KR20210052209A - artificial intelligence remote for controlling air conditioner - Google Patents

Abstract

The present invention relates to an artificial intelligence control system for controlling an air conditioner, which comprises: an artificial intelligence control unit which includes a resident activity volume intelligence control unit which monitors the activity volume of residents in the indoors and the clothing they are wearing all the time, and provides pleasure to the residents all the time by intelligently controlling the wind speed, wind volume, temperature, humidity, ventilation, etc. from the air conditioner, and a resident clothing volume intelligence control unit; an artificial intelligence robot which drives to everywhere indoors for a preset time in accordance with the setting of a timer, and which includes a multimeasuring unit which measures the wind speed, wind volume, temperature, and humidity of the indoors all the time; and a terminal which exchanges signals with the artificial intelligence control unit and the artificial intelligence robot and which performs the management needed for controlling the air conditioner. The present invention aims to provide an artificial intelligence control system for controlling the air conditioner, which is able to realize a pleasant feeling to the residents indoors all the time.

Description

Artificial intelligence remote for controlling air conditioner

The present invention relates to an artificial intelligence control system for controlling an air conditioner, and more particularly, based on the amount of clothing worn by residents living indoors, that is, the amount of clothing, such as wind speed, air volume, temperature and humidity, and ventilation. It is about an artificial intelligence control system for air conditioner control that can provide comfort to residents at all times by automatically adjusting and controlling the lights.

In general, in controlling the operation of an air conditioner, users often use a portable remote control as a way to control the cooling temperature of the air conditioner or select various operation types such as dehumidification or air cleaning.

The operation of such a portable remote control is performed very frequently, but there are many inconvenient points to use at any time due to the fixed setting. This inconvenient situation may be due to various reasons, which is due to not knowing the appropriate control standard, and also to the reason why users prefer to turn on and off in a simple way rather than a complicated setting of the remote control, and the location of the temperature sensor. It may be due to the difference in temperature from the difference between the user's location and the temperature difference in the space, or it may be due to the difference between the control standard according to the temperature deviation in the space and the perceived temperature, and depending on the state of body temperature, direct wind and indirect air It may be due to the difference in preference for wind, and it may also be due to the difference in body temperature according to the constitution (e.g., children have a temperature of about 0.5 degrees higher than that of adults and an average of about 1 degree). have.

As such, users often use the portable remote control for various reasons described above, and users often feel uncomfortable even when using the portable remote control.

Of course, an artificial intelligence control method using voice recognition technology or situation recognition technology has been introduced as a way to improve the above-described problems caused by air conditioner control using such a portable remote control, but control of wind speed, air volume, temperature and humidity, and ventilation. Due to the inaccurate accuracy, the qualitative comfort of the residents living indoors has not reached a satisfactory level.

The present invention for solving the above-described problems is a method of monitoring the state of the amount of clothing worn, that is, the state of the amount of clothing, as well as the state of the activity amount of residents living indoors, in a real-time monitoring method of wind speed, air volume, temperature and humidity, and ventilation. Its purpose is to provide an artificial intelligence control system for air conditioner control that can intelligently control lights to realize constant comfort to indoor residents.

The present invention for achieving the above objects is an artificial intelligence robot equipped with a multi-sensor unit 100 that continuously measures the wind speed and air volume, temperature and humidity in the room while driving throughout the room for a set time according to the timer setting. (10) Active volume intelligence that provides a sense of comfort to residents at all times by intelligently controlling wind speed, air volume, temperature and humidity, and ventilation from the air conditioner through constant monitoring of the activity volume and clothing worn by residents living indoors. An artificial intelligence control unit 1000 composed of a control unit 1100 and a wear amount intelligent control unit 1200, and a terminal 2000 that performs management necessary for controlling the air conditioner in a manner that communicates with the artificial intelligence control unit and the artificial intelligence robot. There is an example feature of the artificial intelligence control system for air conditioner control.

The activity amount intelligent control unit 1100 includes a camera 1110 that monitors the amount of activity and the amount of clothing of the resident in a manner of constantly recording an image, and an activity amount of the resident based on a video object recorded from the camera ( An air conditioner further comprising an artificial intelligence first multi-controller 1120 that intelligently controls wind speed, air volume, temperature and humidity, and ventilation through the current measurement information measured by the multi-sensor unit 100 based on the flow rate. An example feature is the control artificial intelligence control system.

The amount of clothing intelligent control unit 1200 is a camera 1210 that monitors the amount of clothing and the amount of clothing of the occupant in a manner that always records a video, and the resident's based on the video recorded from the camera. A second multi-controller 1220 of artificial intelligence that intelligently controls wind speed, air volume, temperature and humidity, and ventilation through the current measurement information measured by the multi-sensor unit 100 based on the amount of clothing is further provided. There is an example feature of the artificial intelligence control system for air conditioning control including.

The first multi-controller 1120 of the artificial intelligence is a main body 1121 storing a deep learning DB obtained by learning enormous amount of activity information of a resident, and a measurement provided by the multi-sensor unit 100 installed in the main body. An artificial air conditioner control system that further includes an activity control algorithm 1122 that intelligently controls wind speed, air volume, temperature and humidity, and ventilation in accordance with the current occupant's activity condition by analyzing information in a way that compares the information with the deep learning DB. The intelligent control system has an example feature.

The second multi-controller 1220 of the artificial intelligence is a main body 1221 storing a deep learning DB obtained by learning a vast amount of information on a resident's wearing amount, and installed in the main body and provided from the multi-sensor unit 100. An air conditioner further comprising a clothing control algorithm 1222 that intelligently controls wind speed, air volume, temperature and humidity, and ventilation in accordance with the conditions of the current occupant's wearing amount by analyzing measurement information in a manner that compares it with the deep learning DB. An example feature is the control artificial intelligence control system.

The deep learning DB is extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the activity amount extracted by the extraction program Based on the deep learning DB of information, the activity control algorithm 1122 is an example of an artificial intelligence control system for air conditioner control that intelligently controls wind speed, air volume, temperature and humidity, and ventilation in accordance with the current occupant's activity condition. There are features.

The deep learning DB is extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the amount of clothing extracted by the extraction program. ) Based on the deep learning DB of information, the dressing control algorithm 1222 is applied to an artificial intelligence control system for air conditioning that intelligently controls wind speed, air volume, temperature and humidity, and ventilation according to the current occupant's dressing condition. There is an example feature.

The activity control algorithm 1122 detects an object according to acquisition of image and image information collected from the cameras 1110 and 1210, and then detects the object according to the acquisition of image information collected from the cameras 1110 and 1210, and then, in the activity amount information, a joint through a deep neural network (DNN). Key points are extracted, and clothing is extracted in segment units through a convolutional neural network (CNN) from the amount of clothing, and is classified into pose data and clothing data, respectively, and the amount of activity that has already been learned and stored. Based on the deep learning DB of (活動量) information, the above pose data is compared and analyzed to calculate the amount of activity of the resident, and the comfort that meets the conditions of the current resident's activity amount (Predicted Mean Vote; There is an example feature of the artificial intelligence control system for air conditioner control that intelligently controls wind speed, air volume, temperature and humidity, and ventilation according to the numerical values of PMV) to provide comfortable warmth to residents.

The wear control algorithm 1222 detects an object according to the acquisition of image and image information collected from the cameras 1110 and 1210, and then detects the object according to the acquisition of image information collected from the cameras 1110 and 1210, and then, in the activity amount information, a joint through a deep neural network (DNN). Key points are extracted, and clothing is extracted in segment units through a convolutional neural network (CNN) from the information on the amount of clothing, classified into pose data and clothing data, respectively, and then learned and stored clothing. Based on the deep learning DB of the amount information, the above clothing data is compared and analyzed, and the amount of clothing of the resident is calculated. Mean Vote (PMV) is an artificial intelligence control system for air conditioner control that intelligently controls wind speed, air volume, temperature and humidity, and ventilation to provide comfortable heat to residents.

The multi-sensor unit 100 is a kind of multi-sensor device, and measures a voice recognition processing unit 101 that processes a user's voice, a distance to a user (resident) existing in the indoor measurement area, and a temperature with the user (resident). It further includes a scan-type sensor 102 that is reflected in the thermal image, and probes 103 and differential pressure sensors 104 that measure the wind speed and direction of the room, and temperature and humidity, and the probes 1100 are for wind speed. An example feature is an artificial intelligence control system for controlling an air conditioner comprising at least a plurality of probes, temperature and humidity probes, and turbulence probes.

The scan-type sensor 102 is a sensor that measures a distance to an object existing in a measurement area and a temperature with the object and reflects it in a thermal image. It is located at the front of the mirror M1 and receives light reflected from the mirror. A distance measurement sensor 110 that measures the distance to an object existing in the measurement area, and measures the temperature of the object existing in the measurement area by receiving light reflected from the mirror located at the rear of the mirror M1. The temperature measurement sensor 120, a rotation driving unit 130 having a plurality of motors forming two rotation shafts and securing a measurement area of the distance measurement sensor and the temperature measurement sensor by rotating 360 degrees left and right and 180 degrees vertically. , And a synthesis processing unit 140 for synthesizing distance information and heat distribution information respectively measured by the distance measuring sensor and the temperature measuring sensor and reflecting the distance information in the thermal image. There are features.

The artificial intelligence robot 10 is installed in such a way that the body part 20 constituting the body, the head part 30 connected in a rotatable manner from the upper part of the body, and a power means in the lower part of the body part are operated. An example feature is the artificial intelligence control system for controlling an air conditioner including the leg portion 40 to be moved.

The leg portion 40 is composed of a plurality of claw legs, the claws of a plurality of pulleys connected to a belt connected to a gear coupled to a single power motor shaft and gears individually connected to the pulleys. An example feature is the artificial intelligence control system for air conditioner control operated by a combination of connections.

The head portion 30 of the artificial intelligence robot 10 is rotated in a horizontal manner by connecting a spur gear formed at the lower end of the head portion to a gear connected to one power motor shaft in a horizontal direction, while the spur gear It has a structure in which a locking pin is accommodated in a long hole inside the shaft that extends vertically from the lower center of the shaft, and at the same time, a half gear formed at the lower end of the shaft and a spring is formed between the shaft. An example feature is the artificial intelligence control system for controlling an air conditioner that is operated in a nod operation method through a power motor that provides the power required for the rotation of the gear.

The head unit 30 of the artificial intelligence robot 10 is measurement information provided from an object, based on the distance information value and the temperature information value sensed from the distance measurement sensor 110 and the temperature measurement sensor 120. An artificial for controlling an air conditioner operated in an inverted horizontal manner in the forward direction of the body part 20, but operated in a nod operation method in the front and rear direction when only the distance information value detected by the distance measurement sensor 110 is detected. The intelligent control system has an example feature.

The distance measuring sensor 110 is located at the front of the mirror and receives the reflected light through a lens after entering the front of the mirror according to the rotation of the rotation driving unit 130 to measure and output distance information. There is an example feature of the artificial intelligence control system for air conditioner control including 111.

The temperature measurement sensor 120 is located in an empty space on the rear side of the mirror and receives light incident on the rear surface of the mirror according to the rotation of the rotation driving unit 130, converts it into a thermal image signal, and outputs the bolometer 121 There is a feature of an example of the artificial intelligence control system for air conditioning control comprising a.

According to the present invention as described above, the wind speed, air volume, temperature and humidity, and ventilation of the air conditioner are intelligently and automatically controlled based on the amount of activity or wearing of the residents living indoors, and the quality of the indoor residents There is an effect that can improve comfort.

Moreover, according to the present invention, the difference between the practical comfort index satisfied by indoor residents and the comfort index implemented by intelligent control of the air conditioner can be corrected with a correction algorithm. There is an effect that can reduce errors.

In addition, according to the present invention, since the artificial intelligence robot traveling indoors can detect the difference in the distribution of wind speed and air volume and the distribution of temperature and humidity throughout the room, the amount of activity or the amount of clothing ( There is an effect of improving the control accuracy of the air conditioner's wind speed, air volume, temperature and humidity, and ventilation based on the amount of wear.

In addition, according to the present invention, in the case of an artificial intelligence robot that responds to the user's voice, it can be used not only as an air conditioner control remote control, but also as a toy that maintains intimacy with the user and enjoys it for play There is an effect that can also provide the joy of play to the child.

1 is a block diagram conceptually showing a basic configuration of an artificial intelligence control system for controlling an air conditioner according to an embodiment of the present invention.
2 is a configuration of the multi-sensor unit 100 installed in the artificial intelligence robot 10 shown in FIG. 1, and the amount of activity of the artificial intelligence control unit 1000, the intelligent control unit 1100, and the amount of clothing It is a diagram showing the configuration of the intelligent control unit 1200 as a block.
3 is a conceptual diagram illustrating a state in which heating and wind direction of an air conditioner indoor unit are controlled based on a user (resident) living indoors through the interaction between the artificial intelligence robot 10 and the artificial intelligence control unit 1000 and the terminal 2000. It is a drawing shown as.
FIG. 4 is a diagram showing as an example the concept of the amount of activity of a user (resident) required for collection of a deep learning DB learned by the artificial intelligence unit.
FIG. 5 is a diagram showing as an example the concept of a user's wearing amount required for collection of a deep learning DB learned by an artificial intelligence unit.
6 is a diagram illustrating a concept of a collection process of a deep learning DB learned by an artificial intelligence unit as an example.
7 is a diagram illustrating a concept of a process of learning a user's activity amount by a deep learning DB in an artificial intelligence unit as an example.
8 is a diagram illustrating a concept in which the amount of activity of a user is presented as an objectized pose in the artificial intelligence unit as an example.
FIG. 9 is a flow diagram showing a control flow of an activity control algorithm 1122 for calculating a predicted mean vote (PMV) index as a feeling of indoor comfort according to a state condition of a user (resident)'s indoor activity It's a chart.
FIG. 10 shows a control flow of a wear control algorithm 1222 for calculating a predicted mean vote (PMV) index as a comfortable feeling in the room according to the condition of the indoor wear amount of a user (resident). This is a flowchart.
FIG. 11 is a perspective view showing a concrete conceptual shape structure of the artificial intelligence robot 10 shown in FIG. 1.
12 is a block diagram conceptually showing the basic configuration of the scan-type sensor 102 shown in FIG. 2.
13 is a diagram conceptually showing a basic configuration of a microbolometer.
14 is a diagram schematically showing a mechanical mechanism of the scan-type sensor 102 shown in FIG. 12.
15 is a diagram illustrating a state in which light paths are separated according to the incident of the mirror in FIG. 12.
16 is a view showing the rotation drive unit 130 in FIG. 12.
FIG. 17 is a diagram illustrating a synthesis process performed by the synthesis processing unit 140 in FIG. 12.

The present invention should be interpreted in a manner that includes the scope of the rights to the technical idea through various modified embodiments, only these embodiments make the posting of the present invention complete, and common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to the possessor, and the invention is only defined by the scope of the claims.

Moreover, the drawings attached to the present invention are attached as a way to help the understanding of the description of the present invention to the last, and the technical idea of the present invention should not be interpreted as being limited by the attached drawings.

The artificial intelligence system for controlling an air conditioner according to an embodiment of the present invention is an air conditioner having a panel structure installed indoors in a manner that monitors the state of activity and amount of clothing of residents living indoors. It is a technology that can provide a feeling of comfort to indoor residents at all times by intelligently controlling the device.

The state of the amount of activity of the residents can be defined as encompassing motions of, for example, sleeping motion, sitting motion, standing motion, and various other activities, while the state of the occupant's wearing amount is For example, it may be defined as a whole range of tops and bottoms worn on the body, tops and bottoms for summer, tops and bottoms for winter, types of tops and bottoms, and various other clothes.

Moreover, the term'the amount of activity of residents' can be interpreted to include terms such as posture, attitude, and pose of residents living indoors, and posture, attitude, and pose can be used interchangeably with the same meaning. On the other hand, the term'the amount of clothing of residents' can be interpreted as including terms such as clothes, clothes, clothing, and clothing worn on the bodies of residents living indoors. , Clothing, and clothing may also be used interchangeably with the same meaning.

In addition, people residing indoors refer to users or residents, and terms of users or residents may have the same meaning or may be used interchangeably.

The artificial intelligence system for controlling an air conditioner according to an embodiment of the present invention may be composed of an artificial intelligence robot 10, an artificial intelligence control unit 1000, and a terminal 2000, as shown in FIG. 1, and the artificial intelligence robot ( 10) is a configuration including a body part 20, a head part 30, and a leg part 40, and may be mounted in a form in which the multi-sensor part 100 is installed, and the artificial intelligence control part 1000 is It may consist of a configuration including an activity amount intelligent control unit 1100 and a wear amount intelligent control unit 1200, and the terminal 2000 may be a terminal that can be fixedly installed indoors, or may be a portable terminal, for example, wireless communication. It may be a smart device capable of this, and may be a smart phone as an example.

The artificial intelligence robot 10 is intended to constantly measure the wind speed, air volume, temperature, humidity, etc. of indoor air conditioners installed indoors while driving around the room through the setting of a timer built into itself, while also recognizing the voice of the user (resident). It can be used as a remote control that responds in a way that does.

Air conditioner indoor units can be interpreted as including air conditioners, heaters, or air conditioners, and in particular, panels disclosed in Application No. 10-2019-0105347 (name: radiant convection panel structure air conditioner) filed by the applicant of the present invention. The air conditioner of the structure can be referred to.

The artificial intelligence robot 10 may be configured to include a multi-sensor unit 100 installed therein to recognize the voice of a user (resident), and the multi-sensor unit 100 Over air volume, temperature and humidity can be measured at all times.

The artificial intelligence robot 10 includes a body part 20 forming a body as shown in FIGS. 1 and 2, a head part 30 connected in a rotatable manner from the upper part of the body, and a power means at the lower part of the body part. It may be configured to further include a leg portion 40 that is installed and moved in a manner that is operated through, and the multi-sensor portion 100 may be mounted in a manner that is installed on the body portion 20, for example, The measurement information measured by the multi-sensor unit 100 may be transmitted to the artificial intelligence control unit 1000 in a wireless communication method through a beacon.

The artificial intelligence control unit 1000 constantly monitors the amount of activity of the user (resident) residing indoors and the clothing worn by the user (resident), through which wind speed and air volume, temperature and humidity, and ventilation from indoor units are monitored. It may be composed of an activity amount intelligent control unit 1100 and a wear amount intelligent control unit 1200 that provides a feeling of comfort to residents by intelligently controlling the back.

The artificial intelligence robot 10 and the artificial intelligence control unit 1000 may be controlled and managed through, for example, a user's smartphone as the terminal 2000, and in particular, the artificial intelligence robot 10 may recognize the user's voice. It can also be controlled and managed through.

The artificial intelligence control unit 1000 intelligently controls the indoor wind speed and air volume, temperature and humidity, and ventilation based on the measurement information measured from the multi-sensor unit 100. In particular, the indoor user (resident It is characterized in that it intelligently controls the indoor wind speed and air volume, temperature and humidity, and ventilation based on the amount of activity of) and the amount of wear of the user (resident).

Accordingly, the artificial intelligence robot 10 and the artificial intelligence controller 1000 may be controlled and managed through a mobile application installed on the user's smartphone as an example of the terminal 2000 as shown in FIG. 3, and the artificial intelligence controller 1000 ) Is the indoor wind direction and wind speed measured by the artificial intelligence robot () based on the image image information collected from the cameras 1110 and 1120 that monitor the amount of activity and the amount of clothing of the user (resident) residing indoors, As measurement information such as temperature and humidity is controlled and processed based on the deep learning DB in the artificial intelligence unit, the wind speed and air volume, temperature and humidity, and ventilation of indoor units installed indoors are based on the amount of activity and clothing of the user (resident). By intelligent control in a manner that is indexed according to the predicted mean vote (PMV), indoor residents can always be provided with a sense of comfort.

As shown in FIG. 2, the activity amount intelligent control unit 1100 may further include a camera 1110 and a first multi-controller 1120, and the wear amount intelligence control unit 1200 includes a camera 1210 and a A second multi-controller 1220 may be further included. Of course, the cameras 1110 and 1210 may be configured as one integrated camera, for example. These cameras 1110 and 1210 may be a CCTV type capable of real-time monitoring, but need not be limited thereto, and may include all of them as long as they are capable of real-time monitoring.

The first multi-controller 1120 may be configured to further include a body 1121 equipped with an artificial intelligence unit and an activity control algorithm 1122 installed in the artificial intelligence unit, as shown in FIG. 2, and the second multi-controller 1220 may be configured to further include a body 1221 equipped with an artificial intelligence unit and a dressing control algorithm 1222 installed in the artificial intelligence unit. Of course, the main bodies 1121 and 1221 may be operated as one integrated main body, for example.

In particular, the first multi-controller 1120 is based on the measurement information transmitted from the multi-sensor unit 100 through the beacon receiver installed in the main body 1121 ) In order to provide a comfortable feeling that meets the conditions, it intelligently controls the indoor units of the panel structure installed in the room, and can control the indoor wind speed and air volume, temperature and humidity, and ventilation.

In addition, the second multi-controller 1220 is based on the measurement information transmitted from the multi-sensor unit 100 through the beacon receiver installed in the main body 1221, the current wearing amount of the user (resident) residing indoors.量) In order to provide a feeling of comfort that meets the conditions, it intelligently controls the indoor units of the panel structure installed in the room, and can control the indoor wind speed and air volume, temperature and humidity, and ventilation.

That is, the camera 1110 is installed indoors to monitor the current activity amount of a user (resident) residing indoors, and the camera 1210 is installed indoors to live indoors. The current amount of clothing of the user (resident) can be monitored, and the data on the amount of activity and the amount of clothing of the user (resident) collected by monitoring in this way are learned by the artificial intelligence department. Since it can be stored in a deep learning DB format, the deep learning DB can be used as a database already learned by the artificial intelligence unit.

The deep learning DB that has already been learned and stored can be used as a database of the mother, which is a criterion for analysis in the artificial intelligence unit, and based on the analysis result from the artificial intelligence unit, the first multi-controller 1120 and the second The multi-controller 1220 uses the indoor units as an activity control algorithm 1122 and a wear control algorithm as a way to provide an indoor comfort suitable for the current indoor activity amount of the user (resident) and the current indoor wear amount. 1222) intelligently controls and controls indoor wind speed and air volume, temperature and humidity, and ventilation.

The deep learning DB is processed data on the amount of activity of a user (resident) residing indoors, and may be, for example, motions, postures, and attitudes of the user (resident) as shown in FIG. 4, and specifically sleeping motion ( sleeping, reclining, seated quiet, reading seated, writing, typing, walking about ), and all other motions and postures, and the information on the amount of activity of the user (resident) has already been learned by the artificial intelligence unit mounted on the main body 1121 of the artificial intelligence control unit 1000. It can be stored as a vast deep learning DB.

Moreover, the deep learning DB is processed data on the amount of clothing of a user (resident) residing indoors, and may be, for example, clothing worn on the body of the user (resident), as shown in FIG. Following detection according to (dress, bag, shoes) can be processed into segments, such as summer tops and bottoms, autumn tops and bottoms, winter tops and bottoms, types of tops and bottoms, and the fibers of the tops and bottoms. Information on the amount of wear of the user (resident), including types, hats, shoes, stockings, etc., is already learned by the artificial intelligence unit mounted on the main body 1221 of the artificial intelligence control unit 1000. It can be saved as a DB.

That is, looking at the collection process of the deep learning DB that is learned by the artificial intelligence unit, for example, as shown in Fig. 6, the data of activity classification and clothing classification through object recognition of individual deep learning are processed, and these data are again used for multi-learning. It can be collected through data processing of activity classification and clothing classification through object recognition in a meta-learning method.

In other words, the data information of activity classification through object recognition can be extracted by detecting joint key points of each body part according to an image input, for example, as shown in FIG. 7 through a deep neural network (DNN) of the artificial intelligence unit. , The extracted keypoints are encoded as connectivity, and the pose state of the resident can be estimated through the sum of the keypoints. The estimated pose may be learned as a standing pose, a walking pose, and a sitting pose as shown in FIG. 8 and stored in the artificial intelligence unit in a deep learning DB format.

Of course, data information of clothing classification through object recognition can be extracted in segment units following detection according to classification as shown in FIG. 5, for example, through a convolutional neural network (CNN) of the artificial intelligence unit. The segmented segment is encoded as connectivity, and the wearing state of the occupant can be estimated through the sum of the segments. The estimated state of clothing can be learned and stored in the artificial intelligence unit in the form of a deep learning DB.

This deep learning DB can be extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the activity amount extracted by the extraction program Based on the deep learning DB of (活動量) information, the activity control algorithm 1122 intelligently controls the indoor units installed in the room according to the current activity condition of the user (resident), and the indoor wind speed, air volume, temperature and humidity. , And ventilation can be adjusted to meet the predicted mean vote (PMV) index.

Of course, the deep learning DB can be extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the clothing extracted by the extraction program Based on the deep learning DB of the amount information, the dressing control algorithm 1222 intelligently controls the indoor units installed in the room according to the conditions of the user (resident)'s current amount of clothing, and the wind speed and air volume in the room. , Temperature and humidity, and ventilation can be adjusted to meet the predicted mean vote (PMV) index.

Moreover, there may be a difference in the predicted mean vote (PMV) index based on the condition of the amount of activity and the amount of wear, but this difference is determined by the amount of activity (活動量) The indoor unit is controlled in accordance with the Predicted Mean Vote (PMV) index as it is corrected in a way that reduces errors caused by the matching of the deep learning DB containing information and the deep learning DB containing the amount of wear. Can be done accurately.

The activity control algorithm 1122 and the wear control algorithm 1222 above are based on the current activity amount and the amount of clothing of the user (resident) residing indoors. Predicted Mean Vote (PMV) ) How to intelligently control the wind speed, air volume, and temperature and humidity of indoor units having a panel structure according to the numerical values of the index will be described later in more detail.

On the other hand, the leg portion 40 of the artificial intelligence robot 10 described above may be composed of, for example, a plurality of claws, the claws are connected to a belt connected to a gear coupled to a single power motor shaft. It may be operated by a connection combination of the pulleys of the pulleys and gears integrally formed on the upper ends of the clamp legs that are individually connected to the pulleys.

The gear coupled to the power motor shaft for the operation of the forefingers, the belt, the pulleys, and the connection configuration for the gears integrated at the upper ends of the forefingers are not specifically shown in the drawings, but the above It can be fully understood through the connection description.

Meanwhile, the multi-sensor unit 100 installed in the body part 20 of the artificial intelligence robot 10 has a voice recognition processing unit 101 that processes the user's voice, and the distance to an object (for example, a user) existing in the measurement area. And a scan-type sensor 102 that measures a temperature with an object (for example, a user) and reflects it in a thermal image, and probes 103 and differential pressure sensors 104 capable of measuring wind speed and direction in the room, and temperature and humidity. It may be made of a configuration that further includes.

The voice recognition processing unit 101 processes the noise of the user's voice recognition section, and then stores the voice model DBs in a manner of extracting the features of the voice recognition section, and stores the user's voice features and the above voice model DBs. In the course of analysis, as the voice model DB that can match the voice characteristics of the user is transmitted to the micro controller unit (MCU) installed in the body 20, the MCU reads the information on the voice model DB and The intelligent robot 10 is instructed to act.

The scan-type sensor 102 is a distance measuring sensor located at the front of the mirror M1 and receiving light reflected from the mirror M1 to measure the distance to a user (resident: object) existing in the indoor measurement area. (110), a temperature measurement sensor 120 that is located at the rear of the mirror M1 and receives light reflected from the mirror M1 to measure the temperature of a user (resident: object) existing in the measurement area of the room , A rotation drive unit 130 for securing a measurement area of the distance measurement sensor and the temperature measurement sensor through 360° left and right rotation and 180° vertical rotation, and a plurality of motors forming two rotation axes, and the distance measurement sensor And a synthesis processing unit 140 for synthesizing distance information and heat distribution information respectively measured by the temperature measuring sensor and reflecting the distance information on the thermal image.

The probes 103 are probes that measure wind speed, air volume, temperature, and humidity in various places in the room, and consist of a wind speed probe that measures wind speed and air volume, a temperature and humidity probe that measures temperature and humidity, and a turbulence probe. It can, but need not be limited to this, the differential pressure sensor 104 can determine the difference between the maximum value and the minimum value of the indoor air pressure.

As such, the multi-sensor unit 100 processes the user's voice through the voice recognition processing unit 101 and measures the distance to the indoor resident through the scan-type sensor 102 and the temperature of the resident through the probe 103. It is possible to comprehensively perform indoor wind speed direction and temperature and humidity measurement and indoor air pressure measurement through the differential pressure sensor 104, and the comprehensive measurement information measured by the multi-sensor unit 100 as described above is the artificial intelligence control unit 1000 described above. As it is transmitted to, the activity amount intelligence control unit 1100 and the clothing amount intelligence described above are provided to provide a sense of comfort in the room suitable for the state of the activity amount and the amount of clothing of the user (resident) residing indoors. The controller 1200 intelligently controls the air conditioner and can adjust the indoor wind speed, air volume, temperature, and humidity.

On the other hand, the artificial intelligence robot 10 is configured in a structure in which the head part 30 is rotatable at the upper end of the body part 20 as shown in FIG. 11, and the leg at the lower end thereof with respect to the body part 20 It may be configured in a structure in which the unit 40 is operated. In particular, the head part 30 rotates in a horizontal manner by connecting a spur gear formed at the lower end of the head part to a gear connected to one power motor shaft in a horizontal direction, while vertically moving from the lower center of the spur gear to the lower part. It is a structure in which a locking pin is accommodated in a long hole inside the extended shaft, and at the same time, a half gear formed at the lower end of the shaft and a structure in which a spring is formed between the shaft, and the power required for rotation of the half gear at the bottom of the half gear It can be operated in a nodding mode through the provided power motor. The connection structures for the power motor shaft, gear, spur gear, long hole, locking pin, half gear, shaft, spring, power motor, etc. described above can be fully understood by explanation, and their connection structures and symbols are not shown in the drawings. It is revealed that it is.

The head unit 30 of the artificial intelligence robot 10 is the measurement information provided from an object (for example, a user), and the distance information value and temperature information value sensed from the distance measurement sensor 110 and the temperature measurement sensor 120 As a basis, the body part 20 can be rotated in an upside down horizontal manner, but when only the distance information value detected by the distance measuring sensor 110 is detected, it operates in an operation method of nodding in the front and rear directions. It could be.

The distance measuring sensor 110 is located at the front of the mirror and receives the reflected light through a lens after entering the front of the mirror according to the rotation of the rotation driving unit 130 to measure and output distance information. (111) may be included.

The temperature measurement sensor 120 is located in an empty space on the rear side of the mirror and receives light incident on the rear surface of the mirror according to the rotation of the rotation driving unit 130, converts it into a thermal image signal, and outputs the bolometer 121 It may include.

As such, the scan-type sensor 102 may be configured to include a distance measurement sensor 110, a temperature measurement sensor 120, a rotation drive unit 130, and a synthesis processing unit 140, as shown in FIG. 12, and the distance measurement sensor Reference numeral 110 may measure a distance to a user (resident: object) existing in the indoor measurement area. In general, an ultrasonic sensor, an infrared ray sensor, a lidar sensor, a radar sensor, a camera sensor using visible light, and the like are mainly used for the distance measurement sensor 110. Here, the distance measurement sensor 110 may be implemented as a lidar sensor. The radar fires a laser pulse and receives the light reflected back from the surrounding object and can measure the distance to the object, etc.

The temperature measurement sensor 120 may measure temperature by detecting ultraviolet rays or mainly infrared rays emitted from a user (resident: object) existing in the indoor measurement area. Here, the temperature measurement sensor 120 includes a bolometer 121 as a thermal image sensor. The bolometer 121 is made of a metal or semiconductor material, reacts to incident infrared rays, and has a much lower reaction speed and sensitivity than a photon detector.

The bolometer 121 has a lens (not shown) or a window that receives infrared radiation emitted by an object in a direction as known in the art, and a plurality of microbolometers are usually arranged to form a matrix. Have. All objects emit infrared rays in a pattern according to their own temperature, and focusing on the fact that the higher the temperature of the object, the greater the amount of infrared radiation emitted.

Looking at the basic configuration of the microbolometer, a substantially rectangular structure (sensing body: 220) capable of sensing infrared radiation energy on a signal processing circuit (ROIC) substrate 210 is provided with two supports 230a and 230b on both sides thereof. ) To have a shape that is supported in a state spaced apart from the substrate 210. The sensor 220 absorbs infrared radiation energy well and can easily increase the temperature with a small amount of energy, and uses a material including a semiconductor material such as bannadium oxide that reacts sensitively to a slight increase in temperature. .

The supporters 230a and 230b contain metal electrodes as shown in FIG. 13, one of which is on the sensing body 220 and the Y metal layer 240, and the other is the sensing body 220 and circuit elements (transistors) on the ROIC substrate 210. Connected to the circuit element). The supports 230a and 230b coupled to the sensing body 220 to prevent the heat of the sensing body 220 from being easily transferred to the surroundings are made of low thermal conductivity, and the space between the substrate 210 and the sensing body 220 is It is in a vacuum. When the resistance changes due to the temperature change of the sensor, the output voltage applied to the circuit element increases accordingly, and current flows through the circuit element. This output voltage or current becomes information related to the temperature of the object corresponding to this microbolometer.

However, since the infrared radiation energy sensed by the sensing body 220 of this microbolometer is not the total radiant energy emitted by the object, it can be changed according to the distance between the object and the temperature measuring sensor 120, and the atmospheric condition. Processes to obtain the actual temperature of the object through a program or the like. The lens or window of the bolometer 121 may be made of a material that does not absorb infrared rays emitted by the target radiation and passes through them well, for example, a window of an infrared passing material containing a lot of germanium, and has its own heat absorption capacity. A Fresnel lens having a small volume can be used instead of a convex lens to reduce the reduction. On the surface, a filter capable of detecting a specific (temperature) band frequency may be installed.

In a typical bolometer, when the entire target area to be sensed is viewed as one image, the temperature of the target area obtained through a micro bolometer in charge of individual pixels (pixels) constituting the image is displayed in different colors on the display device. In one embodiment, the bolometer 121 may increase the target area through rotation of the rotation driving unit 130 without having to slow down a large number of microbolometers forming a matrix in order to widen the target area. In that sense, in one embodiment, the bolometer 121 may form the temperature measuring sensor 120 in the form of a thermal imager having a low resolution (small number of pixels).

The rotation drive unit 130 makes it possible to secure a measurable area of the distance measurement sensor 110 and the temperature measurement sensor 120. In one embodiment, the rotation drive unit 130 includes a plurality of motors 131 forming two rotation axes. One of the plurality of motors 131 rotates in a left and right 360 degree direction, and the other rotates in a vertical 180 degree direction. The synthesis processing unit 140 synthesizes the distance information and heat distribution information measured by the distance measurement sensor 110 and the temperature measurement sensor 120, respectively, so that the distance information can be reflected in the thermal image. In one embodiment, the synthesis processing unit 140 constitutes two layers of a distance information layer and a heat distribution layer by imaging distance information and heat distribution information measured by the distance measurement sensor 110 and the temperature measurement sensor 120, respectively, and It can be synthesized.

The scan-type sensor 102 may arrange a distance measurement sensor 110 and a temperature measurement sensor 120 on each of the front and rear surfaces of the mirror 310 based on the mirror 310. In one embodiment, the temperature measurement sensor 120 may be positioned on an empty space behind the mirror 310. Here, the mirror 310 is positioned at an angle of 45 degrees so that incident light can be reflected. At this time, the mirror 310 is rotated 360 degrees left and right and 180 degrees up and down by the rotation driving unit 130, so that the angle is changed so that the light is incident on the front or rear surface of the mirror 310, and the reflection direction is different depending on the incident surface. You lose. That is, a path of incident light may be separated.

As shown in FIG. 15, the lens 320 may be positioned on the front side of the mirror 310 and receive light reflected by the mirror 310 to focus on the detector 111 of the distance measuring sensor 110. The detector 111 of the distance measuring sensor 110 may measure the distance by detecting the reflected light by emitting light from an emitter (not shown). The temperature measurement sensor 120 is located in an empty space on the rear side of the mirror 310 and when the angle of the mirror 310 is changed according to the rotation of the rotation driving unit 130, the light incident on the rear side of the mirror 310 and reflected light is received. You can measure the temperature. In one embodiment, the temperature measurement sensor 120 measures the temperature of the object by analyzing a wavelength radiated to the object. In this case, wavelength analysis and temperature measurement may be performed through the bolometer 121.

The rotation drive unit 130 includes a plurality of motors 131 and a rotating body 133 rotated by the plurality of motors 131 as shown in FIG. 14. In one embodiment, the plurality of motors 131 are a first motor 131a that drives the rotation body 133 to rotate left and right 360 degrees, and a second motor 131b that drives the rotation body 133 to rotate up and down 180 degrees. It may include. Here, as the rotating body 133 rotates according to the driving of the plurality of motors 131, the light can be divided into both sensors 110 and 120 by using both sides of the mirror 210, so that distance and temperature measurements can be performed. Make it all possible.

Light incident on one surface of the mirror 310 according to the rotational driving of the rotation driving unit 130 is reflected to the detector 111 of the distance measuring sensor 110 through the lens 320. Here, one surface of the mirror 310 may be referred to as a front surface for convenience of description, and the other surface may be referred to as a rear surface of the mirror 310. The distance measurement sensor 110 may be positioned at the rear side of the lens 320 in the optical axis direction to measure and output distance information from light that passes through the lens 320. When light is incident on the rear surface of the mirror 310 according to the rotational driving of the rotation driving unit 130, it is reflected toward the temperature measurement sensor 120.

In an embodiment, the temperature measurement sensor 120 may convert an optical image of far-infrared rays reflected from the mirror 310 through the bolometer 121 into a thermal image signal and output it. Here, the temperature measurement sensor 120 may include a CCD sensor unit that converts an optical image of visible light reflected from the mirror 310 into a real image signal and outputs the converted image signal. In this case, the temperature measurement sensor 120 may be mounted with a CCD sensor unit on a single PCB substrate.

The rotation drive unit 130 may include a plurality of motors 131. The plurality of motors 131 may include a first motor 131a and a second motor 131b as shown in FIG. 16. The first motor 131a is positioned on the first axis (Y-axis) to rotate the scan-type sensor 102 horizontally 360 degrees according to an exemplary embodiment. The second motor 131b is positioned on the second axis (X-axis) to rotate the scan-type sensor 102 vertically 180 degrees according to an exemplary embodiment. Here, the first motor 131a and the second motor 131b can transmit power in a two-step manner to the side without centering the shaft, thereby forming an empty space in the central portion A. The distance measuring sensor 110 and the temperature measuring sensor are used as a wiring space in the center (A) of the first motor 131a and the second motor 131b as a wiring space into which cables can enter or an optical axis through which light can pass. It is also possible to efficiently arrange 120 in a limited space. When the empty space of the central part (A) of the first motor 131a and the second motor 131b is used as a wiring space, the power and signal cables may be wired so as not to get tangled.

The distance-temperature scan-type sensor 102 according to an embodiment rotates 360 degrees left and right and 180 degrees up and down according to the driving of the first motor 131a and the second motor 131b to reduce the measurement area in a two-axis form. Distance and temperature can be measured by scanning method. Here, the distance information, thermal image information, and real image information measured by the distance measuring sensor 110 and the temperature measuring sensor 120, respectively, may be synthesized and processed by the synthesis processing unit 140 to be used for controlling the air conditioner of the air conditioner.

The synthesis processing unit 140 calculates coordinate information from the distance information measured by the distance measurement sensor 110 and shifts a plurality of low-resolution real image information measured by the temperature measurement sensor 120 based on the calculated coordinate information. As shown in FIG. 17, by obtaining and synthesizing high-frequency components that are lost in the low-resolution image, it is possible to implement a clear, high-resolution image with high resolution.

The scan-type sensor 102 according to an embodiment can widen the distance and temperature measurement range through a two-axis scan method, and reflect distance information in thermal image information, thereby controlling cooling and heating when applied to indoor unit control of an air conditioner. It makes it possible to increase the efficiency, move the axis of the motor, and form an empty space in the center so that the cable can enter, so that the power and signal cables are not tangled and clean wiring can be made.

On the other hand, the above-described activity control algorithm 1122 and wear control algorithm 1222 are comfortable (Predicted Mean Vote; It is possible to intelligently control the wind speed, air volume, temperature and humidity of indoor units according to the numerical value of the PMV) index.

The activity control algorithm 1122 detects an object as shown in FIG. 9 according to the acquisition of image and image information collected from the cameras 1110 and 1210, and then, in the activity amount information, the deep neural network of the artificial intelligence unit The joint keypoint is extracted through Network; DNN), and the clothing is extracted in segment units through the convolutional neural network (CNN) of the artificial intelligence unit in the amount of clothing information, and this is used as pose data and clothing data. After classifying each, based on the deep learning DB of the activity amount information that has already been learned and stored, the above pose data is compared and analyzed to calculate the activity amount of the resident, and the activity amount of the current resident. By intelligently controlling the wind speed, air volume, temperature and humidity, and ventilation of the air conditioner according to the values of the predicted mean vote (PMV) according to the conditions, it is possible to provide a comfortable warmth to the residents.

Ventilation can be controlled by opening and closing windows and controlling the ventilation fan installed in the window, opening and closing the exhaust duct, and controlling the exhaust fan installed in the exhaust duct, but in the end, the drive is connected to the window and the ventilation fan, and the exhaust duct and the exhaust fan respectively. It can be controlled in a way that regulates the means.

The wear control algorithm 1222 detects an object as shown in FIG. 10 according to the acquisition of image and image information collected from the cameras 1110 and 1210, and then detects the object as shown in FIG. The joint keypoint is extracted through Network; DNN), and the clothing is extracted in segment units through the convolutional neural network (CNN) of the artificial intelligence unit in the amount of clothing information, and this is converted into pose data and clothing data. After classifying each, the amount of clothing of the resident is calculated by comparing and analyzing the above clothing data based on a deep learning DB of the amount of clothing that has already been learned and stored. By intelligently controlling wind speed, air volume, temperature and humidity, and ventilation according to Predicted Mean Vote (PMV) values that meet the conditions of wear and tear, it can provide a comfortable feeling of heat to residents.

Although described above with reference to the preferred embodiments of the present application, those skilled in the art will variously modify the present invention within the scope not departing from the spirit and scope of the present invention described in the following claims. And it will be appreciated that it can be changed.

10: artificial intelligence robot 20: body
30: head portion 40: leg portion
100: multi-sensor unit 101: voice reception processing unit
102: scan type sensor 103: probe
104: differential pressure sensor
110: distance measuring sensor 111: detector
120: temperature measurement sensor 121: bolometer
130: rotation drive unit 131: a plurality of motors
131a: first motor 131b: second motor
133: rotating body 140: synthesis processing unit
310: mirror 320: lens
1000: artificial intelligence control unit 1100: activity amount intelligence control unit
1110: camera 1120: first multi-controller
1121: main body 1122: activity control algorithm
1200: clothing amount intelligent control unit 1210: camera
1220: second multi-controller 1221: main body
1222: Dressing Control Algorithm 1300: Extraction Program
DNN: Deep Neural Network CNN: Convolutional Neural Network
PMV: Comfort 2000: Terminal

Claims (17)

An artificial intelligence robot 10 equipped with a multi-sensor unit 100 that continuously measures indoor wind speed, air volume, temperature and humidity while driving throughout the room for a set time according to the timer setting;
Active volume intelligent control unit that intelligently controls the wind speed, air volume, temperature, and humidity from the air conditioner through constant monitoring of the amount of activity and clothing of residents living indoors to provide a sense of comfort to residents at all times. An artificial intelligence control unit 1000 composed of 1100 and an intelligent control unit 1200 for the amount of clothing; And
A terminal (2000) in which management necessary for control of an air conditioner is performed in a manner of communicating with the artificial intelligence control unit and the artificial intelligence robot;
Artificial intelligence control system for air conditioning control, characterized in that consisting of. The method of claim 1,
The activity amount intelligent control unit 1100
A camera 1110 that monitors the amount of activity and the amount of clothing of the occupant in a video recording method; And
The artificial intelligence intelligently controls wind speed, air volume, temperature, and humidity based on the current measurement information measured by the multi-sensor unit 100 based on the activity amount of the occupant based on the video recorded from the camera. A first multi-controller 1120;
An artificial intelligence control system for controlling an air conditioner, characterized in that it further comprises. The method of claim 1,
The wear amount intelligent control unit 1200
A camera 1210 that monitors the amount of clothing and the amount of clothing of the occupant in a way that always records an image; And
Artificial intelligence that intelligently controls wind speed, air volume, temperature and humidity, and ventilation through the current measurement information measured by the multi-sensor unit 100 based on the amount of wear of the occupant based on the image recorded from the camera. A second multi-controller 1220;
An artificial intelligence control system for controlling an air conditioner, characterized in that it further comprises. The method of claim 2,
The first multi-controller 1120 of the artificial intelligence
A main body 1121 equipped with an artificial intelligence unit storing a deep learning DB obtained by learning enormous amount of activity information of a resident; And by analyzing the measurement information installed in the main body and provided from the multi-sensor unit 100 with the deep learning DB, the wind speed and air volume, temperature and humidity suitable for the current occupant's activity condition are intelligently determined. A controlling activity control algorithm 1122;
Artificial intelligence control system for air conditioning control, characterized in that it further comprises. The method of claim 2,
The second multi-controller 1220 of the artificial intelligence
A main body 1221 equipped with an artificial intelligence unit storing a deep learning DB obtained by learning a vast amount of information on a resident's amount of clothing;
By analyzing the measurement information installed in the main body and provided from the multi-sensor unit 100 with the deep learning DB, the wind speed and air volume, temperature and humidity suitable for the current occupant's wearing amount condition are intelligently determined. A controlled dressing control algorithm 1222;
Artificial intelligence control system for air conditioning control, characterized in that it further comprises. The method according to claim 4 or 5,
The deep learning DB is extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the activity amount extracted by the extraction program An artificial intelligence control system for air conditioner control, characterized in that the activity control algorithm 1122 intelligently controls wind speed, air volume, temperature and humidity according to the conditions of the current occupant's activity based on a deep learning DB of information. The method according to claim 4 or 5,
The deep learning DB is extracted as activity amount information and clothing amount information through the extraction program 1300 installed in the main body 1121 and 1221, and the amount of clothing extracted by the extraction program. ) Artificial intelligence for air conditioner control, characterized in that the dressing control algorithm 1222 intelligently controls wind speed, air volume, temperature and humidity, and ventilation according to the conditions of the current occupant's wearing amount based on a deep learning DB of information. Control system. According to claim 4
The activity control algorithm 1122 detects an object according to acquisition of image and image information collected from the cameras 1110 and 1210, and then detects the object according to the acquisition of image information collected from the cameras 1110 and 1210, and then, in the activity amount information, a joint through a deep neural network (DNN). Key points are extracted, and clothing is extracted in segment units through a convolutional neural network (CNN) from the amount of clothing, and then classified into pose data and clothing data, respectively, and the amount of activity that has already been learned and stored. Based on the deep learning DB of (活動量) information, the above pose data is compared and analyzed to calculate the amount of activity of the resident, and the comfort that meets the conditions of the current resident's activity amount (Predicted Mean Vote; PMV) by intelligently controlling wind speed, air volume, temperature and humidity to provide comfortable heat to the occupants, an artificial intelligence control system for air conditioner control, The method of claim 5
The wear control algorithm 1222 detects an object according to the acquisition of image and image information collected from the cameras 1110 and 1210, and then detects the object according to the acquisition of image information collected from the cameras 1110 and 1210, and then, in the activity amount information, a joint through a deep neural network (DNN). Key points are extracted, and clothing is extracted in segment units through a convolutional neural network (CNN) from the amount of clothing, and it is classified into pose data and clothing data. Based on the deep learning DB of the amount information, the above clothing data is compared and analyzed, and the amount of clothing of the resident is calculated. An artificial intelligence control system for air conditioner control, characterized in that it intelligently controls wind speed, air volume, temperature and humidity according to the numerical values of Mean Vote (PMV) to provide comfortable warmth to residents. The method of claim 1,
The multi-sensor unit 100 is a kind of multi-sensor device,
A voice recognition processing unit 101 that processes a user's voice;
A scan-type sensor 102 that measures the distance with the user (resident) and the temperature with the user (resident) existing in the indoor measurement area and reflects it in the thermal image; And
Probes 103 and differential pressure sensors 104 for measuring the wind speed and direction in the room, and temperature and humidity;
Including further, wherein the probe (1100) is an artificial intelligence control system for air conditioner control, characterized in that consisting of at least a plurality of probes for wind speed, probe for temperature and humidity, and probe for turbulence. The method of claim 10,
The scan-type sensor 102 is a sensor that measures a distance to an object existing in a measurement area and a temperature between the object and reflects it in a thermal image,
A distance measurement sensor 110 located at the front of the mirror M1 and receiving light reflected from the mirror to measure a distance to an object existing in the measurement area;
A temperature measurement sensor 120 positioned on the rear surface of the mirror M1 and receiving light reflected from the mirror to measure a temperature of an object existing in the measurement area;
A rotation driving unit 130 having a plurality of motors forming two rotation shafts and securing a measurement area of the distance measurement sensor and the temperature measurement sensor through 360 degree rotation left and right and 180 degree vertical rotation; And
A synthesis processing unit 140 for synthesizing distance information and heat distribution information measured by the distance measuring sensor and the temperature measuring sensor, respectively, and reflecting the distance information in a thermal image;
An artificial intelligence control system for controlling an air conditioner, characterized in that it further comprises. The method of claim 11,
The artificial intelligence robot 10
A body portion 20 constituting a body;
A head part 30 connected in a rotatable manner from the top of the body; And
A leg portion 40 that is installed and moved in a manner operated by a power means in the lower portion of the body portion;
An artificial intelligence control system for controlling an air conditioner, comprising: a. The method of claim 12,
The leg portion 40 is composed of a plurality of forefingers,
The pincers are operated by a combination of a plurality of pulleys connected to a belt connected to a gear coupled to a single power motor shaft and gears individually connected to the pulleys. The method of claim 12,
The head portion 30 of the artificial intelligence robot 10 is rotated in a horizontal manner by connecting a spur gear formed at the lower end of the head portion to a gear connected to one power motor shaft in a horizontal direction, while the spur gear It has a structure in which a locking pin is accommodated in a long hole inside the shaft that extends vertically from the lower center of the shaft, and at the same time, a half gear formed at the lower end of the shaft and a spring is formed between the shaft. An artificial intelligence control system for controlling an air conditioner, characterized in that it is operated in a nod operation method through a power motor that provides power required for rotation of the gear. The method of claim 14,
The head unit 30 of the artificial intelligence robot 10 is measurement information provided from an object, based on the distance information value and the temperature information value sensed from the distance measurement sensor 110 and the temperature measurement sensor 120. It is characterized in that it is operated in an inverted horizontal manner in the forward direction of the body part 20, and is operated in an operation method of nodding in the front and rear directions when only the distance information value detected by the distance measurement sensor 110 is detected. Artificial intelligence control system for air conditioner control. The method of claim 15,
The distance measuring sensor 110 is
And a detector 111 located at the front of the mirror and receiving the reflected light through a lens after entering the front of the mirror according to the rotation of the rotation driving unit 130 to measure and output distance information. Artificial intelligence control system for air conditioner control. The method of claim 15,
The temperature measurement sensor 120 is
An air conditioner comprising a bollo meta 121 located in an empty space at the rear of the mirror and receiving light incident on the rear of the mirror according to the rotation of the rotation driving unit 130, converting it into a thermal image signal, and outputting it. Control artificial intelligence control system.

Images