KR102592101B1 - A system and methods for guiding evacuation route based on artificial intelligence - Google Patents

Abstract

The present invention relates to an optimal evacuation route guidance system and guidance method within an artificial intelligence-based building structure, and more specifically, to obtaining internal structure information of a building structure based on two-dimensional drawing information of the building structure. wealth; an occupant information acquisition unit that detects or receives location information of occupants scattered within the building structure; A disaster occurrence monitoring unit installed inside and outside the building structure to monitor the occurrence of a disaster; A communication unit that transmits/receives the internal structure information of the building structure obtained from the internal structure information acquisition unit of the building structure, the occupant information obtained from the occupant information acquisition unit, and the disaster occurrence signal obtained from the disaster occurrence monitoring unit; an evacuation route information provider configured to be installed or portable at multiple locations indoors so that the information received through the communication unit can be used to provide evacuation route information to occupants or enable occupants to request evacuation route information in the event of a disaster; The internal evacuation route of the building structure is identified based on the information obtained from the internal structure information acquisition department and the occupant information acquisition department of the building structure, and the evacuation route information provision department is provided based on the disaster occurrence signal obtained through the disaster occurrence monitoring department. It consists of an evacuation route guide that provides the optimal evacuation route.

Description

{A system and methods for guiding evacuation route based on artificial intelligence}

The present invention relates to an optimal evacuation route guidance system and guidance method within an artificial intelligence-based building structure, and more specifically, to a deep learning technique that applies hydraulic network modeling built on a server in the event of a disaster at various large-scale construction sites or multi-assembly facilities. This is about an optimal evacuation route guidance system and guidance method within an artificial intelligence-based building structure that can quickly extract and provide an evacuation route optimized for the current location of workers and occupants.

In recent years, natural and human-made disasters have occurred frequently, and not only are the scale of the disaster increasing, but the types of disasters are also caused by a variety of factors.

The present invention provides that when a force majeure accident such as a disaster, a fire or a collapse accident occurs in a specific space or place while there are workers or an unspecified number of people in a certain structure, the personnel remaining inside the structure without external support It is about an evacuation route guidance method that allows people to actively escape from a disaster site.

Generally, when a disaster occurs, a disaster report is made by a witness, etc., and rescue activities are carried out after a certain period of time has elapsed after the disaster occurs and the 119 rescue team arrives.

However, if the arrival of the rescue team is delayed or if the rescue team's access to the disaster site is difficult even after the rescue team arrives, the people trapped at the disaster site must evacuate to a safe zone on their own to preserve their lives, but they must evacuate in a hurry due to the urgent disaster site situation. If you are unable to find a route or are trapped in an obstacle even if you find an evacuation route, it becomes very difficult to escape the disaster scene on your own, and if rescue time is delayed, a situation can easily occur that could result in the loss of life.

In addition, even if rescue teams are dispatched to the disaster site, they often face difficulties in efficient life-saving work if they cannot accurately identify the internal situation of the structure, its structure, or the point of occurrence of the disaster. When rescuers enter the structure when no one is present, even the rescuers become trapped at the disaster site, which often results in more casualties.

Therefore, in the event of a conventional disaster, a system that can respond to disasters in various forms and methods to reduce human casualties by allowing personnel remaining inside the structure to actively evacuate by avoiding the disaster site and minimize human casualties in the event of a disaster Many inventions for evacuation guidance systems are being developed.

Looking at some examples that have been developed in the past, the “shortest evacuation route guidance system using disaster occurrence information” has been proposed in Patent Registration No. 10-2136092 (registration date: 2020.07.15).

In the case of the previously registered patent, it consists of a location information collection device, an environmental information measurement device including a fire detection sensor and a gas sensor, a rescue and evacuation guidance device, and a guidance server connected to these devices through a network, and the guidance server The server is configured to divide the target area into multiple zones and monitor it, so that when a disaster occurs in a specific area, it identifies the shortest exit from the disaster area and visually provides it.

However, the system must track and determine the real-time location of people entering the target area managed by the management server, and the people entering the area have the inconvenience of having to wear a location transmitting device that can collect location information. It was inevitable that feelings would arise.

In addition, since the information provided by the rescue and evacuation guidance device is displayed with arrows or letters, confusion is bound to occur during the evacuation process for occupants who do not accurately understand the structure of the building where the disaster occurred.

In particular, if the inside of the building is large and high-rise, it is difficult to get from the disaster site to the exit with only the flat direction indicator signal displayed on the rescue and evacuation information device, and if the location of the occupants changes frequently, the shortest route from the current location of each occupant is difficult. As the evacuation simulation engine was operated to find an evacuation route, the management server not only had a load on data processing, but also had problems making it difficult to provide prompt guidance.

In addition, the previously registered patent divides the target area into several zones and provides evacuation guidance based on the movement path from each zone to the exit, but each movement path has specific obstacles (internal facilities, stairs, internal structure changes, etc.). ) or if the exit does not function properly, evacuation guidance is provided based only on the already entered exit information, which has the disadvantage of reducing realism.

As another conventional example, a “response system in the event of a building disaster using a two-dimensional building floor plan” has been developed in Patent Publication No. 10-2021-73102 (publication date: 2021.06.18).

In the case of the above published patent, smart IoT and CCTV are installed in the building, and a 2D building floor plan and augmented reality are used to guide evacuation routes and rescue routes to occupants of the building and external rescuers in the event of a disaster. It is designed to disseminate disaster situations and guide evacuation routes through mobile devices.

However, in the case of the above-mentioned published patent, the place where a disaster occurred, the evacuation route for occupants in the space, and a specific method for guiding evacuation routes and exits to occupants in other spaces are not provided, so it cannot be easily implemented by those skilled in the art. It introduces a disaster response system at the level of an abstract idea in which the composition of the invention is not clearly described, making its application difficult in reality.

Meanwhile, the present invention seeks to use the ‘Hydraulic network’ model technique to improve the problems of the conventional technology and provide more efficient evacuation guidance to occupants.

The 'Hydraulic network' model technique used in the present invention is a hydraulic design network for building a water supply network for water supply facilities that is essential for supplying water to many users making up a city. It can be said to be a model technique that simplifies the structure.

In addition, this 'Hydraulic network' model technique is used to analyze the process of compressed air flowing from a high pressure point to a low pressure point according to the pressure difference in the internal flow path of a gas turbine with various shapes and structures. Also, an embodiment of this hydraulic network model is shown in Figure 1.

In the case of a hydraulic network that constitutes a hydraulic design network, the continuous state of connection between water users and the pipes connecting each user is indicated by a continuous line, and the various characteristics of the water passing through the connected pipes are displayed as a continuous line. Loss, that is, the friction coefficient of the pipe, the length or diameter of the pipe, hydraulic pressure, flow rate, and head loss, etc. are taken into consideration to suggest a plan for optimally constructing a water distribution network.

In addition, these hydraulic networks classify structures based on the shape characteristics of specific structures composed of various shapes and complex structures, and limit the flow of certain physical quantities that make up the structure by expressing them as resistance elements. By doing so, it is a technique that can model the flow of physical quantities from any point to another by organizing the entire system of interest into one network.

Focusing on the fact that in this 'Hydraulic network' model technique, the most desirable path through which the movement of physical quantities can be natural is the path that connects to the element with the lowest resistance, the present invention proposes the above-described hydraulic network. The modeling technique was applied to the present invention considering that the most desirable flow (direction) of physical quantities can be used to derive the most efficient evacuation route in the event of a disaster within the structure.

In other words, looking at how to find an emergency evacuation route within a building structure, the interior of the building structure is composed of multiple walls, doors, stairs, passages, windows, doorways, and internal facilities (hereinafter referred to as 'resistance elements'). Therefore, based on the two-dimensional drawing of the building structure, the overall shape can be diagrammed by dividing it into inlet, outlet, and resistance elements.

In addition, each schematic resistance element is given a weight for each resistance element according to the area or distance of the elements that determine its characteristics, the ease of movement of the occupants, etc., so that the entire building structure can form a huge network composed of multiple resistance elements. , using this concept, the 'Hydraulic network' model technique can be applied to derive evacuation routes for workers or occupants within complex building structures.

- Patent Registration No. 10-2136092 (Registration Date: 2020.07.15) - Patent Publication No. 10-2021-73102 (Publication date: 2021.06.18) - Patent Publication No. 10-2021-85323 (Publication Date: 2021.07.08)

The present invention was developed to overcome the shortcomings of the existing disaster evacuation route guidance system and guidance method. The structure and resistance elements of the relevant building structure are modeled using artificial intelligence techniques, and the latest information is reflected in real time to prepare for the occurrence of a disaster. The purpose is to provide an optimal evacuation route guidance system and guidance method within an artificial intelligence-based building structure that can provide quick and optimal evacuation routes to occupants while reducing the load on the evacuation route information department.

In order to achieve the above-mentioned purpose, the artificial intelligence-based optimal evacuation route guidance system within a building structure according to the present invention includes an internal structure information acquisition unit of the building structure that acquires the internal structure information of the building structure based on the two-dimensional drawing information of the building structure. ; an occupant information acquisition unit that detects or receives location information of occupants scattered within the building structure; A disaster occurrence monitoring unit installed inside and outside the building structure to monitor the occurrence of a disaster; A communication unit that transmits/receives the internal structure information of the building structure obtained from the internal structure information acquisition unit of the building structure, the occupant information obtained from the occupant information acquisition unit, and the disaster occurrence signal obtained from the disaster occurrence monitoring unit; an evacuation route information provider configured to be installed or portable at multiple locations indoors so that the information received through the communication unit can be used to provide evacuation route information to occupants or enable occupants to request evacuation route information in the event of a disaster; The internal evacuation route of the building structure is identified based on the information obtained from the internal structure information acquisition department and the occupant information acquisition department of the building structure, and the evacuation route information provision department is provided based on the disaster occurrence signal obtained through the disaster occurrence monitoring department. and an evacuation route information unit that provides an optimal evacuation route, wherein the evacuation route information unit receives and stores the building structure information and occupant information obtained from the internal structure information acquisition unit and the occupant information acquisition unit of the building structure. Classify, learn the stored and classified data, identify and analyze the movement line and internal evacuation route of the occupants for each floor where the occupants are located based on the data learning, and learn the data and determine the location of the occupants and the internal evacuation route. Based on the structural analysis results, predict and determine the optimal evacuation route for each occupant location on each floor in advance, obtain a disaster signal from the disaster occurrence monitoring unit, and request evacuation route guidance for occupants through the evacuation route information provision unit. Automatically provides optimal evacuation route information for each occupant's location based on the prediction and judgment results, and when predicting and judging the evacuation route in advance, distribution status by each floor and location of occupants and movement of occupants according to the disaster situation. It is configured to predict and determine the optimal evacuation route based on the movement path and movement obstacle elements located in the movement line of the occupants.

In addition, the evacuation route guide is a movement obstacle element used when learning the data, such as the number of occupants, passage width, location of occupants, or walls, artificial facilities, doors, and stairs existing in the passage from the movement line to the entrance. , It consists of a configuration that includes distance data to the entrance.

In addition, when learning the data, the evacuation route guide unit extracts the new equipment or artificial facility as a movement obstacle element when new equipment or artificial facility is placed in the main passageway inside the building structure, and extracts the extracted movement obstacle element. It is configured to predict and determine the optimal evacuation route for each occupant's location by adding to the composition of the movement disability elements set in the existing learning model and relearning the new added movement disability elements.

Meanwhile, the optimal evacuation route guidance method within an artificial intelligence-based building structure according to the present invention to achieve the above-described purpose includes the steps of acquiring internal structure information of the building structure based on two-dimensional drawing information of the building structure; Obtaining data on movement disorder elements arranged on a movement path among the data obtained in the internal structure information acquisition step of the building structure; Occupant indoor location information acquisition step of acquiring indoor location information of occupants using the building structure; Storing and classifying the acquired internal structure information of the building structure, movement disorder element data, and occupant location data; learning the stored and classified data; Predicting and determining an evacuation route in advance for each location or movement line of the occupants based on the data learning and the results of the identified movement disorder elements; A disaster occurrence information acquisition step of acquiring disaster occurrence information through a disaster occurrence monitoring unit installed inside and outside the building structure; When obtaining the disaster occurrence information, calculating the shortest evacuation route for each occupant based on the prediction and judgment results of the evacuation route for each occupant's location or movement line; When obtaining the disaster occurrence information, the calculated evacuation route information is transmitted to the evacuation route information provider carried or worn by the occupants or placed inside the building structure, so that the occupants evacuate quickly and safely through the evacuation route suggested by the evacuation route information provider. It includes the step of transmitting evacuation route information.

In particular, in the step of acquiring the movement disability element data, movement disability weights are differentially applied to each movement disability element according to the characteristics of artificial facilities placed on the movement passage, classified and stored, and the evacuation route is determined in advance. In the prediction and judgment step, the evacuation route is predicted and determined using the movement disorder element data with different weights obtained in the movement disorder element data acquisition step.

In addition, the movement obstacle element data acquisition step and the evacuation route prediction and determination step are based on a hydraulic network modeling technique that displays the efficient flow and distribution of fluid as a flat network based on a number of material points (nodes) and resistance elements. It consists of features.

The artificial intelligence-based optimal evacuation route guidance system and guidance method within a building structure according to an embodiment of the present invention obtains data on the internal structure of the building structure based on the two-dimensional drawing information of the building structure, and provides efficient evacuation route according to the location of the occupants. Evacuation routes are provided so that occupants can evacuate safely in the event of a disaster.

In particular, in the present invention, in the process of predicting and determining the occupant's evacuation route in advance, weights are differentially assigned according to the characteristics of the movement obstacle elements that are placed on the movement path connecting each location of the occupant to the entrance and exit and provide resistance to movement. By predicting and determining the optimal evacuation route in advance according to the location and movement line of occupants, it is possible to minimize casualties by reducing the evacuation time escalation when a large number of occupants are concentrated in one place in the event of a disaster.

Figure 1 shows an embodiment of a hydraulic network model used in the water treatment field.
Figure 2 is an example of a plan view of a multi-story building structure applied to the optimal evacuation route guidance system in an artificial intelligence-based building structure according to an embodiment of the present invention;
Figure 3 is an embodiment showing the planar structure of the multi-story building structure of Figure 2 as a hydraulic network model;
Figure 4 is a block diagram of the main components of the optimal evacuation route guidance system within an artificial intelligence-based building structure according to an embodiment of the present invention;
Figure 5 is a flow chart for explaining the optimal evacuation route guidance method within an artificial intelligence-based building structure according to an embodiment of the present invention;
Figure 6 is a flow chart showing an optimal evacuation route guidance method within an artificial intelligence-based building structure according to a modified embodiment of the present invention;
Figure 7 is an example of the internal structure of an architectural structure defined for calculating the optimal evacuation route through the hydraulic network model technique applied to the present invention.

The present invention provides a system and guidance method (hereinafter abbreviated as “evacuation route guidance system and guidance method”) that finds the optimal evacuation route using hydraulic network modeling techniques and provides guidance to occupants on each floor when a disaster occurs within a building structure. It's about.

Before explaining specific embodiments of the present invention, the concept of hydraulic network modeling and its application techniques used in the method of finding the optimal evacuation route to be achieved in the present invention will be briefly explained.

The present invention relates to an evacuation route guidance system and guidance method for application to multi-story complex facilities (hereinafter collectively referred to as architectural structures) such as offices, complex shopping centers, and shopping malls. Walls, doors, stairs, passages, windows, entrances, and other internal facilities (hereinafter referred to as 'movement obstruction elements') are extracted from the 2D drawing and display a plurality of material points (nodes) and flow paths applied to the hydraulic network model. A network schematized in the form of a network is constructed, and the characteristics of the plurality of movement obstacle elements, that is, the area occupying space, size, etc., are calculated as movement obstacle factors during evacuation, and the movement is performed according to the movement location and movement line of the occupants. By assigning weights of different sizes to the obstacle elements, we model a network that displays the movement path inside the building structure and each element and entrance as a connecting line, and an evacuation route, which is the optimal or shortest movement line, is determined depending on the location of the occupants. Let us suggest a way to find it.

For example, when trying to calculate an evacuation route based on an architectural structure, there are entrances and exits on both sides of an empty space, multiple walls and doors located between the walls, and a wall in another space divided by the walls or doors. The evacuation routes provided inside the structure with passages will inevitably be different.

The reason is that in the case of an empty space, the distance between the entrance and the door provided in the space where the occupants are located is an important factor in finding an evacuation route, but the number of doors located between multiple walls and the walls, and the In an architectural structure equipped with walls, passages, stairs, and internal facilities in other spaces, the type and number of doors, walls, and internal facilities located in the middle of the evacuation route connecting the specific space where the occupants are located to the entrance, and the distance traveled, etc. Since it can be an element of movement obstruction during evacuation, when guiding occupants to the optimal evacuation route, the multiple movement obstruction elements present inside these building structures act as resistance elements for evacuees.

In particular, in large buildings or multi-use complexes where the interior of an architectural structure is more complex than when it is simple and is divided into various spaces and has multiple walls, doors, and internal facilities placed in passages, etc., architectural structures are used. In order to find a systematic way to calculate an optimal evacuation route while avoiding complex internal structures or indoor facilities for efficient evacuation of occupants in the event of an internal disaster, the present invention applies different weights to the movement obstacle elements provided in the movement passage. An evacuation route guidance method using a hydraulic network model is proposed.

That is, in the hydraulic network model, a plurality of nodes and a line, which is a flow path connecting each node, are various obstacles (the present invention) that exist on the movement path within the building structure to which the present invention is applied. In , it is expressed as a 'movement disorder element') and the movement line from the space where the occupant is located to the entrance.

In particular, the present invention builds a planar network of the entire building structure by simplifying the internal structure into a plurality of nodes and movement disorder elements based on a two-dimensional drawing of a specific architectural structure, and then updates the already established network at regular intervals to create a realistic network network. , and assigning different weights in proportion to the resistance elements of each node and movement disorder element of the constructed network to find the shortest movement route from the location where the occupants are located to the entrance, calculate the optimal evacuation route, and provide it to the occupants. Provides a way to do this.

Hereinafter, the optimal evacuation route guidance system and its guidance method according to an embodiment of the present invention will be described in detail with reference to the attached drawings.

The terms used in the detailed description of the present invention are general terms widely used in the art in consideration of the structural functions of the invention, and the meaning of the terms is interpreted based on the content throughout the detailed description.

Figure 2 shows an example of a plan structural diagram of a multi-story building structure applied to an embodiment of the present invention.

In addition, Figure 3 is an embodiment showing the planar structure of the multi-story building structure of Figure 2 as a hydraulic network model, and Figure 4 is the main configuration of the optimal evacuation route guidance system in an artificial intelligence-based building structure according to an embodiment of the present invention. It is briefly shown as a block diagram.

As shown in Figure 4, the evacuation route guidance system 100 according to an embodiment of the present invention provides information on the internal structure of the building structure based on the two-dimensional design drawing of the building structure as shown in Figure 2. A building structure internal structure information acquisition unit 110 where necessary internal structure information of the building structure is input, an occupant information acquisition unit 120 for acquiring location information of occupants working or using the inside of the building structure, and an architectural structure A disaster occurrence monitoring unit 130 installed on the inside and outside of the building to monitor and understand various disaster conditions occurring in the building structure, and a communication unit 140 that transmits and receives information input from each of the above information acquisition units and the disaster occurrence monitoring unit, It also includes an evacuation route information unit 160 that determines the disaster state of the building structure based on various information received through the communication unit 140 and calculates and provides the optimal evacuation route for each occupant's location when a disaster situation occurs.

Here, the building structure internal structure information acquisition unit 110 may include a scanner or computer device that can scan or manually input the structural diagram of the building structure displayed on the floor plan of the building structure.

The internal structure information of the building structure obtained from the building structure internal structure information acquisition unit 110 is converted and processed into a form that can construct a hydraulic network model through 2D modeling and digital work through a computer program. .

In addition, the occupant information acquisition unit 120 is a human body detection device that can acquire the real-time location of regular occupants who work inside the building structure and occupants who enter and exit from time to time, including dynamic detection sensors, image sensors, infrared sensors, and GPS. It is composed of one of a sensor and an integrated sensor including these, and is configured to obtain information about occupants in real time.

Of course, it is preferable that the occupant information acquisition unit 120 is installed in one or more places inside the building structure and is arranged in the main passage or space on each floor so that the number of people inside can be effectively determined.

Additionally, the occupant information acquisition unit 120 may include any one of a wearable device that the occupant can wear, a personal identification card, and a biometric sensor installed at the entrance.

In addition, the disaster occurrence monitoring unit 130 includes an indoor disaster monitoring unit 131 that monitors disasters occurring inside the building structure, and an outdoor disaster monitoring unit 132 that monitors disasters occurring outside the building structure. Includes.

The indoor disaster monitoring unit 131 and the outdoor disaster monitoring unit 132 are sensor devices installed inside and outside the building structure to monitor various disaster conditions, for example, a temperature sensor, a fire detection sensor, an illumination sensor, a gas sensor, It includes, but is not limited to, one or more of a power sensor, an impact sensor, a positioning sensor, and a dust sensor.

In addition, the communication unit 140 is a device for transmitting and receiving basic information of the building structure, occupant information, and information obtained from the disaster occurrence monitoring unit installed inside and outside the building structure to the evacuation route information unit 160, and is a wired and wireless mobile communication module. , includes one or more of a wireless Internet module, a short-range communication module, and a location information module.

The communication modules of the communication unit 140, that is, mobile communication modules, wireless Internet technology, short-distance communication modules, etc., are widely used communication technologies, and technical standards and communication methods for performing communication are adopted according to each communication module. It is configured to transmit and receive wireless signals to an external terminal or management server through a previously established communication network. In the detailed description of the present invention, detailed descriptions of various previously known communication methods and communication modules will be omitted.

The wireless signal is configured to transmit and receive various types of data according to the transmission and reception of voice call signals, video call signals, or text/multimedia messages.

Additionally, the location information module is a module for acquiring the user's mobile terminal location (or current location) and may include a GPS module, Wi-Fi module, beacon module, etc.

For example, when the occupant's mobile terminal uses a GPS module, it acquires its own location using a signal sent from a GPS satellite, and sends the acquired location information to the evacuation route information unit 160 through the communication unit 140. It can be provided as .

In addition, when using the Wi-Fi module provided in the occupant's mobile terminal, one's own location is acquired based on the information of the Wi-Fi module and the wireless AP (Wireless Access Point) that transmits or receives wireless signals. Information may be provided to the evacuation route information unit 160 through the communication unit 140.

In addition, the evacuation route guidance system 100 according to an embodiment of the present invention may further include an evacuation route information provider 150 installed or provided throughout the interior of the rigid structure.

The evacuation route information provider 150 connects the communication unit 140 to an evacuation route receiver 151 that receives evacuation route data received from the evacuation route information unit 160, and an evacuation route information provider 160 for occupants. It may be composed of an evacuation route request unit 152 that requests an evacuation route, and is equipped with a speaker-integrated monitor that can show the received evacuation route to the occupants by voice or two-dimensional/three-dimensional image. In addition to portable communication devices, it can be a fixed type installed on the wall of a building structure or a portable type that can be removed and carried by the occupant at will, for example, a laptop computer, tablet PC, etc.

Meanwhile, the evacuation route information unit 160 includes a data storage unit 161 that stores data on the received internal structure information of the building structure, location information of occupants, and disaster occurrence information, and a data storage unit 161 that stores data on the received internal structure information of the building structure. A data classification unit 162 that classifies data on information, location information of occupants, and disaster occurrence information, a learning unit 163 that learns stored and classified data, and the internal structure and movement of each floor based on data learning. An analysis unit 164 that analyzes obstacle elements, occupant locations, and disaster occurrence data, and an evacuation route prediction/judgment unit that predicts and determines the optimal evacuation route in advance based on data learning and analysis results for each floor/occupant location. It is composed of a configuration including (165) and a control management unit (166) that manages to provide evacuation route information to the evacuation route provision unit in the event of a disaster based on the evacuation route prediction/judgment results.

In addition, the evacuation route information unit 160 may further include a monitoring unit 167 that displays the evacuation route identified through the analysis unit 164 or the status of each occupant's location in real time as two-dimensional or three-dimensional information. there is.

In addition, the learning unit 163 of the evacuation route guidance unit 160 is configured to learn data using an artificial neural network (ANN).

Here, artificial neural network (ANN) refers to a deep neural network (DNN) composed of multiple hidden layers between one input layer and an output layer.

For example, the deep neural network (DNN). It may include, but is not limited to, convolutional neural networks (CNNs), recurrent neural networks (RNNs), deep belief networks (DBNs), deep autoencoders, etc. No.

In addition, the learning method performed in the learning unit 163 is performed using any one of supervised learning, semi-supervised learning, and unsupervised learning.

Here, the learning of the artificial neural network is to minimize the error in the output. Learning data is repeatedly input into the artificial neural network, the error of the output and target value of the artificial neural network for the learning data is calculated, and the artificial neural network is used to reduce the error. A method of adjusting the weight of each node of the artificial neural network by backpropagating the error of the neural network from the output layer of the artificial neural network to the input layer can be used.

Therefore, when learning data, the learning unit 163 of the evacuation route guide unit 160 determines the length of the internal passage of each floor of the building structure, which is used as a resistance element of the hydraulic network model, and the space divided by the wall in the passage. Data related to the number of doors for entering and exiting, the number and size of mobility-impairing elements located in the passageway, the distance from the occupant's location to the entrance, the number of stairs, and the number of occupants can be learned.

In addition, when learning data from the learning unit 163, the optimal evacuation route can be learned for each location of occupants on each floor of the building structure, the number of occupants, and the type of movement obstacle element present in the passage.

In addition, when learning data in the learning unit 163, different weights are assigned in accordance with the number and type of movement obstacle elements installed in the movement path and the area of the movement path occupied by the movement obstacle elements to determine the optimal evacuation route. You can learn to calculate and predict.

In addition, when learning data in the learning unit 163, one of the first conditions including occupants located in the closed space of each floor, occupants located in the movement passage of each floor, and movement of each floor Calculate the optimal evacuation route in each situation where any one of the second conditions, including the type of movement obstacle element existing in the passage, the number of movement obstacle elements, and the distance from the middle point of the movement passage to the entrance, overlaps with each other. You can learn to predict.

In addition, when learning data in the learning unit 163, it is possible to learn in advance to predict different optimal evacuation routes corresponding to the total number of occupants and the location of occupants.

In addition, when learning data from the learning unit 163, it learns to predict or calculate the optimal evacuation route in advance according to the location of occupants on each floor in response to the type and location of the identified disaster situation and the elapsed time of the disaster occurrence. can do.

In addition, when predicting and calculating the evacuation route, the analysis unit 164 of the evacuation route information unit 160 determines the number of occupants on each floor, the width of the evacuation route, the distance to the entrance, the movement path of the evacuation route, or The optimal evacuation route for each floor can be comprehensively analyzed by assigning separate weights according to the type and number of mobility obstacles present around the entrance.

In addition, the prediction/judgment unit 165 of the evacuation route information unit 160 predicts/determines the evacuation route in advance, based on data learning and analysis results of the number of occupants and the type and number of movement disorder elements. It is possible to predict and determine the movement line of a person in advance, and predict and determine the real-time evacuation route for each floor.

In addition, the prediction/judgment unit 165 of the evacuation route information unit 160 predicts/determines the evacuation route in advance based on data learning and analysis results of the number of occupants and the type and number of movement disorder elements. You can predict and determine the shortest evacuation route according to the disaster situation, and predict and determine the optimal evacuation route for each floor according to the disaster situation.

Meanwhile, the evacuation guidance control unit 166 of the evacuation route information unit 160 establishes an evacuation route to enable evacuation to the main entrance of each floor when the number of occupants on each floor is below the standard value as a result of prediction and determination of the evacuation route. I can guide you.

In addition, the evacuation guidance control unit 166 of the evacuation route information unit 160, when the number of occupants on each floor exceeds the standard value as a result of predicting and determining the evacuation route, disperses the occupants to the auxiliary entrance as well as the main entrance of each floor. Evacuation route guidance can be controlled to enable evacuation.

In addition, the evacuation guidance control unit 166 of the evacuation route information unit 160 uses different evacuation routes for each occupant's location when the number of occupants on each floor exceeds the standard value as a result of predicting and determining the evacuation route. You can control it to evacuate.

In addition, the evacuation guidance control unit 166 of the evacuation route guidance unit 160 predicts and determines that, as a result of predicting and determining the evacuation route, it is predicted that movement obstacle elements exist on the main movement passage of each floor, making evacuation difficult at normal movement speeds. In this case, evacuation route guidance can be controlled so that occupants can evacuate to different exits depending on their location.

In addition, the evacuation guidance control unit 166 of the evacuation route information unit 160 allows occupants to evacuate to an alternative evacuation route when a movement obstacle element exists or is closed at the entrance from which the occupants are trying to evacuate based on the evacuation route prediction and judgment results. You can control it.

In addition, the evacuation guidance control unit 166 of the evacuation route guidance unit 160 guides and controls the evacuation route by preferentially guiding the evacuation route to the floor where the disaster situation occurred, and guides and controls the floor where the disaster situation occurred. Priority information can be generated and guided and controlled so that evacuation route guidance is sequentially provided starting from the adjacent floor.

In addition, the evacuation guidance control unit 166 of the evacuation route guidance unit 160 is configured to transmit a disaster situation occurrence signal to an external rescue organization or administrator when a disaster situation occurs and an evacuation route is provided to each floor. (140) can be controlled.

In addition, when a disaster situation occurs and an evacuation route is provided to each floor, the evacuation guidance control unit 166 of the evacuation route information unit 160 sends an evacuation guidance message or evacuation route image information to the evacuation route provided on each floor. The communication unit 140 can be controlled to transmit to the information provider 150.

In addition, when the evacuation guidance control unit 166 of the evacuation route information unit 160 transmits an evacuation guidance message or evacuation route image information to the evacuation route information provision unit 150 provided on each floor, the number or movement of occupants is Based on the analysis results of the number of obstacle elements, customized evacuation route information can be created and provided to occupants.

In addition, when generating and providing an evacuation guidance message, the evacuation guidance control unit 166 of the evacuation route information unit 160 uses text, voice (sound), light (light source), etc. according to the location of the occupants and the number of occupants. Any one of the vibrations can be generated and provided.

In addition, when the evacuation guidance control unit 166 of the evacuation route information unit 160 generates and provides an evacuation guidance message, the content of the guidance message provided to the floor where the disaster occurred and other floors is generated differently, It can be provided to ensure safe and efficient evacuation for each floor.

And, as the number of occupants on each floor increases when learning data, the evacuation guidance control unit 166 of the evacuation route information unit 160 relearns the data with the increased number of people in the previously learned model to evacuate. Route prediction and judgment can be made, and analysis and guidance can be provided so that occupants can evacuate using the predicted and determined evacuation route based on the enhanced data.

In addition, the evacuation guidance control unit 166 of the evacuation route guidance unit 160 can automatically control the indoor ventilation device when predicting and determining the optimal evacuation route according to the number of occupants on each floor. .

In this way, the present invention allows re-learning by adding new data to the preset or pre-predicted learning data according to the number of occupants and the changed number of occupants on each floor, thereby providing guidance and evacuation routes according to the number of occupants. Make information provision more efficient.

In addition, the present invention uses artificial intelligence to provide learning data according to the identified indoor structure of the building structure according to the number of occupants, the type or size of movement obstacle elements present in the movement passage, the location of the occupant, and the movement line of the occupant. Based on this, the evacuation routes of occupants are efficiently distributed and guided so that safe evacuation routes that can minimize human casualties in the event of a disaster are predicted and determined, and guidance is provided.

In addition, the present invention takes into account the floor where the disaster situation occurred, the number of occupants, the distribution status of occupants, etc., and differentiates them according to emergency priority, so that disaster situation guidance and evacuation route guidance can be provided, so that a large number of people can travel at the same time in a multi-unit building. Reduce the risk of various safety accidents that may occur due to evacuation.

Figure 5 simply shows a flowchart for explaining the optimal evacuation route guidance method within an artificial intelligence-based building structure according to an embodiment of the present invention.

As shown in Figure 5, the optimal evacuation route guidance method according to an embodiment of the present invention first obtains basic information about the internal structure of the building structure and obtains information about the occupants (S100).

Acquiring the internal structure information of the building structure is obtained based on the floor plan of the building structure or updated in real time during use of the building structure. Information on the internal structure of the building structure is collected from each component applied to the hydraulic network model, that is, each In addition to basic information on spatial structure information for each floor and movement obstruction elements that exist on the internal evacuation route of the building structure, such as movement passages, doors, stairs, and internal facilities, it can further include information on the interior structure of the building structure that is changed after construction. .

In addition, the information acquisition of the occupants can obtain information on the number of occupants residing on each floor, the occupant status of non-resident personnel, and the occupants' movement lines.

This information acquisition of occupants is done through occupant information acquisition units installed on each floor, that is, consisting of one or more of image sensors, infrared sensors, CCTV, and GPS sensors that can detect the movement or shape of occupants. Understanding is achieved.

Next, the present invention stores and classifies the acquired indoor structural information and occupant information of the building structure (S110), and a process of learning the stored and classified data through a hydraulic network model is performed (S120).

The contents of the learning process (S120) include the indoor space structure for each floor, the distance and width of the movement passage, that is, the distance from each room to the entrance, and the types of movement disorder elements that exist on the movement passage from each room to the entrance. and features can be learned.

Next, based on data learning based on the hydraulic network model, the location and movement of occupants can be analyzed for each floor or room, and the internal structural status of the building structure can be identified and analyzed ( S130).

In addition, based on the above data learning, it is possible to identify and analyze the possibility of acting as a mobility obstacle according to the type and characteristics of the mobility obstacle element that exists in the movement line of occupants on each floor.

In addition, based on the above data learning, the movement line for the evacuation route leading to the movement passage and the evacuation route leading to the entrance via the movement passage is identified based on the arrangement structure or opening and closing method of the door for each room separated by a wall inside the building structure. It can be analyzed.

In addition, based on the data learning, evacuation guidance messages and evacuation route images suitable for occupants of each floor can be selectively or differentially provided.

Next, based on data learning, identification of occupants, identification of the internal structure of the building structure, and analysis results, the evacuation route to be provided in the event of a disaster for each floor can be predicted and determined in advance (S140).

In the evacuation route prediction and judgment process (S140), based on data learning, identification of occupants, identification of the internal structure of the building structure, and analysis results, the distribution status of occupants on each floor and the movement line of occupants are predicted and judged, and the movement route of occupants is predicted and determined. It is possible to predict and determine in advance the mobility disability weight for each mobility obstacle element according to the number of existing mobility obstacle elements, the occupied area of the space, and the distance from each room to the entrance.

The present invention allows predicting and determining the priority of evacuation routes through which occupants of each floor can evacuate via the optimal movement route for each type of disaster or location of occurrence.

In addition, the present invention assigns different mobility obstacle weights to each mobility obstacle element according to the type and characteristics of the mobility obstacle element present in the movement passage based on data learning, identification of occupants, and internal structure of the building structure and analysis results. Therefore, it is possible to predict and determine in advance the movement line to the entrance and the evacuation route that can be evacuated in the shortest time from each room to the entrance according to the location of the occupants on each floor.

Such data learning and evacuation routes for each floor are predicted/determined in advance, and when a certain signal is received through the disaster detection unit, it is determined whether a disaster has occurred (S150), and if it is determined that a disaster has occurred, the predicted and determined The evacuation route for each floor is provided to occupants on each floor (S160).

Of course, if it is determined that a disaster has not occurred as a result of analyzing the signal received from the disaster detection unit, data learning using the above-mentioned learning model is repeated, and when new information is input, the evacuation route for each floor is established through reinforced learning. The process of predicting/judging in advance is carried out continuously to prepare for the occurrence of a disaster.

When a disaster situation occurs, the evacuation route guidance (S160) generates a notification message based on the number of occupants for each floor and the occupant's movement route information and evacuates to the occupant's portable communication device or the evacuation route information provider provided on each floor. Route information can be transmitted.

In particular, when a disaster situation occurs, evacuation route guidance can be provided first to occupants of the floor where the disaster occurred, and then evacuation route guidance can be provided sequentially starting from the floor closest to the floor where the disaster occurred. It can be done.

Here, based on the evacuation route prediction and judgment results identified for each floor, the first floor includes the floor with the largest number of occupants, the floor with the largest movement distribution of occupants, and the floor with the largest number of movement disorder elements existing on the movement passage. The evacuation route for each floor in situations where one of the conditions overlaps with one of the second conditions, which includes the floor closest to the floor where the disaster occurred, the floor with the longest evacuation route, and the floor with the fewest number of exits. Guidance can be provided by prioritizing guidance.

In addition, the present invention transmits a disaster notification message or video to an external rescue organization in the event of a disaster, guides rescuers to the evacuation route for occupants on each floor to enable rapid rescue operations, and provides emergency response to a disaster. After evacuation route guidance is provided, the standby state is maintained until the disaster situation is lifted or the evacuation route guidance is terminated (S170).

In this way, the present invention uses artificial intelligence to guide different evacuation routes for each floor according to the internal structure of the building structure, the number of occupants, the movement line of the occupants, and the type and number of movement obstacle elements, thereby providing quick and safe guidance. Ensure that evacuation can take place.

Figure 6 is a flowchart of a temporary optimal evacuation route guidance method according to a modified embodiment of the present invention.

In the optimal evacuation route guidance method according to a modified embodiment of the present invention, basic information of the building structure is acquired through an information acquisition unit (S200).

The basic information used in the present invention may include information on the internal structure of the building structure, the number of occupants remaining in the building structure, and the basic movement lines of the residents.

Occupant information is obtained through the occupant information acquisition unit, and the occupant information acquisition unit can obtain information on the number of occupants who will reside when acquiring the internal structure through various sensors or cameras installed indoors and the internal structure information acquisition unit.

In addition, the information on the internal structure of the building structure is based on the floor plan of the building structure, including the indoor space structure of each floor, that is, walls, spaces provided with walls, doors, movement passages, entrances, and movement obstruction elements existing in the movement passageways. etc. may be included, and such information on the internal structure of the building structure can be obtained by converting planar information using a scanner device and a planar structure diagram into three-dimensional information through a computer program.

Next, the present invention classifies and stores the obtained basic information (S210), and performs artificial intelligence learning on the classified and stored information through a learning model (S220).

In addition, if there are more occupants than the number of residents entered as the basic information, additional information on non-resident occupants can be obtained (S230), and if a mobility disability element placed indoors is added, information on the added mobility disability element can be obtained. Can be obtained (S240).

Enhanced learning can be performed by resetting the preset classification criteria based on the additionally acquired occupant information, mobility disability element information, and classification criteria set in the learning model (S250).

Based on this enhanced learning, the process of predicting/determining the number of occupants on each floor and the evacuation route according to the movement line by identifying the number of occupants on each floor and the number of movement disorder elements (S260) is repeated. do.

When a certain signal is received from the disaster monitoring unit while the evacuation route is being predicted/determined in advance through such data learning and reinforced learning, the received signal is analyzed to determine whether a disaster has occurred (S270). If it is determined that a disaster has occurred, the disaster detection information can be updated with the previously learned information so that the evacuation route can be quickly guided through the evacuation route information provision unit provided on each floor or the portable communication device carried by each occupant. Do it (S280).

In other words, the present invention predicts/determines in advance the basic indoor structural information of the building structure, the number of occupants, and the evacuation route according to the movement of the occupants, and in the event of a disaster, the evacuation route information provider provided on each floor or each occupant carries it. Evacuation routes can be provided using portable communication devices.

In addition, if the number of non-residential personnel increases or mobility disability elements are added in addition to the basic information initially obtained, the evacuation route is predicted/judgmented through reinforcement learning of the preset learning model to determine the evacuation route corresponding to the occupants or mobility disability elements. By calculating/providing , safe evacuation of occupants can be achieved.

The present invention provides an evacuation route that can evacuate by dispersing the occupants when guiding the evacuation route, when the number of occupants exceeds a preset standard or when the weight of the movement obstacle increases due to the addition of a movement obstacle element to the movement passage. This allows us to minimize casualties in the event of a disaster.

In addition, when providing the evacuation route information, information on the occurrence of a disaster and the evacuation route of the occupants is provided not only to the occupants but also to external rescue organizations so that effective rescue operations can be carried out. After the evacuation route guidance, the system is switched to a standby state or This ends (S290).

On the other hand, the hydraulic network model used to calculate and guide the optimal evacuation route within the building structure of the present invention described above is a network model as shown in Figure 3 (the circular mark in Figure 3 is movable) depending on the internal structure of the building structure and the type of movement obstacle element. An example of calculating the optimal evacuation route for each floor of a building structure through this hydraulic network model technique is shown below with reference to FIG. 7. This is briefly explained.

Figure 7 shows an internal cross-section of a simple architectural structure. As shown in Figure 7, an example of a case where an occupant (P ₁ ) is at a random position in a specific space (SP ₁ ) among the indoor spaces of the building structure. For example, assuming that an occupant (P ₁ ) at a random location in a specific space (SP ₁ ) is a certain distance (S ₁ ) away from the center point (C ₁ ) of the specific space, the location of the occupant (P ₁ ) The distance from the door (R _D1 ) is S ₁ + S ₂ , and the escape routes that can be taken from the door (R _D1 ) are path 1 (Pt ₁ ) and path 2 (Pt ₂ ).

If you select route 1 (Pt ₁ ), you will encounter two or more passages, two or more doors (R _D1, R _{D2, …} ), and one or more stairs (R _St1 ), which are considered mobility obstacles that impede evacuation. When path 2 (Pt ₂ ) is selected, the length of the passage is slightly longer than path 1 (Pt ₁ ), but the movement obstacle elements that must be passed are one door (R _{D1) and} one corner (turn). , only the entrance (R _E1 ).

Therefore, when calculating the shortest evacuation route using hydraulic network modeling techniques, the route that can evacuate in the shortest time is selected by taking into account the movement obstacle elements and travel distance for each route and guidance is provided to occupants. do.

That is, in FIG. 7, the information required to calculate the shortest evacuation route for occupants located in a specific space (SP ₁ ) is the distance to the door (R _D1 ) (S ₁ + S ₂ ), the distance of the path (L ₁ , L ₂ ), This will be the number of movement obstacle elements, and in particular, since the weights that impede movement may be different depending on the characteristics of the movement obstacle elements, in the present invention, separate weights are assigned to each type of movement obstacle element located in the internal structure of the building structure for evacuation. It is defined so that it can be used for route calculation.

That is, an example of evacuation route calculation for path 1 (Pt ₁ ) and path 2 (Pt ₂ ) in FIG. 7 using a hydraulic network modeling technique can be defined as follows.

- Evacuation route calculation formula for route 1 (Pt ₁ )

Path 1 (Pt ₁ )=

= S ₁ + S ₂ + R _D1 + direction turn + L ₁ + R _D2 + Direction turn + R _St1 + Direction turn + … R _Stn

- Evacuation route calculation formula for route 2 (Pt ₂ )

Path 2(Pt ₂ )=

= S ₁ + S ₂ + R _{D 1} + L ₂ + Direction turn+ L ₃ + R _E1

It can be expressed as

Here, R is a movement disorder element based on the shape pattern of the internal structure of the building structure, M is the number of occupants, S ₁ , S ₂ are the distance from the inside of the space to the door, and L ₁ , ~ L ₃ are the distance on the path. means distance.

In addition, the movement disorder element (R) applied to evacuation route calculation means a state in which a certain weight is added according to each characteristic. As an example,

For doors,

For stairs,

For entrances,

For windows,

으로 표현할 수 있으며, 여기서

는 가중상수이다.In the case of internal facilities,

It can be expressed as

is a weighting constant.

In addition, when acquiring information on the internal structure based on the drawing of the first building structure, pattern recognition of the drawing and setting of movement disorder elements are accomplished through the definitions shown in Table 1 below.

drawing pattern identification symbol Justice Physical quantities affecting evacuation route calculation wall R _wall resistance to wall Unable to move door R _door Stopping at entrances and exits such as doors Size of door, time required to open corridor R _corridor Resistance to straight passages such as hallways and rooftops straight line distance stairs R _stair resistance to steps Number of stairs, height of stairs window R _window resistance to windows mobility Facilities R _facility Resistance to facilities installed in passageways that cause inconvenience to occupants walking Width and distance of path that can be moved to avoid installed facilities gas leak R _gas resistance to gas generation Area of gas contaminated area, gas generation time fire R _fire Resistance to fire fire area smoke R _smoke Resistance to smoke generation smoke generation area Water supply issues such as water supply and sewage pipes R _water Resistance to water faucet problems Area of accident area in power supply unit turn R _turn Resistance to changing direction outside of a straight path Time and radius of rotation elevator R _elevator Resistance to lift or descender Availability

In addition, when the various mobility obstacle elements described above are applied to the hydraulic network model to calculate an evacuation route, the criteria for assigning weights to each mobility obstacle element are shown in Table 2 below.

Movement resistance element identification symbol Weight application example
(Weight × Resistance Element)

) 범위weight(

) range wall R _wall

0 door R _door

1.2~1.4 corridor R _corridor

1.0 stairs R _stair

2.5~4.0 window R _window

2.2~2.7 Facilities R _facility

3.1~4.0 gas leak R _gas

5.0 or higher fire R _fire

10.0 or higher smoke R _smoke

10.0 or higher Water supply issues such as water supply and sewage pipes R _water

4.0 or higher turn R _turn

1.3~1.8 elevator R _elevator

3.0 or higher

The weights assigned to each movement disorder element are calculated by taking into account the reaction time of the general public to the internal structure of the building structure. These weights can be subdivided more precisely using artificial intelligence deep learning techniques and assigned according to shape or structure. It can be configured.

Meanwhile, in the case of space (SP ₂ ), space (SP ₃ ), and space (SP ₄ ) in FIG. 7, there are path 1 (Pt ₁ ) and path 2 (Pt ₂ ) through which occupants can evacuate, but each It is normal for different evacuation routes to be selected in space.

The reason is that when occupants located in each space evacuate using path 1 or path 2, the movement resistance element that determines the evacuation time may be different.

In other words, it is a movement resistance element that must be passed through to get from each space to the entrance. Since the movement resistance is different depending on the distance to the door, the number of doors, the number of turns, stairs, and the distance of straight sections, etc., in the process of calculating the evacuation route, This is because the shortest evacuation route for occupants is calculated by assigning a certain weight to the above-mentioned movement resistance elements.

Therefore, in the present invention, each of these movement resistance elements is configured to construct a hydraulic network model for each floor, each building, and each space on each floor, so that the evacuation route for each location of the occupants is guided along the path with the smallest movement resistance element. Each occupant can safely evacuate using the evacuation route that provides the shortest time.

As described above, according to the optimal evacuation route guidance system and guidance method within a building structure according to an embodiment of the present invention, when the information on the internal structure of the initially input building structure is completed, the set hydraulic network model technique is applied. The shortest evacuation route for each floor and each space is calculated through an intelligent algorithm, and when a specific event occurs, an evacuation route is calculated that reflects the occurrence of the specific event in the existing information, thereby reducing the load on the evacuation route information department and providing quick and easy access to the evacuation route. There is an advantage that can be provided to occupants through evacuation route calculation.

Although the preferred embodiments of the present invention have been described above, various modifications can be made within the scope of the technical idea of the present invention by those skilled in the art.

100: Evacuation route guidance system
110: Building structure internal structure information acquisition department
120: Occupant information acquisition department
130: Disaster monitoring department 131: Indoor monitoring department
132: Outdoor surveillance department 140: Communication department
150: Evacuation route information provision unit 151: Evacuation route reception unit
152: Evacuation route request unit
160: Evacuation route information unit 161: Data storage unit
162: data classification unit 163: learning unit
164: Analysis unit 165: Evacuation route prediction/judgment unit
166: Evacuation guidance control unit 167: Monitoring unit

Claims (6)

An internal structure information acquisition unit of the building structure that acquires the internal structure information of the building structure based on the two-dimensional drawing information of the building structure;
an occupant information acquisition unit that detects or receives location information of occupants scattered within the building structure;
A disaster occurrence monitoring unit installed inside and outside the building structure to monitor the occurrence of a disaster;
A communication unit that transmits/receives the internal structure information of the building structure obtained from the internal structure information acquisition unit of the building structure, the occupant information obtained from the occupant information acquisition unit, and the disaster occurrence signal obtained from the disaster occurrence monitoring unit;
an evacuation route information provider configured to be installed or portable at multiple locations indoors so that the information received through the communication unit can be used to provide evacuation route information to occupants or enable occupants to request evacuation route information in the event of a disaster;
The internal evacuation route of the building structure is identified based on the information obtained from the internal structure information acquisition department and the occupant information acquisition department of the building structure, and the evacuation route information provision department is provided based on the disaster occurrence signal obtained through the disaster occurrence monitoring department. Includes an evacuation route guide that provides the optimal evacuation route,
The evacuation route information section,
Receive, store and classify building structure information and occupant information obtained from the internal structure information acquisition department and occupant information acquisition department of the building structure,
Learning the stored and classified data,
Based on the above data learning, the movement and internal evacuation routes of the occupants are identified and analyzed for each floor where the occupants are located.
Based on the data learning and structural analysis results of occupant locations and internal evacuation routes, the optimal evacuation route for each occupant location on each floor is predicted and determined in advance.
After obtaining a disaster signal from the disaster occurrence monitoring unit, the evacuation route information provision unit provides optimal evacuation route information for each occupant's location based on the prediction and judgment results when the occupant requests evacuation route guidance or automatically.
When predicting and judging the evacuation route in advance, predict and judge the optimal evacuation route based on the distribution status of each floor and location of occupants, the movement line of occupants, and movement obstacle elements located in the movement line of occupants, depending on the disaster situation. ,
When acquiring the movement disability elements, movement disability weights are differentially applied to each movement disability element according to the characteristics of artificial facilities placed on the movement path, classified and stored,
When predicting and determining the evacuation route in advance, it is configured to predict and determine the evacuation route using movement obstacle element data with different weights obtained after obtaining the movement obstacle element,
After predicting and determining the acquisition and evacuation route of the movement disorder element, a hydraulic network modeling technique is applied to display the efficient flow and distribution of fluid as a planar network based on a number of material points (nodes) and resistance elements. Optimal evacuation route guidance system within building structures based on artificial intelligence. According to paragraph 1,
The evacuation route information section,
Movement disorder elements used when learning the above data include the number of occupants, passage width, location of occupants, or data on walls, artificial facilities, doors, stairs, and distance to the entrance in the passage from the movement line to the entrance. An optimal evacuation route guidance system within an artificial intelligence-based building structure. According to paragraph 1,
The evacuation route information section,
When learning the data, if new equipment or artificial facilities are placed in the main passageway inside the building structure, the new equipment or artificial facilities are extracted as movement disability elements, and the extracted movement disability elements are converted into movement disability elements set in the existing learning model. An optimal evacuation route guidance system within an artificial intelligence-based building structure, which is characterized by predicting and determining the optimal evacuation route for each occupant's location by adding to the configuration of elements and re-learning the new added movement disability elements. In a method of guiding evacuation routes in the event of a disaster within a building structure based on artificial intelligence,
Obtaining information on the internal structure of the building structure based on the two-dimensional drawing information of the building structure;
Obtaining data on movement disorder elements arranged on a movement path among the data obtained in the internal structure information acquisition step of the building structure;
Occupant indoor location information acquisition step of acquiring indoor location information of occupants using the building structure;
Storing and classifying the acquired internal structure information of the building structure, movement disorder element data, and occupant location data;
learning the stored and classified data;
Predicting and determining an evacuation route in advance for each location or movement line of the occupants based on the data learning and the results of the identified movement disorder elements;
A disaster occurrence information acquisition step of acquiring disaster occurrence information through a disaster occurrence monitoring unit installed inside and outside the building structure;
When obtaining the disaster occurrence information, calculating the shortest evacuation route for each occupant based on the prediction and judgment results of the evacuation route for each occupant's location or movement line;
When obtaining the disaster occurrence information, the calculated evacuation route information is transmitted to the evacuation route information provider carried or worn by the occupants or placed inside the building structure, so that the occupants evacuate quickly and safely through the evacuation route suggested by the evacuation route information provider. An evacuation route information transmission step that allows;
It is made including,
In the step of acquiring the movement disability element data, movement disability weights are differentially applied to each movement disability element according to the characteristics of artificial facilities placed on the movement path, classified and stored,
In the step of predicting and determining the evacuation route in advance, the evacuation route is predicted and determined using the movement disorder element data with different weights obtained in the movement disorder element data acquisition step,
The movement obstacle element data acquisition step and evacuation route prediction and judgment step are characterized by the application of a hydraulic network modeling technique that displays the efficient flow and distribution of fluid as a two-dimensional network based on a number of material points (nodes) and resistance elements. An artificial intelligence-based optimal evacuation route guidance method within building structures. delete delete