Computer program, information processing method, and information processing system

Abstract

Provided are a computer program, an information processing method, and an information processing system.　The present invention: acquires user information of a user who uses an air conditioning device; executes a dialogue process with the user using a generative AI model which refers to the acquired user information, thereby determining control content with respect to the air conditioning device; and causes a computer to execute a process for controlling the air conditioning device on the basis of the determined control content.

Description

Computer Program, Information Processing Method, and Information Processing System

[0001] The present disclosure relates to a computer program, an information processing method, and an information processing system.

[0002] Conventionally, there has been known a technology that calculates the relative distance between a communication terminal installed in a home and a portable terminal carried by a user, and controls the operation of an electric device according to the calculated relative distance (see, for example, Patent Document 1).

[0003] Japanese Unexamined Patent Publication No. 2002-27571

[0004] In Patent Document 1, it is not possible to determine control content through interactive processing with a user.

[0005] An object of the present disclosure is to provide an information processing apparatus, an information processing method, and an information processing system that can execute interactive processing with a user and determine control content for an air conditioning device based on user response information obtained from the interactive processing.

[0006] A computer program according to a first aspect of the present disclosure is a computer program for causing a computer to: acquire user information of a user who uses an air conditioning device; execute interactive processing with the user using a generative AI model that references the acquired user information, thereby determining control content for the air conditioning device; and execute processing for controlling the air conditioning device based on the determined control content.

[0007] A computer program according to a second aspect of the present disclosure is the computer program according to the first aspect, which is for causing the computer to execute processing for determining the control content for the air conditioning device using the generative AI model that further references environmental information of a space air-conditioned by the air conditioning device.

[0008] A computer program according to a third aspect of the present disclosure is the computer program according to the second aspect, wherein the environmental information is information of at least one of an internal environment and an external environment of the space air-conditioned by the air conditioning device.

[0009] The computer program relating to the fourth aspect of this disclosure is a computer program relating to any one of the first to third aspects that specifies the user information as reference information and causes the computer to execute a process to start a dialogue process with the user in the generated AI model.

[0010] The computer program relating to the fifth aspect of this disclosure is a computer program relating to any one of the first to fourth aspects that specifies the user information as reference information and causes the computer to execute a process to initiate a user interaction process with a generating AI model different from the generating AI model.

[0011] A computer program relating to the sixth aspect of this disclosure is a computer program relating to any one of the first to fifth aspects, wherein the interaction process includes a question inquiring about the user's actions and the user's response to the question.

[0012] A computer program relating to the seventh aspect of this disclosure is a computer program relating to any one of the first to sixth aspects, wherein the dialogue processing includes a question inquiring from the user about the appropriateness of the control content determined using the generated AI model, and the user's response to the question.

[0013] The computer program relating to the eighth aspect of this disclosure is a computer program relating to any one of the first to seventh aspects, which causes the computer to perform a process that causes the generated AI model to compare the user's time away from home with a threshold value, and to decide whether or not to continue operating the air conditioning equipment according to the comparison result.

[0014] The computer program relating to the ninth aspect of this disclosure is a computer program relating to any one of the first to eighth aspects, which causes the computer to execute a process that causes the generated AI model to decide whether or not to stop the air conditioning equipment, taking into account the energy-saving effect of the air conditioning equipment.

[0015] The computer program relating to the tenth aspect of this disclosure is a computer program relating to any one of the first to ninth aspects that determines, based on the user information, whether the period during which the air conditioning equipment is not used is longer than the set period, and if it determines that the period during which the air conditioning equipment is not used is longer than the set period, it causes the computer to execute a process to stop the dialogue process.

[0016] A computer program relating to the eleventh aspect of this disclosure is a computer program relating to any one of the first to tenth aspects, wherein the user information includes at least one of the user's location information, the user's presence information, schedule information, biometric information, image information, and past interaction information with the user.

[0017] A computer program relating to the twelfth aspect of this disclosure, in a computer program relating to the second aspect, wherein the environmental information includes at least one of temperature, humidity, and weather.

[0018] The computer program relating to the 13th aspect of this disclosure is a computer program relating to any one of the 1st to 12th aspects, which uses the generated AI model to determine control content for air conditioning-related equipment other than the air conditioning equipment, and causes the computer to execute a process to control the air conditioning-related equipment based on the determined control content.

[0019] The information processing method relating to the fourteenth aspect of this disclosure involves acquiring user information of a user of an air conditioning device, performing a dialogue process with the user using a generating AI model that references the acquired user information, determining the control content for the air conditioning device, and then having at least one computer perform a process to control the air conditioning device based on the determined control content.

[0020] The information processing system relating to the 15th aspect of this disclosure comprises at least one control unit, which acquires user information of a user of an air conditioning device, and performs a dialogue process with the user using a generating AI model that refers to the acquired user information, thereby determining the control content for the air conditioning device, and controlling the air conditioning device based on the determined control content.

[0021] According to this disclosure, it is possible to perform a dialogue process with the user and determine the control content for the air conditioning equipment based on the user's response information obtained from the dialogue process.

[0022] This is a schematic diagram showing the overall configuration of an air conditioning system according to an embodiment. This is a block diagram showing the configuration of the air conditioning control system. This is a block diagram showing the configuration of the user terminal control system. This is a block diagram showing the configuration of the first server control system. This is a block diagram showing the configuration of the second server control system. This is a conceptual diagram showing an example of a user table. This is a diagram showing an example of a prompt used when a user returns home. This is a schematic diagram showing an example of dialogue when an operation start prompt is entered. This is a diagram showing an example of a prompt used when a user leaves the house. This is a schematic diagram showing a first example of dialogue when an operation stop prompt is entered. This is a schematic diagram showing a second example of dialogue when an operation stop prompt is entered. This is a flowchart showing the procedure of processing executed by the first server.

[0023] The information processing system according to the embodiment will be described in detail below with reference to the drawings. (Embodiment 1) Figure 1 is a schematic diagram showing the overall configuration of the air conditioning system according to the embodiment. The air conditioning system 1 according to the embodiment comprises an air conditioner 100 and a user terminal 200. The air conditioner 100 and the user terminal 200 are connected to each other via a local communication network NW1 such as Wi-Fi (registered trademark).

[0024] The air conditioner 100 is a household or commercial air conditioner for adjusting the temperature, humidity, air cleanliness, etc., of a room, and comprises an indoor unit and an outdoor unit. The air conditioner 100 adjusts the temperature, humidity, etc., of a room by operating in accordance with control commands output from a remote controller 150 or a user terminal 200.

[0025] The user terminal 200 is a device such as a smartphone, tablet, personal computer, or smart speaker operated by the user of the air conditioner 100. The user terminal 200 is equipped with artificial intelligence (AI). The artificial intelligence installed in the user terminal 200 is constructed by the artificial intelligence model MD2, which generates new content in response to inputs such as text, voice, still images, and videos. The artificial intelligence model MD2 may be a proprietary model that has been fine-tuned and / or distilled (lightened) for a specific application. The content generated by the artificial intelligence model MD2 includes text, sound such as voice or music, and images such as still images or videos. The user terminal 200 can create the indoor environment desired by the user by using the artificial intelligence to interact with the user, generate control commands for the air conditioner 100, and control the operation of the air conditioner 100 based on the generated control commands.

[0026] In the example shown in Figure 1, there is one user terminal 200, but there may be multiple user terminals 200 capable of controlling the air conditioner 100.

[0027] The air conditioning system 1 may further include an air purifier 151, a circulator 152, a diffuser 153, lighting equipment 154, and sound equipment 155. The air purifier 151 is a device for purifying the indoor air and providing clean air. The circulator 152 is a device for circulating the indoor air. The diffuser 153 is a device for diffusing fragrance. The lighting equipment 154 is a device for illuminating the indoor space. The sound equipment 155 is a device for playing music, ambient sounds, etc. These devices 151 to 155 are connected to at least one of the air conditioner 100 and the user terminal 200 via a communication network NW1 so as to be able to communicate.

[0028] The air conditioning system 1 may further include a wearable device 250 worn by the user. The wearable device 250 measures, for example, at least one of the user's heart rate, body temperature, sweating, blood pressure, respiration, brain waves, and body composition, and outputs the measurement results as the user's biometric information. The output destination may be the air conditioning unit 100 or the user terminal 200.

[0029] The air conditioning system 1 may further include a first server 310 or a second server 320 accessible from the air conditioner 100 or a user terminal 200. The air conditioner 100 or the user terminal 200 and the first server 310 or the second server 320 are connected to each other via an external communication network NW2, such as the Internet.

[0030] The first server 310 is equipped with a generative AI. The generative AI in the first server 310 is constructed by a generative AI model MD3 that generates new content in response to inputs such as text, audio, still images, and videos. The generative AI model MD3 may include existing models such as GPT (Generative Pretrained Transformer, registered trademark), BERT (Bidirectional Encoder Representations from Transformers), GAN (Generative Adversarial Network), Wave Net, DALL-E, and Stable Diffusion, or it may be a unique model that has been fine-tuned for a specific application from an existing model. The first server 310 generates content such as text, audio, still images, and videos in response to a request from the air conditioner 100 or user terminal 200, and returns the generated content to the requester.

[0031] The second server 320 is a search server that searches for relevant documents or information in response to an input query and returns the relevant documents or information to the requester. Existing search engines such as vector search are used to search for relevant documents and information. In one example, the second server 320 is a general-purpose search server that returns search results in response to external search requests. Alternatively, the second server 320 may be a dedicated search server that can access a database DB that stores information such as operation manuals, service guides, technical handbooks, installation instructions, customer information, operation data during trial operation, and operation data after the start of operation for the air conditioner 100, and searches the database DB in response to external search requests and returns the search results to the requester. The database DB may also register construction information when the air conditioner 100 was installed (installation location of indoor and outdoor units, piping length, floor plan, year of construction), information about the installed air conditioner 100 (model, Ua value, horsepower, etc.), and sales information for each employee.

[0032] The air conditioning system 1 according to the embodiment implements the information processing method described herein by utilizing at least one of the generation AI model MD2 installed on the user terminal 200 and the generation AI model MD3 installed on the first server 310. This information processing method may be implemented using only the content output from the generation AI model MD2 installed on the user terminal 200, or it may be implemented using only the content output from the generation AI model MD3 installed on the first server 310.

[0033] Furthermore, the information processing method disclosed herein may be realized through cooperation between the generation AI model MD2 installed on the user terminal 200 and the generation AI model MD3 installed on the first server 310.

[0034] Furthermore, the information processing method disclosed herein may be implemented using a RAG (Retrieval-Augmented Generation) mechanism. That is, in response to an input query from the air conditioner 100 or user terminal 200, the second server 320 may output documents or information related to the input query, and the first server 310 may generate content based on the documents or information output from the second server 320 and return the content to the air conditioner 100 or user terminal 200. By using a RAG mechanism, more accurate and relevant content can be provided.

[0035] In this embodiment, the user terminal 200 is configured to be equipped with the generation AI model MD2. Alternatively, the generation AI model may be equipped on the air conditioner 100. In this case, the information processing method of this disclosure may be implemented using the generation AI model MD1 (see Figure 2) equipped on the air conditioner 100. Furthermore, the information processing method of this disclosure may be implemented by the cooperation of the generation AI model MD1 of the air conditioner 100 and the generation AI model MD3 of the first server 310.

[0036] Furthermore, both the air conditioner 100 and the user terminal 200 may be equipped with generation AI models. In this case, at least two of the generation AI models MD1 of the air conditioner 100, MD2 of the user terminal 200, and MD3 of the first server 310 may work together to realize the information processing method of this disclosure.

[0037] Figure 2 is a block diagram showing the configuration of the control system of the air conditioner 100. The air conditioner 100 comprises a control unit 101, a storage unit 102, a communication unit 103, a sensor unit 104, and a drive unit 110.

[0038] The control unit 101 includes a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and the like. The ROM in the control unit 101 stores control programs for controlling the operation of the drive unit 110. The CPU in the control unit 101 executes the control programs stored in the ROM and various computer programs stored in the memory unit 102, and controls the operation of the drive unit 110, thereby making the entire device function as the air conditioner 100 in this disclosure.

[0039] The storage unit 102 is equipped with a storage device such as flash memory. The storage unit 102 stores various computer programs to be executed by the control unit 101, as well as data necessary for the execution of these programs.

[0040] The computer program (program product) stored in the memory unit 102 includes a control program PG1 for realizing the information processing method of this disclosure. The control program PG1 may be a single computer program or a group of programs composed of multiple computer programs. The control program PG1 may be executed collaboratively by multiple computers.

[0041] The computer program, including the control program PG1, is provided, for example, on a non-temporary recording medium RM1 on which the computer program is recorded in a readable format. The recording medium RM1 is a portable memory such as a CD-ROM, USB memory, SD card, microSD card, or CompactFlash®. The control unit 101 reads various computer programs from the recording medium RM1 using a reading device (not shown in the figure) and stores the read computer programs in the storage unit 102. The computer programs stored in the storage unit 102 may also be provided via communication. In this case, the control unit 101 acquires the computer programs via communication through the communication unit 103 and stores the acquired computer programs in the storage unit 102.

[0042] The data stored in the storage unit 102 may include at least one of an operation manual, a service guide, a technical handbook, an installation instruction, customer information, operation data during test operation, and operation data after the start of operation.

[0043] When the air conditioner 100 is equipped with generative AI, the generative AI model MD1 that constitutes the generative AI is stored in the storage unit 102. The generative AI model MD1 is a model that generates new content in response to inputs such as text, voice, still images, and moving images. The content generated by the generative AI model MD1 includes sounds such as text, voice or music, and images such as still images or moving images. The generative AI model MD1 is an original model obtained by fine-tuning and / or distilling (weight reducing) an existing model such as GPT (registered trademark), BERT, GAN, Wave Net, DALL-E, Stable Diffusion for a specific application. The generative AI model MD1 may be a multimodal model, or may be a unimodal language model that outputs an answer (text) in response to input of a prompt (text).

[0044] The communication unit 103 includes a communication interface for communicating with a user terminal 200 and the like. As the communication interface included in the communication unit 103, a communication interface compliant with a wireless LAN (Local Area Network) communication standard such as Wi-Fi (registered trademark) can be used. Alternatively, a communication interface compliant with a short-range wireless communication standard such as Bluetooth may be used. The communication unit 103 may be configured to be capable of communicating with an air cleaner 151, a circulator 152, a diffuser 153, a lighting device 154, an audio device 155, and the like in addition to the user terminal 200.

[0045] The communication unit 103 may further include a communication interface for communicating with a first server 310 or a second server 320. As such a communication interface, a wireless communication interface such as Wi-Fi (registered trademark), 3G, 4G, 5G, LTE (Long Term Evolution) can be used.

[0046] The sensor unit 104 includes one or more sensors that measure various physical quantities in an indoor environment or the state of a user. The sensors included in the sensor unit 104 include sensors that measure at least one of indoor temperature, humidity, heat, CO₂ concentration, and illuminance. The sensor unit 104 may also include sensors that measure the user's movement, facial expression, and the like.

[0047] The drive unit 110 includes a compressor 111, a four-way switching valve 112, an indoor fan 113, a flap 114, and the like. The compressor 111 is a device for sucking in a constant-temperature low-pressure gaseous refrigerant and compressing the refrigerant into a high-temperature high-pressure gas. During cooling operation, the refrigerant compressed by the compressor 111 returns to the suction side of the compressor 111 via an outdoor heat exchanger, an expansion valve, a first indoor heat exchanger, a solenoid valve, and a second indoor heat exchanger. In this refrigeration cycle, heat is dissipated by the outdoor heat exchanger functioning as a condenser, and indoor air is cooled by the first and second indoor heat exchangers functioning as evaporators to perform cooling. During heating operation, the four-way switching valve 112 switches the flow path, and heating is performed in a cycle reverse to that in cooling operation.

[0048] The indoor fan 113 is a device for uniformly circulating cooled air or heated air throughout the room. The air volume produced by the indoor fan 113 is controlled by a control command from the control unit 101. By adjusting the air volume produced by the indoor fan 113, the efficiency of cooling and heating can be improved, and energy consumption can be reduced.

[0049] The flap 114 is a device for adjusting the direction of air delivered from the air conditioner 100. The orientation of the flap 114 is controlled by a control command from the control unit 101. By ensuring an appropriate air flow via the flap 114, the efficiency of cooling and heating can be improved, and energy consumption can be reduced.

[0050] Fig. 3 is a block diagram showing the configuration of a control system of the user terminal 200. The user terminal 200 is a terminal device such as a smartphone, a tablet terminal, a personal computer, or a smart speaker, and includes a control unit 201, a storage unit 202, a communication unit 203, an imaging unit 204, a voice input unit 205, a voice output unit 206, a positioning unit 207, an operation unit 208, a display unit 209, and the like.

[0051] The control unit 201 includes a CPU, ROM, RAM, etc. The ROM in the control unit 201 stores control programs and the like that control the operation of each hardware component of the user terminal 200. The CPU in the control unit 201 reads and executes the control programs stored in the ROM and the computer programs stored in the memory unit 202, and controls the operation of the hardware components, thereby making the entire device function as the user terminal 200 in this disclosure. The RAM in the control unit 201 temporarily stores data used during the execution of calculations.

[0052] In this embodiment, the control unit 201 is configured to include a CPU, ROM, and RAM. Alternatively, the control unit 201 may be one or more control circuits or processing circuits, such as a GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), quantum processor, or volatile or non-volatile memory. The control unit 201 may also include functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from the start of measurement to the end of measurement, and a counter for counting numbers.

[0053] The storage unit 202 includes a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The storage unit 202 stores various computer programs executed by the control unit 201 and various data used by the control unit 201.

[0054] The computer program (program product) stored in the memory unit 202 includes a control program PG2 for realizing the information processing method of the present disclosure. The control program PG2 may be a single computer program or a group of programs composed of multiple computer programs. The control program PG2 may be executed collaboratively by multiple computers. For example, the information processing method of the present disclosure may be realized by the collaboration of a control program PG1 installed in the air conditioner 100 and a control program PG2 installed in the user terminal 200.

[0055] The computer program, including the control program PG2, is provided, for example, on a non-temporary recording medium RM2 on which the computer program is recorded in a readable format. The recording medium RM2 is a portable memory such as a CD-ROM, USB memory, SD card, microSD card, or CompactFlash®. The control unit 201 reads various computer programs from the recording medium RM2 using a reading device (not shown in the figure) and stores the read computer programs in the storage unit 202. The computer programs stored in the storage unit 202 may also be provided via communication. In this case, the control unit 201 acquires the computer programs via communication through the communication unit 203 and stores the acquired computer programs in the storage unit 202.

[0056] The memory unit 202 may include a generation AI model MD2. The generation AI model MD2 is a model that generates new content in response to inputs such as text, voice, still images, and videos. The content generated by the generation AI model MD2 includes text, sound such as voice or music, and images such as still images or videos. The generation AI model MD2 is a unique model that has been distilled (lightened) for a specific purpose from existing models such as GPT®, BERT, GAN, Wave Net, DALL-E, and Stable Diffusion. The generation AI model MD2 may be a multimodal model, or it may be a unimodal language model that outputs a response (text) in response to a prompt (text) input.

[0057] The communication unit 203 is equipped with a communication interface for communicating with the air conditioner 100 and the like. The communication interface provided by the communication unit 203 can be a communication interface compliant with a wireless LAN communication standard such as Wi-Fi (registered trademark). Alternatively, a communication interface compliant with a short-range wireless communication standard such as Bluetooth may be used. In addition to the air conditioner 100, the communication unit 203 may be configured to communicate with an air purifier 151, a circulator 152, a diffuser 153, lighting equipment 154, sound equipment 155, and the like.

[0058] The communication unit 203 further includes a communication interface for communicating with the first server 310 or the second server 320. Such a communication interface can be a wireless communication interface such as Wi-Fi®, 3G, 4G, 5G, or LTE (Long Term Evolution).

[0059] The imaging unit 204 is equipped with an image sensor such as a CMOS (Complementary Metal Oxide Semiconductor) or a CCD (Charge-Coupled Device), and generates an image (still image or video) of the object to be imaged by imaging the object. The imaging unit 204 outputs the image data related to the generated image to the control unit 201. The control unit 201 stores the input image data in the storage unit 202. The control unit 201 may display the image based on the image data on the display unit 209, or transmit it to the outside via the communication unit 203.

[0060] The audio input unit 205 includes a microphone for collecting sound, a processing circuit for converting the collected sound into a digital signal (audio data), and the like. The audio data generated in the audio input unit 205 is sent to the control unit 201, where appropriate processing such as noise reduction is performed. The audio data generated in the audio input unit 205 is also stored in the storage unit 202 or transmitted to the first server 310 via the communication unit 203.

[0061] The audio output unit 206 is equipped with a speaker that outputs sound. The audio output unit 206 reproduces sound based on the audio data provided by the control unit 201. In this embodiment, the audio input unit 205 and the audio output unit 206 are configured to input and output sound, but sounds other than voice (ambient sounds, music, etc.) may be input to the audio input unit 205 and sounds other than voice (ambient sounds, music, etc.) may be output from the audio output unit 206.

[0062] The positioning unit 207, for example, is equipped with a GPS (Global Positioning System) receiver and receives radio waves transmitted from GPS satellites to determine the current location of the user terminal 200. The positioning unit 207 outputs location information related to the determined current location of the user terminal 200 to the control unit 201.

[0063] The operation unit 208 is equipped with operating devices such as a touch panel, keyboard, and switches, and accepts various inputs and operations from the user. The control unit 201 acquires information input through the operation unit 208 and performs appropriate control based on the various operation information provided by the operation unit 208.

[0064] The display unit 209 is equipped with a display device such as a liquid crystal monitor or an organic EL (Electro-Luminescence) monitor, and displays information that should be notified to the user in response to instructions from the control unit 201.

[0065] Figure 4 is a block diagram showing the configuration of the control system of the first server 310. The first server 310 is a dedicated or general-purpose server computer and includes a control unit 311, a storage unit 312, a communication unit 313, and the like.

[0066] The control unit 311 includes a CPU, ROM, RAM, etc. The ROM in the control unit 311 stores control programs and the like that control the operation of each hardware component of the first server 310. The CPU in the control unit 311 reads and executes the control programs stored in the ROM and the computer programs stored in the memory unit 312, and controls the operation of each hardware component, thereby making the entire device function as the first server 310 in this disclosure. The RAM in the control unit 311 temporarily stores data used during the execution of calculations.

[0067] In this embodiment, the control unit 311 is configured to include a CPU, ROM, and RAM, but the configuration of the control unit 311 is not limited to the above. The control unit 311 may also include functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from the time a measurement start instruction is given until a measurement end instruction is given, and a counter for counting numbers.

[0068] The storage unit 312 includes a storage device such as an HDD or SSD. The storage unit 312 stores various computer programs executed by the control unit 311, various data acquired through the communication unit 313, and various data used by the control unit 311.

[0069] The computer program (program product) stored in the memory unit 312 may include a control program PG3 for realizing the information processing method of the present disclosure. The control program PG3 may be a single computer program or a group of programs composed of multiple computer programs. The control program PG3 may be executed collaboratively by multiple computers. For example, the control program PG3 may realize the information processing method of the present disclosure by collaborating with a control program PG1 installed in the air conditioner 100 or a control program PG2 installed in the user terminal 200.

[0070] The computer program, including the control program PG3, is provided, for example, on a non-temporary recording medium RM3 on which the computer program is recorded in a readable format. The recording medium RM3 is a portable memory such as a CD-ROM, USB memory, SD card, microSD card, or CompactFlash®. The control unit 311 reads various computer programs from the recording medium RM3 using a reading device (not shown in the figure) and stores the read computer programs in the storage unit 312. The computer programs stored in the storage unit 312 may also be provided via communication. In this case, the control unit 311 acquires the computer programs via communication through the communication unit 313 and stores the acquired computer programs in the storage unit 312.

[0071] The memory unit 312 may include a generation AI model MD3. The generation AI model MD3 is a model that generates new content in response to inputs such as text, voice, still images, and videos. The content generated by the generation AI model MD3 includes text, voice or music, images such as still images or videos, etc. The generation AI model MD3 may be an existing model such as GPT®, BERT, GAN, Wave Net, DALL-E, or Stable Diffusion, or it may be a unique model fine-tuned for a specific application. The generation AI model MD3 may be a multimodal model, or it may be a unimodal language model that outputs a response (text) in response to a prompt (text) input.

[0072] The communication unit 313 is equipped with a communication interface for communicating with a user terminal 200 or the like. The communication interface provided by the communication unit 313 can be a wireless communication interface such as 3G, 4G, 5G, or LTE (Long Term Evolution).

[0073] The first server 310 may include an operation unit for receiving operations from administrators, a display unit for displaying information to be notified to administrators, etc. The first server 310 may be a single computer, or it may be a computer system composed of multiple computers and peripheral devices.

[0074] Figure 5 is a block diagram showing the configuration of the control system of the second server 320. The second server 320 is a dedicated or general-purpose server computer and includes a control unit 321, a storage unit 322, a communication unit 323, and the like.

[0075] The control unit 321 includes a CPU, ROM, RAM, etc. The ROM in the control unit 321 stores control programs and the like that control the operation of each hardware component of the second server 320. The CPU in the control unit 321 reads and executes the control programs stored in the ROM and the computer programs stored in the memory unit 322, and controls the operation of each hardware component, thereby making the entire device function as the second server 320 in this disclosure. The RAM in the control unit 321 temporarily stores data used during the execution of calculations.

[0076] In this embodiment, the control unit 321 is configured to include a CPU, ROM, and RAM, but the configuration of the control unit 321 is not limited to the above. The control unit 321 may also include functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from the time a measurement start instruction is given until a measurement end instruction is given, and a counter for counting numbers.

[0077] The storage unit 322 includes a storage device such as an HDD or SSD. The storage unit 322 stores various computer programs executed by the control unit 321, various data acquired through the communication unit 323, and various data used by the control unit 321.

[0078] The storage unit 322 may also include a database DB for storing information such as the operation manual, service guide, technical handbook, installation instructions, customer information, operation data during trial operation, and operation data after the start of operation for the air conditioner 100. The database DB may further register construction information when the air conditioner 100 was installed (installation location of indoor and outdoor units, piping length, floor plan, and year of construction), information about the installed air conditioner 100 (model, Ua value, horsepower, etc.), and sales information for each employee.

[0079] The communication unit 323 is equipped with a communication interface for communicating with a user terminal 200 or the like. The communication interface provided by the communication unit 323 can be a wireless communication interface such as 3G, 4G, 5G, or LTE (Long Term Evolution).

[0080] The second server 320 may include an operation unit for receiving commands from administrators, a display unit for displaying information to be notified to administrators, etc. The second server 320 may be a single computer, or it may be a computer system composed of multiple computers and peripheral devices.

[0081] In the air conditioning system 1 according to Embodiment 1, at least one of the generated AI models MD1 to MD3 is used to perform dialogue processing with a user using the air conditioner 100 and to determine the control content for the air conditioner 100. Below, we will describe a configuration in which dialogue processing with a user is performed using the generated AI model MD3 installed on the first server 310 to determine the control content for the air conditioner 100.

[0082] The first server 310 acquires user information of users using the air conditioner 100 and performs interaction processing with the user using the generating AI model MD3 which references the acquired user information. The acquired user information is stored in the user table UTB in the storage unit 312. Figure 6 is a conceptual diagram showing an example of the user table UTB. The user table UTB stores, for example, an identifier that identifies the user, information about the air conditioner equipment the user is using, the user's location information, occupancy information, schedule information, biometric information, image information, past interaction information, information about cohabitants, etc.

[0083] The identifier is the user's name or user ID. The information about the air conditioning equipment includes the model name, model number, air conditioning capacity, installation location, and size of the installation location of the air conditioner 100 installed in the user's home. If other air conditioners or air conditioning-related equipment are installed in the user's home, this information is also stored in the user table UTB. Air conditioning-related equipment includes the air purifier 151, circulator 152, and diffuser 153 mentioned above, as well as humidifiers, ventilation devices, etc.

[0084] The user's location information is the location information determined by the positioning unit 207 of the user terminal 200. Although the positioning unit 207 actually determines the location of the user terminal 200, since the user terminal 200 is expected to be carried by the user, the location information determined by the positioning unit 207 can be considered as the user's location information. The first server 310 periodically acquires location information from the user terminal 200 and stores the acquired location information in the user table UTB. Alternatively, the first server 310 may request location information from the user terminal 200 as needed and store the location information returned from the user terminal 200 in the user table UTB in response to the request.

[0085] Occupancy information indicates whether or not a user is at home. Existing occupancy detection methods are used, such as methods using camera images, methods using sensors such as motion sensors, sound sensors, and CO2 concentration sensors, and methods using communication such as Wi-Fi and Bluetooth. The first server 310 acquires the occupancy information detected by the existing detection method via communication and stores the acquired occupancy information in the user table UTB.

[0086] Schedule information is information representing a user's schedule. Schedule information is information registered by the user themselves in a calendar application or schedule management service. The first server 310 retrieves the user's schedule information from the calendar application or schedule management service and stores the retrieved schedule information in the user table UTB. The schedule information may include information that allows for the estimation of the user's departure and return times. Alternatively, or together with the schedule information, information on the user's behavioral patterns, including departure and return times, may be stored in the user table UTB. The user's behavioral patterns are estimated by the first server 310, for example, based on past schedule information.

[0087] Biometric information includes the user's heart rate, body temperature, sweating, blood pressure, respiration, brain waves, and body composition, and is measured by a wearable device 250 worn by the user. The first server 310 acquires the biometric information measured by the wearable device 250 through the user terminal 200 and stores the acquired biometric information in the user table UTB.

[0088] The image information is an image captured by photographing the user or a cohabitant of the user. The captured image is an image obtained by photographing the user or a cohabitant with the imaging unit 204 of the user terminal 200. If imaging devices such as security cameras or network cameras are installed in the user's home, the captured image may also be one captured by these imaging devices. The first server 310 acquires the captured image captured by the user terminal 200 or the imaging device installed in the user's home via communication and stores the acquired captured image in the user table UTB.

[0089] Past dialogue information represents the content of past conversations between the user and the generated AI model MD3. This dialogue information is stored in the user table UTB as either text or audio data.

[0090] Information about cohabitants includes whether or not they live with the user, the attributes of the cohabitants (relationship with the user, age, etc.), and the cohabitants' schedule information.

[0091] In this embodiment, for convenience, we will describe a configuration in which all user information is consolidated and stored in the user table UTB, but each piece of information constituting the user information may be stored individually in the storage unit 312.

[0092] Furthermore, although this embodiment describes a configuration in which user information stored in the user table UTB is referenced, the first server 310 may request necessary user information from the user terminal 200 at appropriate times and use the user information transmitted from the user terminal 200 in response to the request without storing it in the user table UTB.

[0093] The first server 310 uses the generating AI model MD3, which references user information stored in the user table UTB, to perform interaction processing with the user.

[0094] The following describes the dialogue processing when a user returns home. Figure 7 shows an example of a prompt used when a user returns home. When a user returns home, the control unit 311 of the first server 310 inputs a prompt as shown in Figure 7 (hereinafter referred to as the "operation start prompt") to the generating AI model MD3 and starts the dialogue processing with the user. The control unit 311 estimates the user's return time by referring to the schedule information in the user table UTB, for example, and inputs the operation start prompt to the generating AI model MD3 at a timing earlier than the estimated return time. Alternatively, the control unit 311 may input the operation start prompt to the generating AI model MD3 at any timing. Since the operation start prompt is a fixed phrase, it may be stored in advance in the storage unit 312.

[0095] The operation start prompt includes the role of the generated AI model MD3, the commands to the generated AI model MD3, the reference information to be consulted when executing the processing based on the commands, and the purpose of the processing. In the example in Figure 7, the generated AI model MD3 is given the role of "Air Conditioner Operation Concierge." When the operation start prompt is input to the generated AI model MD3, it is expected to act as an advisor to respond to user requests and inquiries regarding the operation of the air conditioner 100.

[0096] The operation start prompt is set as an instruction to the generating AI model MD3 to estimate the user's return time based on user information, to start the operation of the air conditioner 100 so that an appropriate air-conditioned environment can be provided to the user upon their return, and, if there is no confidence in the estimation, to proceed with dialogue processing with the user and perform appropriate operation.

[0097] The operation start prompt has user information set as reference information. The user information to be referenced includes information on the air conditioning equipment used by the user, the user's location information, occupancy information, schedule information, biometric information, image information, past dialogue information, and information on cohabitants. In this embodiment, since the user information is stored in the user table UTB, the control unit 311, which inputs the operation start prompt to the generating AI model MD3 and executes processing by the generating AI model MD3, refers to the user table UTB when executing processing by the generating AI model MD3.

[0098] Furthermore, it is set that, if necessary, environmental information, operating information of the air conditioner 100, and operating information of other air conditioners may be used as reference information. Environmental information is information on at least one of the internal and external environments of the space air-conditioned by the air conditioner 100. Internal environment information is information such as temperature, humidity, and cleanliness of the space air-conditioned by the air conditioner 100. External environment information is information such as temperature, humidity, atmospheric conditions (pollen, PM2.5, etc.), and weather outside. Operating information of the air conditioner 100 includes on / off information and operating settings information of the air conditioner 100. Operating information of other air conditioners includes on / off information and operating settings information of air conditioners other than the air conditioner 100, or air conditioning-related equipment.

[0099] The purpose of the process set in the operation start prompt represents what the prompt is trying to achieve. In the example in Figure 7, the purpose is set to provide the user with an appropriate air conditioning environment when the user returns home.

[0100] Figure 8 is a schematic diagram showing an example of dialogue when the operation start prompt is entered. The control unit 311 of the first server 310 performs dialogue processing with the user based on the user information stored in the user table UTB. The control unit 311 performs dialogue processing by generating a question sentence to the user using the generation AI model MD3 and sending it to the user terminal 200, and receiving the user's reply sentence from the user terminal 200 via the communication unit 313. The example dialogue shown in Figure 8 includes utterances (mainly question sentences) generated by the generation AI model MD3 and utterances (mainly reply sentences) from the user.

[0101] The operation start prompt illustrated in Figure 7 includes a command to estimate the user's return time based on user information. When this operation start prompt is input to the generating AI model MD3, the control unit 311, which executes the processing by the generating AI model MD3, first checks the user's location information and schedule information from the user table UTB and estimates the user's return time. If the control unit 311 determines that the user will be returning home soon (for example, if it determines that the estimated return time is within 10 minutes from the current time), it sends the question "May I start operation?" from the communication unit 313 to the user terminal 200. This question may be a standard phrase or it may be generated by the generating AI model MD3.

[0102] Furthermore, the control unit 311 may start operating the air conditioner 100 without sending a question to the user terminal 200. For example, if the user's return time, outside temperature, etc., have not changed in the past few days and the same control is always being performed, the control unit 311 may start operating the air conditioner 100 a set time before the estimated time of the user's return (for example, 10 minutes before), without relying on dialogue processing.

[0103] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 8 shows that the response received was "Start in 10 minutes."

[0104] The control unit 311 inputs the received response into the generating AI model MD3 and generates the next question. The example in Figure 8 shows that the question "The temperature is higher today than yesterday. Shall we lower the set temperature?" has been generated. Such questions are generated by the generating AI model MD3, which refers to the external environment information stored in the user table UTB. For example, if the external environment information indicates that the temperature or weather has changed from the previous day, the generating AI model MD3 generates a question asking whether to change the set temperature. The control unit 311 transmits the question generated by the generating AI model MD3 to the user terminal 200 via the communication unit 313.

[0105] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 8 shows that the response received was "Please lower it."

[0106] The control unit 311 inputs the received response into the generation AI model MD3 and generates the next question. The example in Figure 8 shows that the question "Do you want to turn on the living room air conditioner?" has been generated. Such questions are generated by the generation AI model MD3, which refers to the information on the air conditioning equipment stored in the user table UTB. The control unit 311 transmits the question generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313.

[0107] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The control unit 311 generates the next question using the generation AI model MD3 as needed. The example in Figure 8 shows that the question "Do you want to start another air conditioner?" has been generated. Such a question is generated by the generation AI model MD3 when information about an air conditioner other than the air conditioner 100 is stored in the user table UTB. The control unit 311 transmits the question generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313.

[0108] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 8 shows that the response received was "The living room is enough." Upon receiving this response, the control unit 311 can determine that the only air conditioner 100 to be controlled is the one installed in the living room, that operation should start after 10 minutes, and that the set temperature should be low. Based on the determined control content, the control unit 311 generates a control command to the air conditioner 100 and controls the air conditioner 100 by transmitting the generated control command to the air conditioner 100. In this example, the control unit 311 may send a control command to the air conditioner 100 after 10 minutes indicating that operation should start with a low set temperature, or it may send a control command to the air conditioner 100 immediately indicating that operation should start after a 10-minute waiting period with a low set temperature.

[0109] The control unit 311 may generate further questions using the generation AI model MD3 by referring to the reference information described in the operation start prompt. For example, if information on air conditioning-related equipment is stored in the user table UTB, the generation AI model MD3 may generate a question such as, "It's dry today. Shall we turn on the humidifier as well?" The control unit 311 transmits the question generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313. The control unit 311 also receives the user's response to the transmitted question from the user terminal 200 and executes control over the air conditioning-related equipment according to the user's response.

[0110] As described above, the control unit 311 can provide the user with a comfortable air-conditioned environment at the estimated time of return home by performing dialogue processing with the user using the generating AI model MD3 which refers to user information.

[0111] Next, we will explain the dialogue processing that occurs when a user leaves the premises. Figure 9 shows an example of a prompt used when a user leaves the premises. When a user leaves the premises, the control unit 311 of the first server 310 inputs a prompt as shown in Figure 9 (hereinafter referred to as the "operation stop prompt") to the generating AI model MD3 and starts the dialogue processing with the user. The control unit 311 estimates the time the user leaves the premises by referring to the schedule information in the user table UTB, for example, and inputs the operation stop prompt to the generating AI model MD3 at a timing after the estimated time the user leaves the premises. Since the operation stop prompt is a standard phrase, it may be stored in advance in the storage unit 312.

[0112] Instead of using schedule information, the control unit 311 may estimate a user's departure from the user's location information, presence information, and image information stored in the user table UTB. For example, the control unit 311 may learn the scenes (image information) in which the user is present in the room using a machine learning model, and if it detects a scene different from the usual presence scene, it may estimate that the user is out. Normal presence scenes include, for example, scenes where the user is wearing loungewear, not carrying a bag, not wearing shoes, and not wearing makeup. Conversely, if the control unit 311 can recognize scenes where the user is wearing outdoor clothes, carrying a bag, wearing shoes, or wearing makeup using the machine learning model, it should estimate that the user is out. Existing models such as VGGNet, ResNet, and Vision Transformer can be used as machine learning models for scene detection. The control unit 311 may also estimate the departure of cohabitants based on image information, information about cohabitants, etc., in addition to the user's departure.

[0113] Furthermore, the control unit 311 may distinguish and estimate short-term and long-term outings by referring to schedule information and image information stored in the user table UTB. In this embodiment, short-term and long-term are distinguished from the viewpoint of energy saving, CO2 emissions, or electricity costs of the air conditioner 100. That is, short-term refers to a period in which it is more effective not to stop the operation of the air conditioner 100 from the viewpoint of energy saving, CO2 emissions, or electricity costs. Conversely, long-term refers to a period in which it is more effective to stop the operation of the air conditioner 100 from the viewpoint of energy saving, CO2 emissions, or electricity costs. Energy saving effect is an indicator that represents the amount of reduction in energy consumption (or the rate of reduction in energy consumption) in the air conditioner 100. It is known that immediately after the start of operation of the air conditioner 100, the compressor is operated at a high load in order to quickly transition the room temperature to the set temperature, and the amount of energy consumed is large. Therefore, from the standpoint of energy saving, it may be recommended to keep the air conditioner 100 running continuously rather than frequently turning it off and restarting it. The control unit 311 may estimate that the user is planning a long trip or business trip if it can determine from the schedule information that the user is planning a trip or business trip. In addition, if the control unit 311 can detect from the image information that the user is about to go out without luggage or is about to go out in loungewear, it may estimate that the user is about to go out for a short period of time.

[0114] The operation stop prompt includes the role of the generating AI model MD3, the command to the generating AI model MD3, the reference information to be consulted when executing the processing based on the command, and the purpose of the processing. In the example in Figure 9, the generating AI model MD3 is given the role of "Air Conditioner Operation Concierge." When the operation stop prompt is input to the generating AI model MD3, it is expected to act as an advisor to respond to user requests and inquiries regarding the operation of the air conditioner 100.

[0115] The operation stop prompt is set as an instruction to the generating AI model MD3 to estimate whether the air conditioner 100 has been left on based on user information, to stop operation if the user has forgotten to turn it off, and if the estimation is not confident, to proceed with dialogue with the user and perform appropriate operation.

[0116] The operation stop prompt has user information set as reference information. The user information to be referenced includes information about the air conditioning equipment the user is using, the user's location information, occupancy information, schedule information, biometric information, image information, past dialogue information, and information about cohabitants. In this embodiment, since the user information is stored in the user table UTB, the control unit 311, which inputs the operation stop prompt to the generating AI model MD3 and executes processing by the generating AI model MD3, refers to the user table UTB when executing processing by the generating AI model MD3. It is also set that, if necessary, the system may refer to the environmental information and operating information of the air conditioner 100, and the operating information of other air conditioners.

[0117] The purpose of the process set for the operation stop prompt represents what the prompt is trying to achieve. In the example in Figure 9, the objectives are energy saving and improved user convenience.

[0118] Figure 10 is a schematic diagram showing a first example of dialogue when a stop prompt is entered. The control unit 311 of the first server 310 performs dialogue processing with the user based on the user information stored in the user table UTB. The control unit 311 performs dialogue processing by generating a question sentence to the user using the generation AI model MD3 and sending it to the user terminal 200, and receiving the user's reply sentence from the user terminal 200 via the communication unit 313. The example dialogue shown in Figure 10 includes utterances (mainly question sentences) generated by the generation AI model MD3 and utterances (mainly reply sentences) from the user.

[0119] The operation stop prompt illustrated in Figure 9 includes a command to estimate whether the air conditioner 100 has been left running based on user information. When this operation stop prompt is input to the generating AI model MD3, the control unit 311, which executes the processing by the generating AI model MD3, checks the user's schedule information, image information, occupancy information, cohabitant information, etc. from the user table UTB and estimates whether the user (and cohabitants) are out. The control unit 311 also checks the information of the air conditioner 100 from the user table UTB and determines whether the air conditioner 100 is running or not. If the control unit 311 estimates that the user (and cohabitants) are out and determines that the air conditioner 100 is running, it sends a question, for example, "Do you want to turn off the air conditioner?" to the user terminal 200 from the communication unit 313. This question may be a standard phrase or may be generated by the generating AI model MD3.

[0120] The control unit 311 may stop the operation of the air conditioner 100 without sending a question to the user terminal 200. For example, if the control unit 311 detects that a user who always turns off the air conditioner 100 when going out has forgotten to turn it off, the control unit 311 may stop the operation of the air conditioner 100 without relying on dialogue processing.

[0121] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 10 shows that the response received was "It's okay to leave it on."

[0122] The control unit 311 inputs the received response into the generating AI model MD3 and generates the next utterance. The example in Figure 10 shows that the utterance "Continue operation." is generated. When such an utterance is generated, the control unit 311 terminates the interaction process with the user without performing any control of the air conditioner 100.

[0123] Even if the user wishes to continue operation, the control unit 311 may, through dialogue processing, notify the user of the benefits of stopping the air conditioner 100 (reduction in electricity consumption, reduction in CO2 emissions, decrease in electricity costs, etc.) if it is preferable to stop the operation of the air conditioner 100 from the standpoint of energy saving, CO2 emissions, or electricity costs, and encourage the user to stop the operation of the air conditioner 100.

[0124] Figure 11 is a schematic diagram showing a second example of dialogue when a stop operation prompt is entered. The dialogue example in Figure 11 shows a dialogue example when the user is planning a long-term absence. When a stop operation prompt is entered into the generating AI model MD3, the control unit 311, which executes processing by the generating AI model MD3, checks the user's schedule information, image information, occupancy information, cohabitant information, etc. from the user table UTB and estimates the user's (and cohabitants') absence. A threshold may be set to distinguish between short-term and long-term absences. For example, if the threshold is set to 3 days, the control unit 311 estimates absences of less than 3 days as short-term absences and absences of 3 days or more as long-term absences. If the control unit 311 estimates a long-term absence due to travel, it sends a question such as "Are you going on a trip tomorrow?" from the communication unit 313 to the user terminal 200. This question may be a standard phrase or may be generated by the generating AI model MD3.

[0125] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 11 shows that the response received was "I'm going on a trip until next Wednesday."

[0126] The control unit 311 inputs the received response into the generation AI model MD3 and generates the next question. The example in Figure 11 shows that the question "Is it unnecessary to operate the air conditioner until next Wednesday?" has been generated. Such questions are used to reconfirm the user's response and are generated by the generation AI model MD3. The control unit 311 transmits the question generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313.

[0127] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 11 shows that the response received was "It's not necessary."

[0128] The control unit 311 inputs the received response into the generation AI model MD3 and generates the next question. The example in Figure 11 shows that the question "Is it unnecessary to operate the air conditioning until you return from your trip?" has been generated. Such questions are used to confirm the user's response and are generated by the generation AI model MD3. The control unit 311 transmits the question generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313.

[0129] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 11 shows that the response received was "It's not needed." Upon receiving this response, it is determined that the air conditioner 100 does not need to be operated until the user returns from their trip, so the control unit 311 can decide to control the air conditioner 100 not to operate until then.

[0130] At this time, the control unit 311 may, for example, generate a speech statement such as "Until then, we will be in dialogue-stopped mode" using the generation AI model MD3, and transmit the generated speech statement to the user terminal 200 via the communication unit 313. Here, the above-mentioned "dialogue-stopped mode" means stopping the dialogue processing with the user. This reduces power consumption associated with the generation AI model MD3 functioning while the user is absent, and eliminates unnecessary notifications (dialogue processing) to the user.

[0131] The control unit 311 generates a control command to the air conditioner 100 based on the determined control content, and controls the air conditioner 100 by transmitting the generated control command to the air conditioner 100. In this example, the control unit 311 transmits a control command to the air conditioner 100 to stop operation after the user has departed on their trip.

[0132] Furthermore, the control unit 311 may generate a question such as "Did you enjoy your trip? Will you resume driving tomorrow?" in accordance with the user's planned return date, and send it to the user terminal 200 via the communication unit 313. Such a question is intended to confirm with the user whether they want to deactivate the "dialogue suspension mode" and is generated by the generating AI model MD3, which references user information.

[0133] The control unit 311 receives the user's response to the transmitted question from the user terminal 200. The example in Figure 11 shows that the response received was "Please start tomorrow." In this case, the control unit 311 cancels the "dialogue stop mode" and resumes processing by the generated AI model MD3.

[0134] As described above, the control unit 311 uses the generating AI model MD3, which references user information, to perform dialogue processing with the user, thereby preventing the user from forgetting to turn off the air conditioner 100 when they leave the house. Furthermore, even for the same type of outing, it is possible to control the system so that for short outings the air conditioner 100 continues to operate from the perspective of energy saving, CO2 emissions, or electricity costs, and for long outings the air conditioner 100 stops operating.

[0135] Figure 12 is a flowchart showing the procedure of processing performed by the first server 310. The control unit 311 of the first server 310 acquires user information of users using the air conditioner 100 via communication (step S101). User information includes an identifier to identify the user, information on the air conditioner equipment used by the user, the user's location information, occupancy information, schedule information, biometric information, image information, past conversation information, information on cohabitants, etc. The control unit 311 stores the acquired user information in the user table UTB. Note that the acquisition of user information is performed as needed, and the user information stored in the user table UTB may be updated as appropriate.

[0136] The control unit 311 performs user interaction processing using the generating AI model MD3, which refers to user information stored in the user table UTB (step S102). The control unit 311 starts user interaction processing by inputting a start prompt or a stop prompt to the generating AI model MD3 according to the user information. If the user is estimated to be returning home based on the user's schedule information, location information, etc., included in the user information, the control unit 311 should input a start prompt to the generating AI model MD3, and if the user is estimated to be leaving home, it should input a stop prompt to the generating AI model MD3.

[0137] The control unit 311 transmits utterances (mainly question sentences) generated by the generation AI model MD3 to the user terminal 200 via the communication unit 313, and executes dialogue processing by receiving user utterances (mainly answer sentences) returned from the user terminal 200 via the communication unit 313. The control unit 311 may also input the user utterances received from the user terminal 200 into the generation AI model MD3, and have the generation AI model MD3 generate the next question sentence for the user.

[0138] The control unit 311 determines whether the control content for the air conditioner 100 has been determined through interaction with the user (step S103). If the start, stop, and continuation of operation of the air conditioner 100, the set temperature, etc., are determined through interaction with the user, the control unit 311 determines that the control content has been determined. If the control content has not been determined (S103: NO), the control unit 311 returns to step S102 and continues the interaction with the user.

[0139] If the control unit 311 determines that the control content for the air conditioner 100 has been determined (S103: YES), it generates a control command for the air conditioner 100 according to the determined control content and controls the air conditioner by transmitting the generated control command to the air conditioner 100 (step S104).

[0140] As described above, in this embodiment, user information of a user using the air conditioner 100 is acquired, and by using the generating AI model MD3, which refers to the acquired user information, to perform a dialogue process with the user, the control content for the air conditioner 100 can be determined. In this embodiment, even if the user does not explicitly instruct the operation of the air conditioner 100, the operation of the air conditioner 100 can be controlled in response to the dialogue process with the user.

[0141] In this embodiment, the system is configured to perform user interaction processing using the generation AI model MD3 installed on the first server 310. However, it is also possible to configure the system to perform user interaction processing using the generation AI model MD1 installed on the air conditioner 100. In this case, the control unit 101 of the air conditioner 100 can start user interaction processing by inputting prompts, as shown in Figure 7 or Figure 9, to the generation AI model MD1 at an appropriate timing, thereby specifying user information as reference information. The control unit 101 of the air conditioner 100 can determine the control content of its own unit through user interaction processing and control the operation of its own unit based on the determined control content.

[0142] Alternatively, the configuration may involve using the generated AI model MD2 installed on the user terminal 200 to perform dialogue processing with the user. In this case, the control unit 201 of the user terminal 200 can start dialogue processing with the user by inputting prompts, as shown in Figure 7 or Figure 9, to the generated AI model MD2 at an appropriate timing, thereby specifying user information as reference information. The control unit 201 of the user terminal 200 can determine the control content of the air conditioner 100 through dialogue processing with the user, generate control commands to the air conditioner 100 based on the determined control content, and control the operation of the air conditioner 100 by transmitting the generated control commands to the air conditioner 100.

[0143] Furthermore, the user interaction process and the process of determining the control content of the air conditioner 100 may be performed by separate generating AIs. For example, the user interaction process may be executed by inputting a prompt to the generating AI model MD2 of the user terminal 200, specifying user information as reference information and instructing it to start the user interaction process. At the same time, the control content of the air conditioner 100 may be determined by inputting a prompt to the generating AI model MD1 of the air conditioner 100 (or the generating AI model MD3 of the first server 310), specifying the interaction result and user information as reference information and instructing it to determine the control content of the air conditioner 100.

[0144] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended.

[0145] 100 Air conditioner 200 User terminal 310 First server 320 Second server 101, 201, 311, 321 Control unit 102, 202, 312, 322 Storage unit 103, 203, 313, 323 Communication unit PG1, PG2, PG3 Control program MD1, MD2, MD3 Generated AI model RM1, RM2, RM3 Recording medium

Claims

1. A computer program that obtains user information of a user of an air conditioning system, uses a generative AI model that references the obtained user information to perform a dialogue process with the user, determines the control content for the air conditioning system, and causes a computer to execute a process to control the air conditioning system based on the determined control content.

2. The computer program according to claim 1, which causes the computer to perform a process to determine the control content for the air conditioning equipment using the generating AI model which further references environmental information of the space air-conditioned by the air conditioning equipment.

3. The computer program according to claim 2, wherein the environmental information is information about at least one of the internal environment and the external environment of the space that is air-conditioned by the air conditioning equipment.

4. A computer program according to any one of claims 1 to 3, which causes the computer to execute a process to initiate a dialogue process with the user by specifying the user information as reference information.

5. A computer program according to any one of claims 1 to 4, which causes the computer to execute a process to initiate a user interaction process with a generating AI model other than the generating AI model, by specifying the user information as reference information.

6. The computer program according to any one of claims 1 to 5, wherein the dialogue processing includes a question inquiring about the user's actions and the user's response to the question.

7. The computer program according to any one of claims 1 to 6, wherein the dialogue processing includes a question inquiring from the user about the appropriateness of the control content determined using the generated AI model, and the user's answer to the question.

8. A computer program according to any one of claims 1 to 7, which causes the computer to perform a process of having the generated AI model compare the user's time away from home with a threshold, and to decide whether or not to continue operating the air conditioning equipment according to the comparison result.

9. A computer program according to any one of claims 1 to 8, which causes the computer to perform a process to determine whether or not to stop the air conditioning equipment, taking into consideration the energy-saving effect of the air conditioning equipment, based on the generated AI model.

10. A computer program according to any one of claims 1 to 9, which determines, based on the user information, whether the period during which the air conditioning equipment is not used is longer than the set period, and if it determines that the period during which the air conditioning equipment is not used is longer than the set period, causes the computer to execute a process to stop the dialogue process.

11. The computer program according to any one of claims 1 to 10, wherein the user information includes at least one of the user's location information, the user's presence information, schedule information, biometric information, image information, and past interaction information with the user.

12. The computer program according to claim 2, wherein the environmental information includes at least one of temperature, humidity, and weather.

13. A computer program according to any one of claims 1 to 12, which uses the generated AI model to determine control content for an air conditioning-related device other than the air conditioning device, and causes the computer to execute a process to control the air conditioning-related device based on the determined control content.

14. An information processing method which involves acquiring user information of a user of an air conditioning system, performing a dialogue process with the user using a generating AI model that references the acquired user information, determining the control content for the air conditioning system, and having at least one computer perform a process to control the air conditioning system based on the determined control content.

15. An information processing system comprising at least one control unit, wherein the control unit acquires user information of a user of the air conditioning equipment, performs a dialogue process with the user using a generating AI model that references the acquired user information, determines the control content for the air conditioning equipment, and controls the air conditioning equipment based on the determined control content.

Images