 100 approaches the fire, if the rail bot 100 approaches excessively, the failure of the rail bot 100 is concerned, and in this case, data can no longer be transmitted, so the previously estimated It is preferable to stop the railbot 100 before that so that the distance L does not become less than a preset limited distance (eg, 3 m). Currently, the railbot 100 approaches the fire the most (however, does not approach the fire excessively) and communicates the screen looking at the fire.

Now, the fire detection module 440 determines that: i) the railbot 100's in-plane position, ii) the pan angle of the camera module 170, iii) the tilt angle of the camera module 170, and iv) the previously estimated Using the distance L, it is possible to estimate the location of the fire on the plane.

The warning module 490 displays the location of the fire on a preset floor plan (ie, a plan view of the space where the railbot 100 is located) and outputs it through the preset web client terminal 510 .

2.4 Flame detection function

A fourth embodiment according to the present invention relates to a method for detecting a flame in a predetermined space in a building.

The corresponding function is performed by the flame detection module 450 . The flame detection module 450 determines whether there is a flame on the screen captured by the camera module 170 using artificial intelligence for detection of flame. Any conventional artificial intelligence may be used. For example, a flame may be identified through a change in the amount of light of a pixel, or a flame may be identified by detecting a specific wavelength of the flame, such as infrared rays or ultraviolet rays.

The method except for the flame detection method is substantially the same as that of the third embodiment, and thus detailed description thereof will be omitted.

2.5 Smoke detection function

A fifth embodiment according to the present invention relates to a method for detecting smoke in a predetermined space in a building.

The corresponding function is performed by the smoke detection module 460 . The smoke detection module 460 determines whether there is smoke on the screen captured by the camera module 170 using artificial intelligence for detection of smoke. Any conventional artificial intelligence may be used. For example, it is possible to determine whether to smoke through a change in a gray to black region among pixels.

In addition, the artificial intelligence of the smoke detection module 460 can also identify the direction of movement of smoke by changing the area of the smoke, specifically by using that the smoke is moving from a direction in which the smoke is darker to a direction in which it is lighter on the screen. Also, it is possible to identify the point where the smoke is generated through the direction of movement of the smoke. Since the specific identification method is a prior art, a detailed description thereof will be omitted.

A specific method will be described.

The smoke detection module 460 determines whether smoke is generated on the screen photographed by the camera module 170 and determines whether the color is dark by assuming that the smoke is moving from a darker direction to a lighter smoke on the screen. The movement direction of smoke is identified through the method, and the smoke generation point is identified using pixels corresponding to the determined smoke and the identified movement direction of the smoke.

Next, the railbot driving module 410 operates any one or more of the tilt step motor 141 and the pan step motor 142 so that the identified smoke generating point is in the center of the screen photographed by the camera module 170 .

At the same time, the smoke detection module 460 estimates the distance between the railbot 100 and the smoke generating point in a preset manner. The specific method is substantially the same as in the third embodiment.

At the same time, the railbot driving module 410 moves the railbot 100 toward the smoke generating point. The specific method is substantially the same as that of the third embodiment. Currently, the railbot 100 approaches the smoke generating point the most (however, without excessively approaching it) and communicates the screen looking at the smoke generating point.

Now, the smoke detection module 460 determines: i) a plane position of the railbot 100 , ii) a pan angle of the camera module 170 , iii) a tilt angle of the camera module 170 , and iv) Using the distance estimated in step (d6), the location on the plane of the smoke generating point is estimated.

The warning module 490 displays the location of the smoke generating point on a preset floor plan (that is, a plan view of the space where the railbot 100 is located) and outputs it through the preset web client terminal 510 .

2.6 Water immersion detection function

A sixth embodiment according to the present invention relates to a method for detecting flooding in a predetermined space in a building.

The corresponding function is performed by the immersion detection module 470 . The submersion detection module 470 determines whether there is submersion on the screen photographed by the camera module 170 using artificial intelligence for detecting submergence. Any conventional artificial intelligence may be used. For example, the submersion may be determined by changing the wavelength corresponding to water among the wavelengths of the pixel.

The method except for the submerged detection method and the railbot movement to the corresponding point is substantially the same as in the third embodiment, and thus a detailed description thereof will be omitted.

2.7 Facility monitoring function

The seventh embodiment according to the present invention is a function of moving the rail bot 100 to a position where the equipment can be monitored when an abnormality is confirmed in the equipment, and providing an image captured thereto to the manager.

In general BAS, abnormalities of facilities can be checked, but only the fact that an abnormality has occurred is confirmed. It is not possible to confirm whether it was issued or whether prompt action should be taken. According to the present invention, when a facility abnormality is confirmed, the railbot 100 can be moved to deliver the image captured by the camera module 170 directly to the manager, so even if an abnormality occurs in the facility located in the blind spot of the CCTV, the current situation can be maintained. can be checked quickly and accurately.

To this end, as described above, the facility database 409 stores in advance the location of the facility and the location of the facility on the plan, which are different for each facility. The monitoring position is preset to a position that can best photograph the equipment when the rail bot 100 moves.

The facility monitoring function is performed by the facility monitoring module 480 .

A specific method will be described.

When the facility monitoring module 480 confirms that there is an abnormality in any one of the plurality of facilities, the railbot driving module 410 moves the tilt step motor 141 and the pan step so that the facility comes to the center of the screen. Any one or more of the motors 142 are operated.

At the same time, the railbot driving module 410 moves the railbot 100 to the monitoring position of the facility stored in advance for the facility.

Next, the warning module 490 outputs the image captured by the camera module 170 of the railbot 100 through the web client terminal 510 .

100: railbot
102: wheel
103: housing
110: power roller
130: motherboard module
141: tilt step motor
142: fan step motor
143: encoder operation module
144: motor
145: drive gear
146: battery and charging module
151: front proximity sensor
152: rear proximity sensor
160: communication module
170: camera module
171: visual camera module
172: thermal imaging camera module
200: rail
310: relay donation
311: Ethernet Hub
312: TLC modem
320: power unit
400: server unit
401: BAS Server
402: railbot server
501: facility database
510: web client terminal
511: user terminal
520: user terminal
610: gateway
620: DDC
630: equipment

Claims (14)

As a building automatic control method using a rail bot 100 that can move along a rail 200 provided in a building and a server unit 400 that can communicate with it,
The server unit 400 includes a BAS (Building Automation System) server 401 and a railbot server 402 for controlling the railbot 100,
The railbot 100,
A camera including a real image camera 171 that is not a 3D depth camera, but a VGA camera, and a thermal imager 172 that drives tilt and pan together with the real image camera 171 . module 170;
a tilt step motor 141 for controlling the tilt angle of the camera module 170; and
and a fan step motor 142 for controlling the fan angle of the camera module 170,
The server unit 400,
a railbot driving module 410 for controlling the driving of the railbot 100;
a human detection module 430 that determines whether the object is human when the camera module 170 detects the movement of an object on the screen using artificial intelligence and calculates recognition accuracy at this time; and
a warning module 490;
The method is
(a1) detecting, by the railbot 100, the movement of an object on the screen captured by the camera module 170;
(a2) determining, by the human detection module 430, that a human is detected when the recognition accuracy of the object in which the motion is sensed is equal to or greater than a preset value;
(a3) judging, by the human detection module 430, the detected human as a dangerous person when the movement is detected within the detection time on a preset schedule;
(a4) When the human detection module 430 determines that there are two or more dangerous people because there are two or more objects for which the motion is sensed, it is determined that an object with a high recognition accuracy value is a high risk person, and the recognition determining an object having a large number of pixels on a screen corresponding to the object as a high risk person when the accuracy values are the same;
(a5) The railbot driving module 410 checks the center on the screen through the pixel of the highest priority in the step (a4), and the confirmed center is photographed by the camera module 170 operating at least one of the tilt step motor 141 and the pan step motor 142 to come to the center of the screen;
(a6) estimating, by the human detection module 430, the distance between the railbot 100 and the center of the highest-priority risk person in a preset manner while performing the step (a5);
(a7) At the same time as performing steps (a5) and (a6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving it until it is between 95 degrees and -85 degrees;
(a8) the human detection module 430 includes: i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (a6), estimating the position on the plane of the risk person with the highest priority; and
(a9) the warning module 490 displaying the estimated location on a preset floor plan and outputting it through a preset web client terminal 510,
The step (a6) is,
The human detection module 430 identifies the pixels corresponding to the head of a risky person with the highest priority among the pixels of the recognized object, and selects the number of pixels between the number of pixels corresponding to the average head size of the person and the identified number of pixels. Using the correlation, including the step of estimating the distance between the rail bot 100 and the risk person with the highest priority,
Way.
The method of claim 1,
The server unit 400 further includes a fire detection module 440 that determines a fire on the screen shot by the camera module 170 using artificial intelligence,
The method is
(b1) determining, by the fire detection module 440, that a fire has occurred on the screen captured by the camera module 170;
(b3) the fire detection module 440 identifies the outline of the group of the portion determined to be a fire, checks the number of fires using the discontinuity of the identified outline, and determines that the number of fires is two or more judging a group having a larger area of a group bounded by a discontinuous outline as a high-priority fire;
(b5) the railbot driving module 410 checks the center of the fire having the highest priority in the step (b3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142;
(b6) estimating, by the fire detection module 440, the distance between the railbot 100 and the center of the highest priority fire in a preset manner while performing the step (b5);
(b7) At the same time as performing the steps (b5) and (b6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving the railbot 100 so that the distance estimated according to step (b6) does not become less than a preset limit distance while moving it until it is between 95 degrees and -85 degrees;
(b8) the fire detection module 440 includes: i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (b6), estimating the location on the plane of the highest priority fire; and
(b9) The warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and follows a preset rule based on the estimated location on the floor plan. Further comprising the step of turning off the power of one or more equipment according to and stopping the supply and exhaust equipment and the damper,
The step (b6) is,
The fire detection module 440, i) the rail bot 100, the pan angle of the camera module 170 at any one position (A) before starting to move according to the step (b7) and Tilt angle and ii) the pan angle of the camera module 170 at the one position (A) and the other position (B) after the rail bot 100 starts to move according to step (b7) and estimating the distance between the rail bot 100 and the center of the highest priority fire using the tilt angle,
Way.
3. The method of claim 2,
The server unit 400 further includes a flame detection module 450 that determines the presence or absence of a flame on the screen photographed by the camera module 170 using artificial intelligence,
The method is
(c1) determining, by the flame detection module 450, that a flame is generated on the screen photographed by the camera module 170;
(c3) the flame detection module 450 identifies the outline of the group of the portion corresponding to the determined flame, and checks the number of flames using whether the identified outline is discontinuous, and the number of flames is two. judging a group having a larger area of a group bordered by a discontinuous outline as a flame having a higher priority when it is determined as above;
(c5) the railbot driving module 410 checks the center of the flame having the highest priority in the step (c3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142;
(c6) estimating, by the flame detection module 450, the distance between the railbot 100 and the center of the highest priority flame in a preset manner while performing the step (c5);
(c7) At the same time as performing steps (c5) and (c6), when the railbot driving module 410 has a fan angle of the camera module 170 between -85 and 85 degrees, the railbot ( If the angle of the camera module 170 is 85 degrees to 95 degrees or -95 degrees while moving the rail bot 100 to move the rail bot 100 backward 100) and moving the rail bot 100 backward when it is between 95 degrees and -95 degrees. moving the railbot 100 so that the distance estimated according to step (c6) does not become less than a preset limit distance while moving it until it is between degrees and -85 degrees;
(c8) the flame detection module 450, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) estimating the position on the plane of the flame having the highest priority by using the distance estimated in step (c6); and
(c9) The warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and follows a preset rule based on the estimated location on the floor plan. Further comprising the step of turning off the power of one or more equipment according to,
The step (c6) is,
The fire detection module 440, i) the pan angle of the camera module 170 at any one position (A) before the railbot 100 starts moving according to the step (c7), and Tilt angle and ii) the pan angle of the camera module 170 at the one position (A) and the other position (B) after the railbot 100 starts to move according to step (c7) and estimating the distance between the railbot 100 and the center of the highest priority flame using the tilt angle,
Way.
4. The method of claim 3,
The server unit 400 further includes a smoke detection module 460 that determines the presence or absence of smoke on the screen photographed by the camera module 170 using artificial intelligence and identifies the direction of movement of the smoke,
The method is
(d1) the smoke detection module 460 determines that smoke has occurred on the screen photographed by the camera module 170, and uses that smoke is moving from a darker direction to a lighter smoke on the screen. identifying the direction of movement of the smoke;
(d2) identifying, by the smoke detection module 460, a smoke generating point using pixels corresponding to the determined smoke and the identified moving direction of the smoke;
(d5) the railbot driving module 410, the tilt step motor 141 and the pan step so that the smoke generating point identified in step (d2) comes to the center of the screen photographed by the camera module 170 operating any one or more of the motors 142;
(d6) estimating, by the smoke detection module 460, the distance between the railbot 100 and the smoke generating point in a preset manner while performing the step (d5);
(d7) At the same time as performing steps (d5) and (d6), when the railbot driving module 410, the fan angle of the camera module 170 is between -85 and 85 degrees, the railbot Move the rail bot 100 to move the rail bot 100 backward if it is between 95 degrees and -95 and the pan angle of the camera module 170 is 85 degrees to 95 degrees or - moving the railbot 100 so that the distance estimated according to step (d6) does not become less than a preset limit distance while moving it until it is between 95 degrees and -85 degrees;
(d8) the smoke detection module 450, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) estimating a location on a plane of a smoke generating point using the distance estimated in step (d6); and
(d9) The warning module 490 displays the estimated location on a preset floor plan and outputs it through a preset web client terminal 510, and follows a preset rule based on the estimated location on the floor plan. Further comprising the step of turning off the power of one or more equipment according to and stopping the supply and exhaust equipment and the damper,
The step (d6) is,
The smoke detection module 460, i) the pan angle of the camera module 170 at any one position (A) before the railbot 100 starts moving according to the step (d7), and Tilt angle and ii) the pan angle of the camera module 170 at the one position (A) and the other position (B) after the railbot 100 starts moving according to step (d7) and estimating the distance between the railbot 100 and the smoke generating point by using the tilt angle,
Way.
5. The method of claim 4,
The server unit 400 further includes a immersion detection module 470 that determines whether there is immersion on the screen photographed by the camera module 170 using artificial intelligence,
The method is
(e1) determining, by the immersion detection module 470, that immersion has occurred on the screen photographed by the camera module 170;
(e3) the submersion detection module 470 identifies the outline of the group of the portion corresponding to the determined submergence, checks the number of submergences using the discontinuity of the identified outline, and the number of submergence is two judging a group having a larger area of a group bordered by a discontinuous outline as a high-priority immersion if it is determined as above;
(e5) the railbot driving module 410 checks the center of the immersion with the highest priority in the step (e3), and the confirmed center comes to the center of the screen photographed by the camera module 170. operating one or more of the tilt step motor 141 and the pan step motor 142;
(e6) estimating, by the submersion detection module 470, the distance between the railbot 100 and the center of the highest priority submersion in a preset manner while performing step (e5);

(e8) the immersion detection module 470, i) a plane position of the railbot 100, ii) a pan angle of the camera module 170, iii) a tilt angle of the camera module 170, and iv) using the distance estimated in step (e6), estimating the position on the plane of the submerged with the highest priority; and
(e9) The warning module 490 displays the estimated location on a preset floor plan and outputs it through the preset web client terminal 510, and follows a preset rule based on the estimated location on the floor plan. Further comprising the step of turning off the power of one or more equipment according to,
The step (e6) is,
The immersion detection module 470, the pan angle and the tilt angle of the camera module 170 at any one position (A) before the railbot 100 starts to move according to the step (e7). and the pan angle and tilt angle of the camera module 170 at the one position (A) and the other position (B) after the railbot 100 starts moving according to step (e7). Using, including the step of estimating the distance between the railbot 100 and the center of the highest priority immersion,
Way.
6. The method of claim 5,
The server unit 400,
For each of the plurality of facilities, the facility database 409 in which the location of the facility on the plane and the location of the monitoring on the plane of the facility are pre-stored; and
When there is an abnormality in any one of the plurality of facilities, it further includes a facility monitoring module 480 that can identify it using the facility database 409,
The method is
(f1) confirming, by the facility monitoring module 480, that there is an abnormality in any one of the plurality of facilities;
(f5) the railbot driving module 410 operates any one or more of the tilt step motor 141 and the fan step motor 142 so that the equipment identified in the step (f1) comes to the center of the screen step;
(f7) simultaneously performing the step (f5), the railbot driving module 410 moving the railbot 100 to the monitoring position of the equipment stored in advance for the equipment; and
(f9) further comprising the step of outputting, by the warning module 490, the image captured by the camera module 170 of the railbot 100 through the web client terminal 510,
Way.
7. The method of claim 6,
In the facility database 409,
For each of the plurality of facilities, the first upper limit value that is the maximum normal operating temperature, the second upper limit value that is greater than the first upper limit value as the equipment breakage temperature, the first lower limit value that is the normal operating minimum temperature, and the second lower limit value that is smaller than the first lower limit value as the equipment freezing temperature , a first control variable that lowers the facility operation rate to control overheating of the facility, a second control variable that increases the facility operation rate to prevent freezing of the facility, and the second control variable that increases the facility operation rate higher than the second control variable to prevent freezing of the facility The third control variable is pre-stored,
The server unit 400 further includes a facility temperature sensing module 420 and a facility control module 421,
(g11) step of the facility temperature detection module 420, the camera module 170 to check the temperature of the facility on the photographed screen and identify the facility;
(g12) the facility temperature sensing module 420, when the temperature checked in step (g11) is greater than the second upper limit value stored in advance for the facility, the facility control module 421 turns off the facility, outputting a warning to a preset web client terminal 510 through the warning module 490;
(g13) when the facility temperature detection module 420, the temperature checked in step (g11) is greater than the first upper limit value stored in advance for the facility and smaller than the second upper limit value, the facility control module 421 controls the facility changing the control variable of , to a first control variable stored in advance for the corresponding facility, controlling the control variable, and outputting a warning to a preset web client terminal 510 through the warning module 490;
(g14) the facility temperature sensing module 420, when the temperature checked in step (g11) is smaller than the first lower limit value stored in advance for the facility and greater than the second lower limit value, the facility control module 421 controls the facility changing the control variable of , to a second control variable stored in advance for the facility, controlling the control variable, and outputting a warning to a preset web client terminal 510 through the warning module 490; and
(g15) when the facility temperature sensing module 420, the temperature checked in step (g11) is smaller than the second lower limit value stored in advance for the facility, the facility control module 421 corresponds to the control variable of the facility The method further comprises the step of controlling the facility by changing it to a third control variable stored in advance, and outputting a warning to a preset web client terminal 510 through the warning module 490,
Way.
8. The method according to any one of claims 1 to 7,
The real image camera 171 and the thermal image camera 172 of the railbot 100 are oriented in the same direction to acquire a real image and a thermal image screen for the same screen together,
The railbot 100 further includes a communication module 160 that performs a wireless communication function, and wirelessly transmits the real image and thermal image screen acquired to the user terminal 511 through the communication module 160,
The server unit 400 communicates with the Ethernet hub 311 of the repeater unit 310 through Ethernet,
The Ethernet hub 311 and the railbot 100 are wired or wirelessly connected through a copper tube located in the rail 200 , so that the railbot 100 is connected to the web through the Ethernet hub 311 . Transmitting the acquired real image and thermal image screen to the client terminal 510,
Way.
9. The method of claim 8,
The rail bot 100 and the power supply unit 310 are wired through the copper tube in the rail 200,
Way.
10. The method of claim 9,
The rail 200 is installed on the lower surface of the ceiling of the space in the building is installed adjacent to the outer wall of the space,
Way.
11. The method of claim 10,
The railbot 100,
one or more wheels (102) rotating on the rail (200);
a mainboard module 130 for communicating with the repeater unit 310 by wire or wireless connection;
a housing 103 in which the main board module 130 is mounted;
a power roller 110 that is in contact with the rail 200 and is wiredly connected to the power supply unit 310 to supply power;
an encoder operation module 143 connected to the main board module 130 by wire to encode the image captured by the camera module 170;
a motor 144 to which electric power is supplied to rotate the wheel 102;
a driving gear 145 connecting the motor 144 and the wheel 102;
a battery and charging module 146 to which power is supplied to charge power;
a front proximity sensor 151 for detecting an obstacle in front of the railbot 100; and
Further comprising a rear proximity sensor 152 for detecting a rear obstacle of the rail bot 100,
Way.
9. The method of claim 8,
The BAS server 401 is a CCMS (Central Control & Monitoring System),
The BAS server 401 communicates with the facilities 630 in the building through the gateway 610 and DDC (Direct Digital Control) 620,
The facility 630 includes any one or more of an air conditioner, a boiler, a temperature and humidity sensor, a power / peak controller, a lighting, a heat exchanger, a valve, and a fine dust sensor,
Way.
A computer-readable recording medium in which a computer program for executing the method according to any one of claims 1 to 7 is recorded in a computer.
A computer program stored in a recording medium for executing the method according to any one of claims 1 to 7 on a computer.

Images