KR20200057825A - Air purifier of providing recommendation information corresponding schedule based on artificial intelligence controlling and method implementing thereof - Google Patents

Abstract

The present invention relates to an air purifier providing recommendation information corresponding to schedule based on artificial intelligence and a method thereof. An air purifier according to one embodiment of the present invention includes a control unit which determines the absence of a user and generates recommendation information based on the user′s schedule information, and an interface unit which outputs a voice message or a visual message including the recommendation information.

Description

AIR PURIFIER OF PROVIDING RECOMMENDATION INFORMATION CORRESPONDING SCHEDULE BASED ON ARTIFICIAL INTELLIGENCE CONTROLLING AND METHOD IMPLEMENTING THEREOF}

The present invention is a technology for an air purifier and method for providing recommendation information corresponding to a schedule based on artificial intelligence.

The air purifier refers to a device that performs a purifying function such as removing contaminants from the inhaled air after inhaling the contaminated air in the room and discharging the purified air. In general, an air purifier may select an air purifying function or automatically perform an air purifying function according to a user's control or according to the degree of pollutants distributed in the inhaled air.

On the other hand, in order to enable the air purifier to operate automatically in a variety of ways, the air purifier may check surrounding information and control the operation based on this.

In particular, it is necessary to accurately deal with changes in the external environment or changes in the situation in that the air cleaner is sensitive to changes in the air depending on the external environment. This cannot be dealt with simply by measuring air quality or detecting the human body.

Utility model CN 203893372, filed and registered in China, proposes an air purifier that displays weather using sensors and communication functions that sense fine dust.

However, this has a limitation in that it is impossible to output information suitable for the user's situation in that it outputs external weather information from the air purifier using a predetermined push method.

Accordingly, there is a need for a method in which the air cleaner judges a certain situation more accurately to perform an air cleaning function or provide information to a user.

In this specification, to solve the above-described problems, the air purifier uses an interactive voice recognition function and a calendar interworking function to reflect recommended weather conditions and provide recommendation information suitable for the schedule.

In this specification, the air purifier detects a user's outing situation, checks a schedule, and provides a smart function that provides information suitable for the user's going out in response to a weather condition.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned can be understood by the following description, and will be more clearly understood by embodiments of the present invention. In addition, it will be readily appreciated that the objects and advantages of the present invention can be realized by means of the appended claims and combinations thereof.

The air purifier according to an embodiment of the present invention determines a user's outing situation and generates a recommendation information based on the user's schedule information, and an interface unit for outputting a voice message or a visual message including the recommendation information Includes.

The air purifier according to an embodiment of the present invention further includes a sensing unit that detects changes in light, increase or decrease in dust, or human movement.

The air purifier according to an embodiment of the present invention further includes a learning network that receives sensed sensing information and outputs a result of a user's going out.

The air purifier according to an embodiment of the present invention uses the time information and place information stored in the schedule information to output a voice message to determine the outing situation by determining the outing situation of the control unit and the user predicting the outing time of the user It includes an interface.

The air purifier according to an embodiment of the present invention includes a storage unit for storing a list of outgoing status indicators, a microphone unit for receiving words included in the outgoing status directives list, and a control unit for determining a user's outing status.

According to an embodiment of the present invention, a method of providing recommendation information corresponding to a schedule based on artificial intelligence of an air purifier according to an embodiment of the present invention is the controller predicting a user's outing time by using time information and location information stored in the schedule information And determining the user's outing situation, and outputting a voice message confirming the outing situation by the interface unit.

When implementing an air purifier or a method for controlling the air purifier according to an embodiment of the present invention, the air purifier may provide recommendation information suitable for the schedule by reflecting the weather condition using the interactive voice recognition function and the calendar interlock function.

When implementing an air purifier or a method for controlling the same according to an embodiment of the present invention, the air purifier detects the outing situation of the user, checks the schedule, and provides a smart function that provides information suitable for the outing of the user in response to the weather condition Can be.

The effects of the present invention are not limited to the above-described effects, and those skilled in the art can easily derive various effects of the present invention from the configuration of the present invention.

1 is a view showing the configuration of an air purifier according to an embodiment of the present invention.
2 and 3 are views showing a process in which the air purifier according to an embodiment of the present invention receives certain information by voice.
4 is a view showing a process in which the air cleaner according to an embodiment of the present invention receives certain information from the outside.
5 is a view showing a process of providing recommendation information according to an embodiment of the present invention.
6 and 7 are views illustrating a process of recommending an event by applying an interactive voice recognition function and a calendar interworking function according to an embodiment of the present invention.
8 is a view showing a process of interworking with a calendar service according to an embodiment of the present invention.
9 is a diagram showing the configuration of a learning network according to an embodiment of the present invention.
10 is a view showing a process of increasing the accuracy in determining the situation where the controller goes out according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings so that those skilled in the art to which the present invention pertains can easily practice. The present invention can be implemented in many different forms and is not limited to the embodiments described herein.

In order to clearly describe the present invention, parts irrelevant to the description are omitted, and the same reference numerals are assigned to the same or similar elements throughout the specification. In addition, some embodiments of the present invention will be described in detail with reference to exemplary drawings. In adding reference numerals to the components of each drawing, the same components may have the same reference numerals as possible even though they are displayed on different drawings. In addition, in describing the present invention, when it is determined that detailed descriptions of related well-known structures or functions may obscure the subject matter of the present invention, detailed descriptions thereof may be omitted.

In describing the components of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only for distinguishing the component from other components, and the essence, order, order, or number of the component is not limited by the term. When a component is described as being "connected", "coupled" or "connected" to another component, the component may be directly connected to or connected to the other component, but different components between each component It will be understood that the "intervenes" may be, or each component may be "connected", "coupled" or "connected" through other components.

In addition, in implementing the present invention, components may be subdivided and described for convenience of description, but these components may be implemented in one device or module, or one component may be multiple devices or modules. It can be implemented by being divided into.

In the present specification, the air cleaner may be implemented in various forms and components. The air purifier used in this specification provides an air purifier in addition to an apparatus that provides only an air purifier function, and an air conditioner that provides a cooling or heating function is also an embodiment of the air purifier. Accordingly, devices providing the function of air cleaning and other functions are also included in the air cleaner of the present specification. Various types of air conditioners, heaters, and air cleaners that perform only air cleaning functions are included in the air cleaners referred to herein. For convenience of description, this specification is described under the name of an air cleaner, but the present invention is not limited to the device of this name, and the present invention can be applied to all devices mainly or additionally providing air cleaning.

1 is a view showing the configuration of an air purifier according to an embodiment of the present invention. The cleaning unit 110 is a component that sucks and blows air and performs an air cleaning function. In more detail, the clean part 110 may include an intake part for sucking air, a filter part for removing contaminants from the intake air, and a blower for discharging air from which the contaminants have been removed.

The filter unit may be variously configured such as a chemical filter and a physical filter. For example, the filter unit may include at least one of a deodorizing filter, an antibacterial filter, a High Efficiency Particulate Air (HEPA) filter, an Ultra-Low Penetration Air (ULPA) filter, an activated carbon filter, and an electrostatic precipitator.

The microphone unit 120 receives a user's voice in order to perform interactive voice recognition with the user in a space where the air cleaner 100 is disposed. In addition, various ambient sounds generated in the corresponding space may be input. In addition, by checking these sound patterns (patterns related to the time at which sound is generated or acoustic patterns of sound), the air purifier can infer a user's schedule or situation.

The sensing unit 130 provides a function of sensing ambient light, detecting dust, etc., and detecting movement of objects in the surroundings. When it is detected that an object is actively moving around, it may generate motion information of the object so that the controller 150 determines that the user is preparing to go out.

The controller 150 performs analysis on the input voice and surrounding sounds. The operation of the cleansing unit 110 may be controlled in response to pattern information of the input sound. Alternatively, information necessary for the user may be output. The output method may control the interface unit 170.

The control unit 150 may determine a user's going out and generate recommendation information based on the user's schedule information. The schedule information may be input through the microphone unit 120 or may be stored in the air cleaner 100 by receiving user schedule information from an external device. The schedule information may include time information, place information, and activity information.

The interface unit 170 outputs visual information or auditory information. In one embodiment, the interface unit 170 outputs a visual message including a voice message, a text or an image. The visual information may be output as text, images, or colors from the interface unit 170. The auditory information includes outputting a pre-recorded voice file or a non-voice file such as a melody, or a predetermined message generated by the controller 150 in a text to speech (TTS) method.

The communication unit 160 may receive schedule information, weather information, and the like from the outside. The weather information may be received through a weather information server that is disposed outside and provides various information related to weather and weather. Alternatively, the above information may be received from the air purifier 100 and a separate server previously promised.

The communication unit 160 may receive certain information from a portable device that is used closely by a user, such as an adjacent smartphone or tablet computer. Alternatively, the communication unit 160 may receive the aforementioned schedule information through a separate schedule information server through which a user uploads schedule information.

The storage unit 180 may store weather information, schedule information, and the like applied to recommend a schedule to a user, and may store positive or negative feedback on the recommendation information from the user as a result. Also, the storage unit 180 stores a list of places that can determine the place information included in the schedule information as indoor or outdoor.

The learning network 260 extracts features based on information received or sensed or received or received by the storage unit 180 or received or sensed or input by the microphone unit 120, the sensing unit 130, the communication unit 160, and the like. Based on this, it increases the accuracy of the recommendation schedule. The learning network 260 may increase accuracy based on artificial intelligence such as deep learning or artificial neural networks.

The air purifier 100 shown in FIG. 1 provides an air purifier interactive voice recognition and a calendar interworking function to provide a schedule recommendation function according to a situation such as weather or a schedule. In particular, it is possible to provide recommendation information corresponding to a schedule based on artificial intelligence, because the air cleaner can determine a user's outing situation through artificial intelligence, for example, a learning network to which deep learning is applied.

2 and 3 are views showing a process in which the air purifier according to an embodiment of the present invention receives certain information by voice.

The air purifier 100 may interlock the schedule to provide recommendation information appropriate to the user's schedule. In one embodiment, the sensing unit 130 of the air purifier 100 detects movement (S201). For example, the sensing unit 130 may sense a situation in which the user's movement is more than a certain level or the range of movement is large.

Alternatively, the sensing unit 130 of the air cleaner 100 may sense that the fine dust increases or the temperature increases due to the user's movement. That is, the sensing unit 130 may detect changes in light, increase or decrease in dust, or movement of a person.

When the sensed motion information of the sensing unit 130 is provided to the control unit 150, the control unit 150 determines the activity state by inputting the motion information to the learning network 260 (S202). Alternatively, the controller 150 determines that the motion information is an activity state when the motion information is higher than a certain standard. That is, the learning network 260 may receive sensed sensing information and output a result of the user's going out.

When the user's activity is determined to be a going-out situation as a result of the determination of the control unit 150, the recommended information is required. However, when the control unit 150 determines that the user's activity is a cleaning situation, a cooking situation, or an indoor exercise situation, it is not necessary to provide the user with recommendation information related to a separate schedule.

The outing situation is possible when the user turns off the home appliance, turns off the light, or closes the window. When the sensing unit 130 senses ambient light, or when the noise generated from the household appliance input by the microphone unit 120 decreases, or when the sensing unit 130 senses that the inflow of external air is stopped. It corresponds. In addition, when the sensing unit 130 confirms that the user moves repeatedly between certain spaces (rooms and bathrooms) as a result of sensing a user's moving line with a motion sensor, it also corresponds to a going out situation.

On the other hand, the cleaning situation corresponds to the case where the sensing unit 130 rapidly receives dust generated in the surroundings and the microphone unit 120 receives a cleaner sound. Alternatively, the sensing unit 130 may sense that external air is introduced by opening a window. The cleaning situation may be determined based on the information on the increase in dust, the information on the inflow of external air, or the input sound to the control unit 150 or the learning network 260.

In the cooking situation, when the sensing unit 130 senses the smell or senses a sudden rise in the ambient temperature, when fine dust from cooking increases, or when the microphone unit 120 receives a sound generated during a cooking process Corresponds to A cooking situation may be determined based on a result input to the controller 150 or the learning network 260 based on the sensed or input sound.

Therefore, the control unit 150 determines the activity status based on the information received by the sensing unit 130 or the microphone unit (S202), and as a result, when recommendation information is required (S203), that is, the user If the activity of the outing situation, the interface unit 170 of the air purifier 100 is "in response to a situation in which recommended information is required

A message for requesting schedule information necessary for generating recommendation information to the user is output by voice (S204). For example, the interface unit 170 "Are you going out?" Outputs a voice message. If the user says "yes", this is input to the microphone unit 120. And the interface unit 170 "What is the schedule?" Outputs a voice message.

Thereafter, the microphone unit 120 receives schedule information from the user through voice (S205). At this time, the user may utter any one or more of time, place, or activity information by voice, and the control unit 150 may repeat steps S204 and S205 until all information necessary to generate recommendation information is input. .

As shown in FIG. 2, the air purifier 100 determines a user's movement situation, a change in ambient air quality, a change in illuminance, or a noise change in a household appliance, and inputs certain information from the user when the user's movement is in an outing situation Can receive

3 is a view showing a process in which air cleaners receive schedule information by voice in response to a user's command according to an embodiment of the present invention. The user utters the start word and command (S211). The controller 150 determines whether the recommendation information is required in the command (S212). For example, if the user says, "High LG! Turn on the air purifier", the air purifier is operated without additional recommendation information.

On the other hand, if the user says, "High LG! Try to walk", the controller 150 may determine that the user is out. Accordingly, the control unit 150 and the interface unit 170 output a message requesting schedule information necessary for generating recommendation information to the user in voice in response to a situation in which recommendation information is required (S213). For example, "Please tell me the time and place". Then, the schedule information is received by the user through voice (S214). For example, it is possible to obtain information about the time and place and combine it with activity information called "walking" previously obtained to generate recommendation information.

In another embodiment, when the user inputs "walking", the control unit 150 in the air purifier 100 uses the learning network 260 to calculate information about the time and place that the user has previously described in response to the "walking" can do. When the place is around the residence and the time is stored as less than 1 hour, the controller 150 may automatically generate schedule information based on the corresponding place and time, output it, and provide recommendation information for the location.

2 and 3 show a process in which the air cleaner checks schedule information from a user. Schedule information is information that combines spatiotemporal information about the user's activity. The air cleaner 100 may generate recommendation information to be provided to the user in response to this.

In one embodiment, the recommendation information includes recommending a schedule change or other schedule to the user based on weather information such as weather or air condition, or recommending to wear clothes or a mask or to prepare an umbrella. When applying FIGS. 2 and 3, the air purifier obtains schedule information through interactive speech recognition and obtains weather information about the weather to recommend information suitable for the schedule.

4 is a view showing a process in which the air cleaner according to an embodiment of the present invention receives certain information from the outside.

The communication unit 160 receives the schedule information from the device storing the schedule information (S221). An embodiment of a device that stores schedule information includes a server that stores schedule information for each individual in a server, such as Google Calendar. Alternatively, the communication unit 160 may directly receive the schedule information stored in the mobile phone.

In addition, the communication unit 160 may receive and store schedule information stored in a server or stored in a mobile phone / smartphone based on various communication protocols. In this case, the schedules of a plurality of members of the space where the air cleaners 100 are arranged are respectively stored, but the air cleaners 100 can identify the schedule information by classifying users.

The schedule information received by the communication unit 160 is stored in the storage unit 180. When only one user is registered, only one person's schedule information is stored. When multiple users are registered, schedule information of multiple people may be stored.

Thereafter, the controller 150 checks whether recommendation information is required (S222). For example, the sensing unit 130 or the microphone unit 120 senses the user's movement or the user's voice is generated in the surroundings, and the control unit 150 confirms that there is an external schedule related to the user. .

As a result of the check, if the user is currently preparing to go out and there is specific schedule information to the user, the controller 150 outputs the schedule information to the user in voice and a message requesting confirmation from the user in voice in response to a situation in which recommendation information is required. (S223). For example, the control unit 150 "Do you attend a soccer game held in Gasan-dong today?" Outputs a voice message.

Then, the control unit 150 receives confirmation of schedule information from the user by voice (S224). For example, as an example, a user is "yes" to the output message of S223 and utters confirmation of schedule information by voice.

FIG. 4 is a process in which the air purifier receives separate schedule information from an external server or communication device such as a smartphone and outputs schedule information to the user when it is determined that the user is out, and checks schedule information through user confirmation. Show. That is, the air purifier obtains schedule information through the interlocking of the calendar and obtains weather information about the weather to recommend information suitable for the schedule.

2 to 4 are processes in which the air purifier 100 checks schedule information based on interaction with a user or information input from a user or input from another device. The identified schedule information is applied to the air purifier 100 to generate suitable recommendation information.

5 is a view showing a process of providing recommendation information according to an embodiment of the present invention.

The air purifier checks the schedule information of the user based on the embodiment of FIGS. 2 to 4 and the schedule is linked to the air cleaner (S231).

The controller 150 determines whether the weather condition is a good or a bad condition based on the weather information received through the communication unit 150 (S232). At this time, when the place stored in the schedule information is different from the current location, the controller 150 may check weather information for the place on the schedule or the places on the moving line from the current location to the place on the schedule.

Good weather conditions mean the distribution of fine dust in the air, temperature, and no rain or clear conditions. Poor weather conditions mean a state in which it is raining, has a lot of fine dust, or the temperature is rapidly changing.

As a result, if the weather condition is good for the control unit 150, and it is determined that the schedule information may be maintained as a result of comparing the current weather information with the schedule information, the control unit 150 and the interface unit 170 use the existing schedule information. The recommendation information is output as being maintained (S233). In one embodiment, the interface unit 170 may output a guide message that "the weather is good today, so you may proceed with the schedule as it is."

Alternatively, the interface unit 170 may output weather information as a guide message. "Today's weather and humidity are as good as yesterday at 15 degrees and 35%." Message.

On the other hand, if the weather condition is bad, the activity stored in the schedule information is classified to check whether it is indoor or outdoor (S235). For indoor activities and outdoor activities, mapping tables shown in Table 1 may be used.

Table 1 is a table classifying indoor activities and outdoor activities in response to activity or place information among schedule information held by the air cleaner 100. The storage unit 180 stores the information shown in Table 1, so that the controller 150 can determine whether the user activity of the schedule information is an indoor activity or an outdoor activity.

No. activity Activity classification One library inside 2 hospital 3 church 4 conference 5 swimming pool 6 Hair shop 7 Dining together 8 Night Shift 9 health 10 Wedding attendance 11 walk Outdoor 12 jogging 13 Badminton 14 Soccer 15 basketball 16 baseball 17 Amusement Park 18 picnic 19 Waterpark 20 Travel

In S235, the controller 150 may determine the user's activity as an indoor activity or an outdoor activity. As a result of the determination, if it is an indoor activity, the controller 150 may generate recommendation information by maintaining the existing schedule even if the weather condition is bad (S236).

In addition to this, it is possible to recommend wearing a mask or taking an umbrella in response to weather information in order to respond to bad weather during the movement process.

On the other hand, as a result of the determination in S325, if the outdoor activity, the controller 150 may generate recommendation information by changing the existing schedule because the weather condition is bad (S237).

The recommendation information generated in S233, S236, and S237 may be output as a voice message or as a visual message. In particular, in the case of S236, the interface unit 170 of the air cleaner 100 may output a specific image to notify the user of "wearing a mask" or "preparing an umbrella". The mask image can be output or the image can be output first. Or you can output a message via sound.

In summary, if the place information included in the schedule information corresponds to the outdoors, or if the location information and the current location of the place information are greater than a preset distance, the controller 150 determines that the user equipment is exposed outdoors. In addition, when the weather information indicates weather deterioration, the control unit 150 generates recommendation information corresponding to the activity time or movement time of the schedule, and outputs the recommended information.

In particular, when the concentration of fine dust is very high or it is expected to rain a lot, the possibility of weather deterioration or the degree of weather deterioration is large, the interface unit 170 may output a large voice message or a strong light visual message to the user. . For example, when the controller 150 predicts a user's outdoor movement or outdoor activity, the interface unit 170 may output a mask image or an umbrella image.

That is, the control unit 150 may control the interface unit 170 to output recommended information based on the density information, the likelihood probability information, and the temperature information of the fine dust in the weather information.

6 and 7 are views showing a process of recommending an event by applying an interactive voice recognition function and a calendar interworking function according to an embodiment of the present invention.

6 is a process of linking time, place, and activity with a schedule through a conversation according to an embodiment of the present invention. When the user 5 utters "Hi LG" (S240), the air purifier 100 enters a command waiting state (S241). The user 5 requests the recommendation information again, "I want to play soccer, how is the weather?" (S242). The air purifier 100 confirms that the inputted information is "football" and that the activity information is input, and notifies the unconfirmed time and place in providing recommendation information (S243).

As a result, the air purifier 100 outputs a voice message saying "Please tell me the time and place" (S244), and the user 5 inputs "additional digital soccer field at 5 PM today" (S245). Thereafter, the air purifier 100 generates recommendation information based on the input time, place, and activity information (S246). At this time, external weather information may be reflected. As a result, the air purifier 100 outputs, "The weather is clear and fine dust is good. Have fun playing soccer" (S247).

6 shows a process in which the air purifier outputs recommendation information that the input schedule may be maintained as it is. If the weather condition deteriorates, the air purifier 100 may provide recommendation information for suggesting cancellation of the schedule based on other schedule information of the user.

7 is a process of linking time, place, and activity with a schedule through dialogue according to an embodiment of the present invention. When the user 5 utters "Hi LG" (S250), the air purifier 100 enters a command waiting state (S251). The user 5 requests the recommendation information again, "What is the weather, the weather this afternoon?" (S252). The air purifier 100 confirms that the inputted information is time and place information, and notifies the unconfirmed activity information in providing recommendation information (S253).

As a result, the air purifier 100 outputs a voice message, "What are you doing?" (S254), and the user 5 inputs "Soccer" (S255). Thereafter, the air purifier 100 generates recommendation information based on the input time, place, and activity information (S256). At this time, external weather information may be reflected. As a result, the air purifier 100 outputs, "The fine dust condition is very bad. Why don't you try other activities?" (S257).

7 shows a process in which the air purifier outputs recommendation information that the input schedule needs to be changed based on weather information. If the weather condition is in good condition, the air purifier 100 may provide recommendation information that proceeds with the user's schedule.

5 to 7 show a process in which the control unit 150 extracts any one or more of time information, activity information, and place information using a command received by the microphone unit and selects uninputed information. In addition, the control unit 150 controls the interface unit 170 to output a message requesting input of unentered information.

8 is a view showing a process of interworking with a calendar service according to an embodiment of the present invention. Various types of web services or mobile services provide calendar programs or calendar services.

FIG. 8 shows a screen of a mobile device, and as shown in FIG. 8, a user may store schedule information using the mobile device. For example, the activity information 11, the place information 12, and the time information 13 may constitute one piece of schedule information, and this information is stored or updated in the air purifier 100.

9 is a diagram showing the configuration of a learning network according to an embodiment of the present invention. The control unit 150 may further include a learning network 260 as illustrated in FIG. 9 in order for the control unit to check the user's outing situation based on artificial intelligence. Sound input through the microphone unit 120 or information sensed by the sensing unit 130 is input to an input node of the deep learning learning network constructed through learning. The inputted information is feature data.

In one embodiment, the sound may be input through the microphone unit 120. The microphone unit 120 may feature data such as sounds or voices that the user can generate when going out.

In addition, the information sensed by the sensing unit 130, for example, the user's movement speed, the number of movements around the air cleaner, or the time to stay may be characteristic data.

In addition, the sensing unit 130 may detect a change in light and an increase or decrease in dust due to lighting on / off. This can be estimated that the movement of the user is busy when the dust increases rapidly. This may mean an exercise situation, an activity situation, or an outing situation in preparation for going out. These sensed values may also be feature data.

The learning network includes an input layer composed of input nodes into which various K feature data described above are inputted, and an output layer composed of output nodes outputting whether or not it is in an outgoing situation in response to the inputted feature data. (Output) and one or more M hidden layers disposed between the input layer and the output layer.

The characteristic data may be various sounds, dust change, light change, or pre-inputted schedule information that occurs just before the going out.

Or, the feature data includes the current time, the range or frequency of movement, and the time accumulated in the past.

The learning network 260 provides examples of feature data and values of output nodes corresponding to the learning data to change the set values of nodes and edges constituting the learning network 260.

Weights are set at the edges connecting the nodes of the layers, and the presence or absence of the weights or edges can be added, removed, or updated in the learning process. Accordingly, the weights of nodes and edges disposed between k input nodes and i output nodes may be updated by a learning process.

When the weights of nodes and edges are determined through these learnings, when feature data is subsequently given to the input node, the output node calculates a corresponding result, that is, whether the current situation is an outing situation.

Before the learning network 260 performs learning, all nodes and edges may be set as initial values. However, when information for learning is accumulated and the weights of nodes and edges are changed, matching between feature data and outing situation information may be performed in this process.

The learning network 260 of FIG. 9 may be installed in the air purifier 100 after being learned in a separate cloud server. Alternatively, the learning network 260 is disposed on the cloud server as it is, and the air purifier 100 may transmit feature data to the server and the server may transmit pattern information for the air purifier 100.

When learning using a cloud server, the learning network 260 can receive a large number of feature data, so the learning unit 260 can perform learning based on vast data.

10 is a view showing a process of increasing the accuracy in determining the situation where the controller goes out according to an embodiment of the present invention.

The sensing unit 130 detects the user's movement (S271). Then, the control unit 150 checks the schedule information of the user stored in various paths (S272). As a result, the control unit predicts a user's going out time using the time information and the place information stored in the schedule information (S272). For example, if the user's schedule information includes place information 1 hour away from the current location and time information after 2 hours, it is highly likely that the user is going out.

Of course, this may be determined based on the time required to go out after the user's activity accumulated in the controller 150. The user's outing is applied when the signal of the user's mobile phone is not sensed, the user's voice or movement is not sensed, or the user is sensed to close the door.

In addition, the control unit 150 may calculate an average time to prepare for going out by calculating a time difference between the time when the user actively starts moving in the repeated outing situation of the user.

Accordingly, the controller 150 determines the outing situation of the user by predicting the outing time of the user by reflecting the above-described average outing preparation time, time information, and place information (S274).

In this case, under the control of the control unit 150 of the air cleaner 100, the interface unit 170 outputs a message inquiring to the user whether the schedule information and the preparation for going out are in progress (S275). For example, the interface unit 170 may selectively include schedule information such as "Are you ready to go out to the library at 7 o'clock?" And output a voice message confirming the outing situation. Thereafter, when a user inputs a predetermined voice, the controller 150 may provide recommendation information using schedule information and weather information correspondingly.

On the other hand, in order to more clearly determine the user's outing situation, the storage unit 180 stores a list of outing situation directives. "I'm leaving now.", "When will you come?", "Hi", "You need to prepare", "What time do you get out?" As described above, the instruction that the user speaks when going out or before going out is stored in the storage unit 180. The directive list can be updated repeatedly.

Thereafter, when the microphone unit 120 receives a word included in the outgoing situation directive list, the controller 150 may determine that the outing situation is present.

Embodiments of the air purifier and its operation method described herein provide recommendation information suitable for a user's schedule by obtaining user's schedule information and using weather information at a time point corresponding to the schedule. Here, the method of acquiring the schedule information may include a method of detecting a user's going out and interactively recognizing voice, a method of receiving the user's schedule information from a server or a portable device, and interlocking the schedule information.

The schedule information includes time, place, and activity information, and the air purifier can determine indoor or outdoor air based on the activity information or the location information. As an example, Table 1 described above may be applied. Examples of the recommendation information include recommendation information that compares weather conditions and places, activities, and time in the schedule information to maintain the existing schedule, recommendation information that maintains the existing schedule, and suggests wearing of belongings corresponding to weather changes, or It consists of recommendation information, etc., which suggests changing the schedule. As an embodiment of wearing the above-mentioned belongings, it may be wearing a mask, holding an umbrella, or holding an outer garment.

When it is recommended to maintain the existing schedule, the weather condition is good or the user's schedule is indoors. Alternatively, even if the weather information is confirmed as raining, the activity of "sea swimming" is not significantly affected, and thus, "sea swimming" may be maintained, but recommendation information for notifying that it is raining may be provided.

On the other hand, in case of bad weather conditions, if the schedule information indicates indoor activities, the existing schedule is maintained, but wearing of items such as an umbrella or a mask is suggested. If the schedule information is for outdoor activities, it is recommended to change the schedule if the weather conditions are bad.

When the embodiment of the present invention is applied, a smart recommendation function may be provided by comparing weather information and schedule information using a voice recognition air purifier.

In an embodiment of the present invention, the control unit 150 provides a voice recognition function. The control unit 150 may analyze the voice data of the user input through the microphone unit 120 and also infer an indirect situation through context understanding. For example, when a user inquires about a weather condition for a specific area, the air cleaner 100 may provide weather information to the user and inquire about the outing situation.

For example, in the flow as in the embodiment of FIG. 6, when the user utters "Hi-LG, 00 area weather, fine dust", the air purifier 100 displays "00 area weather xxx, fine dust xxx." You can respond by providing information.

As a result, when the user responds "Yes", the air purifier 100 may output recommendation information such as "Take an umbrella or take a mask."

In addition, the controller 150 accumulates and stores the time when the user goes out by day, hour, and season, learns a pattern for the time to go out, and predicts the user's outing situation more accurately.

Although all components constituting the embodiments of the present invention are described as being combined or operated as one, the present invention is not necessarily limited to these embodiments, and all elements within the scope of the present invention are one or more. It can also be operated by selectively combining. In addition, although all of the components may be implemented by one independent hardware, a part or all of the components are selectively combined to perform a part or all of functions combined in one or a plurality of hardware. It may be implemented as a computer program having a. The codes and code segments constituting the computer program can be easily deduced by those skilled in the art of the present invention. Such a computer program is stored in a computer readable storage medium (Computer Readable Media) to be read and executed by a computer, thereby implementing an embodiment of the present invention. The storage medium of the computer program includes a magnetic recording medium, an optical recording medium, and a storage medium including a semiconductor recording element. In addition, the computer program implementing the embodiment of the present invention includes a program module that is transmitted in real time through an external device.

In the above, the description has been mainly focused on the embodiment of the present invention, but various changes or modifications can be made at the level of a person skilled in the art. Accordingly, it will be understood that such changes and modifications are included within the scope of the present invention without departing from the scope of the present invention.

100: air purifier 110: clean part
150: control unit 260: learning network

Claims (14)

A cleaning unit that sucks and blows air and performs an air cleaning function;
A microphone unit that receives ambient sound;
A sensing unit that senses movement; And
A control unit for determining a user's outing situation and generating recommendation information based on the user's schedule information;
A storage unit that stores a list of places capable of determining indoor or outdoor location information included in the schedule information; And
And an interface unit for outputting a voice message or a visual message including the recommendation information,
The schedule information includes time information, place information, and activity information, and an air purifier providing recommendation information corresponding to the schedule based on artificial intelligence.
According to claim 1,
The sensing unit senses a change in light, increase or decrease in dust, or movement of a person,
An air purifier providing recommendation information corresponding to a schedule based on artificial intelligence, further comprising a learning network that receives the sensed sensing information and outputs a result of a user's going out situation.
According to claim 1,
An air cleaner providing recommendation information corresponding to a schedule based on artificial intelligence, further comprising a communication unit that receives any one or more of the schedule information or weather information from a schedule information server, a weather information server, or a portable device.
According to claim 1,
The control unit predicts the outing time of the user by using time information and place information stored in the schedule information,
The control unit controls the interface unit to output a voice message confirming the outing situation by determining the outing situation of the user, an air purifier providing recommendation information corresponding to a schedule based on artificial intelligence.
According to claim 1,
When the place information included in the schedule information corresponds to the outdoors, or when the location and the current location of the place information is greater than or equal to a preset distance, the controller controls the activity time or movement of the schedule when the weather information indicates bad weather. An air purifier that provides recommendation information corresponding to a schedule based on artificial intelligence, which generates recommendation information in response to time.
According to claim 1,
The control unit extracts any one or more of the time information, activity information, and location information by using the command received by the microphone unit and selects uninputed information,
The control unit controls the interface unit to output a message requesting input of the uninputted information, and an air purifier providing recommendation information corresponding to a schedule based on artificial intelligence.
According to claim 1,
The storage unit stores a list of outgoing situation directives,
When the microphone unit receives a word included in the outing condition directive list, the control unit determines the outing condition, and provides an air purifier providing recommendation information corresponding to a schedule based on artificial intelligence.
Sensing the user's movement;
The control unit determines a going out condition based on the detected motion information and generates recommendation information based on the schedule information of the user;
And outputting the generated recommendation information as a voice message or a visual message, and the schedule information includes time information, place information and activity information, and the air purifier recommends corresponding to the schedule based on artificial intelligence. How to provide information.
The method of claim 8,
The sensing unit sensing a change in light, increase or decrease in dust, or movement of a person;
Receiving, by the control unit, the sensed sensing information and inputting it to a learning network;
A method of providing a recommendation information corresponding to a schedule based on artificial intelligence, comprising the step of outputting a result of the outing situation of the user where the learning network is output.
The method of claim 8,
The communication unit further comprises receiving any one or more of the schedule information or the weather information from any one or more of a schedule information server, a weather information server, or a portable device, and the air purifier recommends corresponding to the schedule based on artificial intelligence. How to provide information.
The method of claim 8,
Determining, by the controller, the outing situation of the user by predicting the outing time of the user by using time information and place information stored in the schedule information; And
The interface unit further comprising the step of outputting a voice message confirming the outing, air purifier to provide recommendation information corresponding to the schedule based on artificial intelligence.
The method of claim 8,
When the place information included in the schedule information corresponds to the outdoors, or when the location and the current location of the place information is greater than or equal to a preset distance, when the controller instructs the weather information to deteriorate, the activity time or movement of the schedule A method of providing recommendation information corresponding to a schedule based on artificial intelligence, further comprising generating the recommendation information in response to time.
The method of claim 8,
The control unit extracting any one or more of the time information, activity information, and location information by using the command received by the microphone unit to select un-entered information; And
The control unit further comprising the step of controlling the interface unit to output a message requesting the input of the uninputted information, the air purifier provides a recommendation information corresponding to a schedule based on artificial intelligence.
The method of claim 8,
When the microphone unit receives a word included in a list of outgoing conditions stored in the storage unit, the controller further includes determining that the user is outgoing, the air purifier recommends corresponding to a schedule based on artificial intelligence. How to provide information.

Images