CN104579873A - Method and system for controlling intelligent home equipment - Google Patents
Method and system for controlling intelligent home equipment Download PDFInfo
- Publication number
- CN104579873A CN104579873A CN201510040534.1A CN201510040534A CN104579873A CN 104579873 A CN104579873 A CN 104579873A CN 201510040534 A CN201510040534 A CN 201510040534A CN 104579873 A CN104579873 A CN 104579873A
- Authority
- CN
- China
- Prior art keywords
- control
- mapping
- self
- machine learning
- conclusion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Selective Calling Equipment (AREA)
Abstract
The invention discloses a method and system for controlling intelligent home equipment. The method for controlling the intelligent home equipment comprises the steps that emergency control information input by a user is received by a user side, and the emergency control information and a family mark are sent to a server; the emergency control information is analyzed by the server, so that a user-defined control map containing conditions and a conclusion is obtained, and the user-defined control map is synchronized to a controller corresponding to the family mark; machine learning is conducted on historical control information of the intelligent home equipment by the controller, so that a machine learning control map containing conditions and a conclusion is obtained; the user-defined control map from the server is received, an equipment control command is confirmed based on the machine learning control map and the user-defined control map, and the intelligent home equipment is controlled according to the equipment control command. By the adoption of the method and system for controlling the intelligent home equipment, the intelligent home equipment can be controlled more intelligently.
Description
Technical field
The present invention relates to intelligent control technology, particularly relate to the method and system that intelligent home device is controlled.
Background technology
Current Smart Home field, normal employing technology of Internet of things by the various intelligent home devices in certain limit (as audio & video equipment, illuminator, curtain equipment, air-conditioning equipment, safety-protection system, Digital Theater System, network home appliance and three shows to make a copy for system etc.) connect together, by intelligent home control device, each intelligent home device is controlled, comprise and home wiring control is provided, Lighting control, curtain controls, remote control using telephone, indoor and outdoor remote control, burglar alarm, environmental monitoring, HVAC controls, several functions and the means such as infrared forwarding and programmable Timer control.Described certain limit is such as home-ranges, office-wide etc.
Compared with common home equipment, intelligent home device not only has traditional inhabitation function, also have both network service, information household appliances, equipment automatization, efficient, comfortable, safety, the facility that also provide collecting system, structure, service, management to be integrated, the living environment of environmental protection, can carry out omnibearing information interaction.Just can be controlled each intelligent home device by the controller of Smart Home, easy to operate, optimize the life style of people, help people effectively to arrange the time, strengthen the fail safe of life staying idle at home.
Can how effectively integrate, manage these intelligent home devices as a whole, be a core that realize Smart Home, and this core has expedited the emergence of the birth of a lot of Intelligent housing scheme just.
The automated execution aspect that the intellectuality of Intelligent housing scheme of today is mainly reflected in.Usual way is the record by user operation behavior, is found behavioural habits and the pattern of user, thus carry out the operation of Automated condtrol intelligent home device according to user model by machine learning method.The method meeting recording user, at every turn to the operation information of intelligent home device, then calls ARM learning controller to learn the behavior pattern of user, finds the behavioural habits of user; Thus in running afterwards, can according to these study to behavior pattern automatically perform corresponding actions, the fixed mode operation that liberation user is loaded down with trivial details.
To sum up, existing scheme can only carry out automatic management based on operation behavior record to intelligent home device, but cannot process burst demand or the abnormal conditions of user, cause intelligent performance not high.
Summary of the invention
The invention provides a kind of method controlled intelligent home device, the method can make the control of intelligent home device more intelligent.
The invention provides a kind of system controlled intelligent home device, this system can make the control of intelligent home device more intelligent.
To the method that intelligent home device controls, the method comprises:
User side receives the burst control information of user's input, and burst control information and household identification are sent to server;
Server is resolved burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification;
Controller carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; And the self-defined control received from server maps, determine equipment control command based on machine learning control mapping and self-defined control mapping, according to equipment control command, intelligent home device is controlled.
To the system that intelligent home device controls, this system comprises user side, server and controller;
Described user side, receives the burst control information of user's input, burst control information and household identification is sent to described server;
Described server, receives the burst control information from described client and household identification, resolves burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification;
Described controller, carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; The self-defined control received from server maps, and determines equipment control command, control according to equipment control command to intelligent home device based on machine learning control mapping and self-defined control mapping.
As can be seen from such scheme, in the present invention, user side receives the burst control information of user's input, and burst control information and household identification are sent to server; Server is resolved burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification; Controller carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; The self-defined control received from server maps, and determines equipment control command, control according to equipment control command to intelligent home device based on machine learning control mapping and self-defined control mapping.The present invention obtains the burst control information of user's input, on the basis considering history control information, controls intelligent home device in conjunction with burst control information, thus, realize the control of intelligent home device more flexible, intelligent.
Accompanying drawing explanation
Fig. 1 is the method indicative flowchart that the present invention controls intelligent home device;
Fig. 2 is the method flow diagram example that the present invention inputs burst control and resolves;
Fig. 3 is the method flow diagram example that controller of the present invention carries out intelligent home device control;
Fig. 4 is the system configuration schematic diagram that the present invention controls intelligent home device.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with embodiment and accompanying drawing, the present invention is described in more detail.
The present invention obtains the burst control information of user's input, on the basis considering history control information, controls intelligent home device in conjunction with burst control information, thus, realize the control of intelligent home device more flexible, intelligent.See Fig. 1, be the method indicative flowchart that the present invention controls intelligent home device, it comprises the following steps:
Step 101, user side receives the burst control information of user's input, and burst control information and household identification are sent to server.
Burst control information is some burst demand or abnormal conditions, such as, go on business for bright latter two days (1.22-1.23).Each domestic installation one family mark, user inputs the burst control information of corresponding household identification.
During specific implementation, according to the template input burst control information pre-set, such as, can distinguish the event of input time and correspondence.
The mode that user inputs burst control information does not limit, and comprises text message, voice messaging etc.Burst control information is allowed to the situation of various ways, following specific implementation can be adopted: user side judges the burst control information received, if be text message, then directly send to server, if be not text message, then burst control information is converted to text message, sends to server.
Step 102, server is resolved burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification.
Self-defined control mapping comprises condition and conclusion, that is to say, when satisfying condition, conclusion is set up.Be described with bright rear example of going on business for two days, its condition is date 1.22-1.23, and conclusion is for closing parlor lamp.
Step 103, controller carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; And the self-defined control received from server maps, determine equipment control command based on machine learning control mapping and self-defined control mapping, according to equipment control command, intelligent home device is controlled.
Further, before according to equipment control command intelligent home device being controlled, also equipment control command can be fed back to user and confirm, particularly:
Equipment control command is showed user, if the mandate received from user indicates, then performs described step intelligent home device controlled according to equipment control command.If the mandate do not received from user indicates, can perform according to the mode of acquiescence; Default behavior can be arranged voluntarily, can be set to actuating equipment control command, also can be set to refusal actuating equipment control command.After receiving the mandate instruction from user, can also preserving it, carrying out subsequent operation for directly transferring during next actuating equipment control command.
The present invention is based on machine learning control map and self-defined control mapping determine equipment control command, like this, on the basis considering history control information, in conjunction with burst control information, intelligent home device is controlled, make the control of intelligent home device more flexible, intelligent.
Map based on machine learning control mapping and self-defined control the concrete scheme determining equipment control command, can arrange as required, citing is below described.
Scheme one,
Controller obtains environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, and generates equipment control command.When environment sensing information meets the condition in machine learning control mapping and self-defined control mapping, in mapping using machine learning control, corresponding conclusion is as checking item, this conclusion is verified, judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, conclusion then as checking item is set up, and generates the equipment control command about this conclusion.The project of environment sensing information can be arranged voluntarily, such as, be time, brightness value etc.
Scheme two,
Obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, and generates equipment control command, sends to described actuator; If inconsistent, then compare machine learning and control mapping and the self-defined confidence level controlling to map, determine that the control that confidence level is large maps, extract the conclusion in the control mapping determined, generate the equipment control command about extracted conclusion.
Scheme two controls mapping for machine learning and self-defined control mapping arranges confidence level respectively, meet machine learning in environment sensing information to control to map and the self-defined condition controlled in mapping, but when the conclusion that machine learning controls to map and self-defined control maps is inconsistent, choose conclusion according to confidence level, generate control command.
Scheme one and scheme two are two better implementations, are not limited to this two schemes.Such as, for scheme two, can also according to confidence level choose in advance machine learning control to map and self-defined control to map in control map, then combining environmental perception information and the control chosen map and judge whether conclusion is set up, generation equipment control command during establishment again.
In order to improve precise control, further, controller regularly detects self-defined control mapping, deletes the self-defined control of losing efficacy and maps.Such as arranging the periodic scanning cycle is 7 days, then the self-defined condition controlling to map is judged, if free relevant setting, such as " 12.10-12.12-> closes parlor lamp ", but time when scanning has been the time after 12.12, then represent that this mapping cannot enable again, then the mapping that this condition is expired needs to delete.。
Below by Fig. 2,3 flow process, the present invention program is illustrated.
See Fig. 2, the method flow diagram example for the present invention burst control being inputted and resolve, the method comprises the following steps:
Step 201, user side receives the burst control information of user's input, if the burst control information of input is voice messaging, is then converted to text message, performs step 202; If the burst control information of input is text message, then directly perform step 202.
User side provides the software operation page to user, user can directly control the equipment in family on software, also abnormal inputting interface can be switched to, the abnormal demand of input user, can be phonetic entry or Text Input, then be resolved to server by the input information transmission of network by user.
Step 202, burst control information and household identification are sent to server by user side.
User side can be the equipment that mobile phone, computer etc. can carry out network connection, and user side and server carry out network and be connected, and logs in Intelligent housing webpage, according to webpage prompting input burst control information and corresponding household identification.
Step 203, server is resolved burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps.
This step can adopt the interpretive model of sets itself to resolve burst control information, and the self-defined control obtaining corresponding burst control information maps.This step can adopt the scheme of 2013-2033 to realize:
Step 2031, server is analyzed burst control information, extracts Time And Event wherein.
This step carries out natural language analysis to burst control information, extraction time and time.Such as burst control information is that bright latter two days (1.22-1.23) goes on business, then the time of extracting is 1.22-1.23, and event is for going on business.
Step 2032, server inquires corresponding control overflow according to event in commonsense knowledge base.
Commonsense knowledge base stores general knowledge, comprises the control overflow of corresponding event.To go on business, define the event of going on business in commonsense knowledge base, go on business and represent that people stays out, need to close controlled intelligent home device, then corresponding control overflow is for closing controlled plant.
Step 2033, the intelligent home device information corresponding with corresponding household identification is inquired from customer equipment storehouse, combined with intelligent home equipment information, time and control overflow, the self-defined control generating corresponding intelligent home device maps, and stores self-defined control and map in rules repository.
The customer equipment library storage intelligent home device information of corresponding each household identification; Be that each intelligent home device generates self-defined controls and maps based on the intelligent home device information inquired, if there is multiple intelligent home device, then generate many self-defined controls mappings.In this example, the time of hypothesis is 1.22-1.23, and control overflow is for closing, then the self-defined control generated is mapped as " time period-> closes intelligent home device N "; Time period is wherein also condition, and when the time meets 1.22-1.23, conclusion " is closed intelligent home device N " and set up.
Self-defined control maps and also can be described as custom rule.
Step 204, self-defined control maps and is synchronized to the controller corresponding with household identification by server.
In the family, by controller, intelligent home device is controlled.Self-defined control mapping is synchronized to corresponding controller by network by server.
See Fig. 3, for controller of the present invention carries out the method flow diagram example of intelligent home device control, the method comprises the following steps:
Step 301, controller carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map.
Controller have recorded the history control information of user operation behavior, is found behavioural habits and the pattern of user by machine learning method, obtains machine learning and controls to map.Machine learning is carried out to history control information, obtains machine learning mapping relations, be prior art, seldom repeat here.
Machine learning can periodically perform.
Machine learning is controlled mapping to be stored in family's mapping thesaurus.
Step 302, the self-defined control that controller receives from server maps, and carries out inference engine, determine equipment control command based on machine learning control mapping and self-defined control mapping.
This step can specifically comprise: controller obtains environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, then generate the equipment control command controlling the conclusion mapped about machine learning; If inconsistent, then conclusion is false.
Suppose have one to be mapped as in the mapping that machine learning obtains: " the if time is 7 points, and parlor brightness <th, and then turns on parlor lamp "; Self-defined control is mapped as " the if time is that 2014.9.9-2014.9.10, then need not turn on parlor lamp ".When carrying out calculating based on such mapping, environment sensing information comprises time and ambient brightness; When environment sensing information meets (time of meeting is at 7 and parlor brightness <th) when Article 1 maps, the conclusion needing to turn on parlor lamp will be drawn, when but if the time also meets Article 2 mapping simultaneously, the conclusion need not turning on parlor lamp can be drawn again, now conflict, inference engine cannot provide conclusion, also just causes intelligent system to occur abnormal, cannot normally work.In order to address this is that, this example introduces general knowledge skill of deduction and calculation, and the method that general knowledge calculates has a lot, and a kind of feasible method uses Default Logic, basic introduction is carried out to Default Logic below, with the description of visualization more feature of the present invention.
Shown in the following formula of general principle of Default Logic:
(α(x):β(x))/γ(x)
Set up if the understanding of this formula is α (x), and β (x) can meet conforming words under the present circumstances, then conclusion γ (x) sets up; Namely, environment sensing information all meets machine learning and controls to map and self-defined condition (i.e. α (x) establishment controlled in mapping, the α (x) of respective mapping is different under normal circumstances), and machine learning controls the conclusion checking item in mapping, i.e. β (x), consistent with the self-defined conclusion controlling to map, then the conclusion that machine learning controls to map is set up (namely γ (x) sets up).That is, on the basis that routine calculates, before reaching a conclusion, need first to do consistency checking.
Corresponding Default Logic, there is a Default Theory<D, but what W>, W represented is stable incomplete World Affairs, the expansion W of D representative and the rational rule that defines, but these rules are defeasible.Correspond in situation of the present invention, what W was corresponding is environment sensing information, and self-defined control maps; What D was corresponding is that the machine learning obtained by machine learning controls to map.Machine learning in D controls to map all to be needed to be updated to the form meeting above-mentioned formula, and a simple method is exactly that β (x) is identical with γ (x), and namely " A/B " is updated to " A:B/B "; A is the condition in mapping, and first B in " A:B/B " is the conclusion checking item in mapping, and second B in " A:B/B " is conclusion.When inference engine carries out reasoning, if meet the condition A in mapping based on environment sensing information, and when information is consistent in conclusion checking item B and W in A:B in mapping, then can release B establishment.As can be seen from this explanation, the release of conclusion and consistency judge to be relative.By this method, the mapping of conclusion contradiction can be made and deposit, and calculating based on such mapping, obtaining equipment control command.For our situation, the item of conclusion checking is herein the same with conclusion, so when the conclusion that also can be understood as machine learning control mapping is consistent with the conclusion that self-defined control maps, the conclusion that machine learning controls to map is set up.
For previous example: machine learning controls to be mapped as " the if time is 7 points, and parlor brightness <th, and then turns on parlor lamp "; Self-defined control is mapped as " the if time is that 2014.9.9-2014.9.10, then need not turn on parlor lamp ".If environment sensing information meets Article 1 map (time that namely meets is at 7 and parlor brightness <th) and Article 2 mapping (time that namely meets is 2014.9.9-2014.9.10), because both conclusions are inconsistent, then do not extrapolate conclusion.After introducing DefaultLogic, above-mentioned machine learning controls mapping can be updated to that (the if time is 7 points, and parlor brightness <th): turn on parlor lamp (i.e. conclusion checking item)/turn on parlor lamp (i.e. conclusion) "; correspond to above-mentioned general principle formula; namely α (x) is that " the if time is 7 points, and brightness <th ", β (x)=γ (x)=" turning on parlor lamp ".So when reckoning, environment sensing information meets the prerequisite that machine learning controls to map, and enables this and maps, and then needs to judge whether β (x) now meets consistency.Will find that environment sensing information now enables self-defined control mapping equally, and conclusion is " need not turn on parlor lamp ", so the conclusion of this conclusion of β (x) checking item and self-defined control mapping is inconsistent, so the conclusion that machine learning controls to map cannot be set up.
Above-mentioned a kind of method for resolving contradiction; Also the method for probability can be adopted, certain probable value is given to all mappings, and define probable value attenuation function, as time goes by, probable value can decay gradually, and mapping newer like this can have higher confidence level (the time gap current time namely mapping generation is nearer, and its confidence level is higher), thus adopt by inference engine, obtain corresponding equipment control command.
Step 303, equipment control command is showed user by controller, if the mandate received from user indicates, then performs step 304; If do not receive user feedback in setting-up time section, then according to default behavior, directly perform step 304.
Equipment control command is fed back to user interface, and the operation behavior follow-up to control system for user is browsed and controls.Herein the controlling behavior of user can simply be divided into three kinds-" Never ", " OK ", " Not This Time "." Never " represents this conclusion mistake, and inference engine is when reaching a conclusion next time, and before propelling movement, conclusion is anti-; " OK " represents out of question, and inference engine conclusion is correct, and user can accept; " Not This Time " represents only current inaccurate specifically, and inference engine does not process for this situation, and current is the special processing of user specifically.Setting feedback time section, if user does not make corresponding operation, system default inference conclusion is accurate, namely obtains the mandate of user; After the mandate obtaining user, perform corresponding equipment controlling behavior.
Step 304, controls intelligent home device according to equipment control command.
Instant invention overcomes the shortcoming of prior art, provide a kind of intelligent home furnishing control method calculated based on general knowledge.General knowledge calculates to have two features, and nonmonotonicity and time harmony, obtain the most reasonably calculating conclusion based on current knowledge, can tolerate the existence of contradiction.This feature just in time can solve the collision problem that in intelligent home control system, emergency case is brought.What general knowledge calculated realizes mainly comprising commonsense knowledge base and inference engine, and the acquisition of commonsense knowledge base has number of ways, can be the crawl of online data, user's contribution data, or utilizes existing commonsense knowledge base; Inference engine mainly realizes commonsense reasoning algorithm, carries out reckoning obtain rational conclusion based on commonsense knowledge base.Main purpose of the present invention solves inference engine in prior art cannot process the problem of contradiction or conflict, and effectively prevent intelligent control system cannot normal operation, enhances the fault-tolerant ability of inference engine, makes system more intelligently autonomous operation.
See Fig. 4, be the system configuration schematic diagram that the present invention controls intelligent home device, this system comprises user side, server and controller;
Described user side, receives the burst control information of user's input, burst control information and household identification is sent to described server;
Described server, receives the burst control information from described client and household identification, resolves burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification;
Described controller, carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; The self-defined control received from server maps, and determines equipment control command, control according to equipment control command to intelligent home device based on machine learning control mapping and self-defined control mapping.
Preferably, described server comprises Mapping Resolution device, commonsense knowledge base, customer equipment storehouse and maps thesaurus;
Described customer equipment storehouse, stores the intelligent home device information of corresponding each household identification;
Described commonsense knowledge base, stores general knowledge, comprises the control overflow of corresponding event;
Described Mapping Resolution device, receives the burst control information from client and household identification, carries out natural-sounding analysis to burst control information, extracts Time And Event wherein, inquires corresponding control overflow according to event in described commonsense knowledge base; From described customer equipment storehouse, inquire the intelligent home device information corresponding with corresponding household identification, combined with intelligent home equipment information, time and control overflow, the self-defined control generating corresponding intelligent home device maps, and is stored in described mapping thesaurus;
Described mapping thesaurus, the self-defined control stored from described Mapping Resolution device maps, and is synchronized to the controller corresponding with household identification.
Preferably, described controller comprises family mapping thesaurus, inference engine module and actuator;
Described family maps thesaurus, and the self-defined control received from described server maps, and stores; And storage control controls to map by the machine learning that machine learning mode obtains to the historical operation record of Smart Home;
Described inference engine module, maps thesaurus from described family and extracts the mapping of machine study control and self-defined control mapping, map and determines equipment control command, send to described actuator based on the control of extracting;
Described actuator, receives the equipment control command from described inference engine module, controls corresponding intelligent home device.
Preferably, described inference engine module comprises the first inference engine submodule, obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if so, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if unanimously, then generate the equipment control command controlling to map conclusion about machine learning, send to described actuator.
Preferably, described inference engine module comprises the second inference engine submodule, obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if so, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if unanimously, then by generating the equipment control command controlling to map conclusion about machine learning, send to described actuator; If inconsistent, then compare machine learning and control mapping and the self-defined confidence level controlling to map, determine that the control that confidence level is large maps, extract the conclusion in the control mapping determined, generate the equipment control command about extracted conclusion, send to described actuator.
Preferably, described user side comprises MIM message input module and information text module;
Described MIM message input module, receives the burst control information of user's input, sends to described information text module;
Described information text module, judges the burst control information received, if be text message, then directly sends to server, if be not text message, then burst control information be converted to text message, send to server.
Preferably, described family maps thesaurus, also regularly detects self-defined control mapping, deletes the self-defined control mapping of losing efficacy.
In control system scheme provided by the invention, user can by any information such as mobile phone, computer input client, by burst demand or emergency case with the form input control system of voice or text; Input information is converted to the available mapping of reckoning by means of commonsense knowledge base and natural language analysis technology by Mapping Resolution device, maps as self-defined control, and is stored in mapping thesaurus.The mapping that inference engine module maps in thesaurus based on family calculates, obtains conclusion, thus the equipment control command obtained, call actuator and perform corresponding equipment control command to reach the automated execution of equipment.
Mapping Resolution device is mainly resolved the burst control information of user's input and is converted to self-defined control mapping, the latent meaning of burst control information is analyzed by means of natural language technology and commonsense knowledge base, and convert corresponding self-defined control mapping to, stored in mapping thesaurus.
General knowledge reckoning is carried out in the mapping that inference engine module primary responsibility maps in thesaurus based on family, obtains rational conclusion.It is a kind of non-monotonic reckoning that general knowledge calculates, be divided into the method for logic and alogical method, the method of logic can pass through two kinds of approach, and a kind of is the method carrying out studying under classical logic framework, and another sets up the method for new semanteme mechanism with logic system.The combination of logical method and NOT logic method is also the research field relatively enlivened now.
Family maps storage and the renewal of the mapping of thesaurus primary responsibility.Basic mapping derives from the regular study of machine learning method to user facility operation record, obtains the operation behavior mode of user, as the foundation that equipment automatization runs.Focus on the self-defined renewal controlling to map; Derive from the output of Mapping Resolution device, may produce conflict with original behavior pattern, how to solve the mapping of conclusion conflict and deposit family just and map the problem that thesaurus and inference engine module need to solve also is core place of the present invention.
Adopt the present invention program, the burst demand input of user can be received, and be converted to self-defined control mapping, as follow-up derivation reference; Further, also introduce commonsense reasoning technology, the mapping of conclusion conflict can be accepted and deposit, and reckoning obtains rational conclusion on this basis.Like this, make the control of intelligent home device more intelligent and flexible, promote the experience of user.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (13)
1. to the method that intelligent home device controls, it is characterized in that, the method comprises:
User side receives the burst control information of user's input, and burst control information and household identification are sent to server;
Server is resolved burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification;
Controller carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; And the self-defined control received from server maps, determine equipment control command based on machine learning control mapping and self-defined control mapping, according to equipment control command, intelligent home device is controlled.
2. the method for claim 1, is characterized in that, described server carries out parsing to burst control information and comprises:
Server is analyzed burst control information, extracts Time And Event wherein, inquires corresponding control overflow according to event in commonsense knowledge base; From customer equipment storehouse, inquire the intelligent home device information corresponding with corresponding household identification, combined with intelligent home equipment information, time and control overflow, the self-defined control generating corresponding intelligent home device maps.
3. the method for claim 1, is characterized in that, described controller controls mapping based on machine learning and self-defined control mapping determines that equipment control command comprises:
Controller obtains environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, and generates equipment control command.
4. the method for claim 1, is characterized in that, described controller controls mapping based on machine learning and self-defined control mapping determines that equipment control command comprises:
Obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, and generates equipment control command, equipment control command is sent to described actuator; If inconsistent, then compare machine learning and control mapping and the self-defined confidence level controlling to map, determine that the control that confidence level is large maps, extract the conclusion in the control mapping determined, generate the equipment control command about extracted conclusion.
5. the method for claim 1, is characterized in that, the method also comprises:
Controller regularly detects self-defined control mapping, deletes the self-defined control of losing efficacy and maps.
6. the method according to any one of claim 1 to 5, is characterized in that, after described user side receives the burst control information of user's input, the method also comprises:
User side judges the burst control information received, if be text message, then directly sends to server, if be not text message, then burst control information be converted to text message, send to server.
7. the method according to any one of claim 1 to 5, is characterized in that, described according to equipment control command intelligent home device controlled before, the method also comprises:
Equipment control command is showed user, if the mandate received from user indicates, then performs described step intelligent home device controlled according to equipment control command.
8. to the system that intelligent home device controls, it is characterized in that, this system comprises user side, server and controller;
Described user side, receives the burst control information of user's input, burst control information and household identification is sent to described server;
Described server, receives the burst control information from described client and household identification, resolves burst control information, and the self-defined control obtaining the condition that comprises and conclusion maps, and is synchronized to the controller corresponding with household identification;
Described controller, carries out machine learning to the history control information of intelligent home device, and the machine learning obtaining the condition that comprises and conclusion controls to map; The self-defined control received from server maps, and determines equipment control command, control according to equipment control command to intelligent home device based on machine learning control mapping and self-defined control mapping.
9. system as claimed in claim 8, is characterized in that, described server comprises Mapping Resolution device, commonsense knowledge base, customer equipment storehouse and maps thesaurus;
Described customer equipment storehouse, stores the intelligent home device information of corresponding each household identification;
Described commonsense knowledge base, stores general knowledge, comprises the control overflow of corresponding event;
Described Mapping Resolution device, receives the burst control information from client and household identification, carries out natural language analysis to burst control information, extracts Time And Event wherein, inquires corresponding control overflow according to event in described commonsense knowledge base; From described customer equipment storehouse, inquire the intelligent home device information corresponding with corresponding household identification, combined with intelligent home equipment information, time and control overflow, the self-defined control generating corresponding intelligent home device maps, and is stored in described mapping thesaurus;
Described mapping thesaurus, the self-defined control stored from described Mapping Resolution device maps, and is synchronized to the controller corresponding with household identification.
10. system as claimed in claim 8, is characterized in that, described controller comprises family and maps thesaurus, inference engine module and actuator;
Described family maps thesaurus, and the self-defined control received from described server maps, and stores; And storage control controls to map by the machine learning that machine learning mode obtains to the historical operation record of Smart Home;
Described inference engine module, maps thesaurus from described family and extracts the mapping of machine study control and self-defined control mapping, map and determines equipment control command, send to described actuator based on the control of extracting;
Described actuator, receives the equipment control command from described inference engine module, controls corresponding intelligent home device.
11. systems as claimed in claim 10, it is characterized in that, described inference engine module comprises the first inference engine submodule, obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, generation equipment control command, sends to described actuator.
12. systems as claimed in claim 10, it is characterized in that, described inference engine module comprises the second inference engine submodule, obtain environment sensing information, judge whether environment sensing information meets machine learning and control to map and the self-defined condition controlled in mapping, if, then judge that whether the conclusion checking item that machine learning controls in mapping is consistent with the conclusion that self-defined control maps, if consistent, the conclusion that then machine learning controls to map is set up, generation equipment control command, sends to described actuator; If inconsistent, then compare machine learning and control mapping and the self-defined confidence level controlling to map, determine that the control that confidence level is large maps, extract the conclusion in the control mapping determined, generate the equipment control command about extracted conclusion, send to described actuator.
13. systems according to any one of claim 8 to 12, it is characterized in that, described user side comprises MIM message input module and information text module;
Described MIM message input module, receives the burst control information of user's input, sends to described information text module;
Described information text module, judges the burst control information received, if be text message, then directly sends to server, if be not text message, then burst control information be converted to text message, send to server.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510040534.1A CN104579873B (en) | 2015-01-27 | 2015-01-27 | The method and system being controlled to intelligent home device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510040534.1A CN104579873B (en) | 2015-01-27 | 2015-01-27 | The method and system being controlled to intelligent home device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104579873A true CN104579873A (en) | 2015-04-29 |
CN104579873B CN104579873B (en) | 2017-09-22 |
Family
ID=53095130
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510040534.1A Active CN104579873B (en) | 2015-01-27 | 2015-01-27 | The method and system being controlled to intelligent home device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104579873B (en) |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104881013A (en) * | 2015-05-29 | 2015-09-02 | 四川长虹电器股份有限公司 | Intelligent household electrical appliance linkage control method based on user living habit mining |
CN104898634A (en) * | 2015-06-30 | 2015-09-09 | 联想(北京)有限公司 | Information processing method and electronic equipment |
CN105137779A (en) * | 2015-07-30 | 2015-12-09 | 四川长虹电器股份有限公司 | Information processing method and smart home environment control equipment |
CN105404161A (en) * | 2015-11-02 | 2016-03-16 | 百度在线网络技术(北京)有限公司 | Intelligent voice interaction method and device |
WO2016188336A1 (en) * | 2015-05-22 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Control method and apparatus for smart home system |
CN106527168A (en) * | 2016-12-27 | 2017-03-22 | 海尔优家智能科技(北京)有限公司 | AllJoyn linkage control conflict resolution method, trigger and server |
CN107045291A (en) * | 2016-02-05 | 2017-08-15 | 中国电信股份有限公司 | Forming method, the device and system of intelligent terminal control rule |
CN107153359A (en) * | 2016-03-03 | 2017-09-12 | 阿姆Ip有限公司 | What configuration was set accesses in limited time |
CN107564522A (en) * | 2017-09-18 | 2018-01-09 | 郑州云海信息技术有限公司 | A kind of intelligent control method and device |
CN107688329A (en) * | 2017-08-21 | 2018-02-13 | 杭州古北电子科技有限公司 | Intelligent home furnishing control method and intelligent home control system |
CN108111378A (en) * | 2017-12-22 | 2018-06-01 | 南京物联传感技术有限公司 | The conflict coordination system and method for work of a kind of scene settings of smart home |
CN108919669A (en) * | 2018-09-11 | 2018-11-30 | 深圳和而泰数据资源与云技术有限公司 | A kind of smart home dynamic decision method, apparatus and service terminal |
WO2019000980A1 (en) * | 2017-06-30 | 2019-01-03 | 华为技术有限公司 | Control method, related device, and system |
CN109324585A (en) * | 2018-11-20 | 2019-02-12 | 曹景荣 | A kind of monitoring system and method for smart home |
CN109361766A (en) * | 2018-12-04 | 2019-02-19 | 安徽信息工程学院 | A kind of internet of things net controller based on SDN |
CN109683715A (en) * | 2018-12-29 | 2019-04-26 | 北京诺亦腾科技有限公司 | A kind of VR apparatus control method, device and computer readable storage medium |
CN110888336A (en) * | 2019-12-02 | 2020-03-17 | 蚁蜂(广州)智能科技有限公司 | Intelligent household control system and control method |
CN112328325A (en) * | 2020-11-06 | 2021-02-05 | 深圳壹账通智能科技有限公司 | Execution method and device of model file, terminal equipment and storage medium |
CN114722050A (en) * | 2022-06-10 | 2022-07-08 | 辰星(天津)自动化设备有限公司 | Data synchronization method of robot system and robot system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20080046964A (en) * | 2006-11-24 | 2008-05-28 | (주)오로라 디자인랩 | A service method of the intelligence type home network system for the user of an equipment control base |
CN201435002Y (en) * | 2009-07-16 | 2010-03-31 | 吕晓冬 | Remote automation control system of intelligent household |
CN202939771U (en) * | 2012-11-04 | 2013-05-15 | 张宁 | Intelligent home safety monitoring system |
CN103393412A (en) * | 2013-08-15 | 2013-11-20 | 重庆邮电大学 | Intelligent home based old person caring device |
CN103744411A (en) * | 2014-02-07 | 2014-04-23 | 上海金牌软件开发有限公司 | Control method related to ZigBee technology for realizing smart home |
-
2015
- 2015-01-27 CN CN201510040534.1A patent/CN104579873B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20080046964A (en) * | 2006-11-24 | 2008-05-28 | (주)오로라 디자인랩 | A service method of the intelligence type home network system for the user of an equipment control base |
CN201435002Y (en) * | 2009-07-16 | 2010-03-31 | 吕晓冬 | Remote automation control system of intelligent household |
CN202939771U (en) * | 2012-11-04 | 2013-05-15 | 张宁 | Intelligent home safety monitoring system |
CN103393412A (en) * | 2013-08-15 | 2013-11-20 | 重庆邮电大学 | Intelligent home based old person caring device |
CN103744411A (en) * | 2014-02-07 | 2014-04-23 | 上海金牌软件开发有限公司 | Control method related to ZigBee technology for realizing smart home |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016188336A1 (en) * | 2015-05-22 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Control method and apparatus for smart home system |
CN106292302A (en) * | 2015-05-22 | 2017-01-04 | 阿里巴巴集团控股有限公司 | The control method of intelligent domestic system and device |
CN104881013B (en) * | 2015-05-29 | 2018-01-16 | 四川长虹电器股份有限公司 | The intelligent household electrical appliance linkage control method excavated based on user's habits and customs |
CN104881013A (en) * | 2015-05-29 | 2015-09-02 | 四川长虹电器股份有限公司 | Intelligent household electrical appliance linkage control method based on user living habit mining |
CN104898634A (en) * | 2015-06-30 | 2015-09-09 | 联想(北京)有限公司 | Information processing method and electronic equipment |
CN105137779A (en) * | 2015-07-30 | 2015-12-09 | 四川长虹电器股份有限公司 | Information processing method and smart home environment control equipment |
CN105137779B (en) * | 2015-07-30 | 2018-02-16 | 四川长虹电器股份有限公司 | A kind of information processing method and smart home environment control device |
CN105404161A (en) * | 2015-11-02 | 2016-03-16 | 百度在线网络技术(北京)有限公司 | Intelligent voice interaction method and device |
CN107045291A (en) * | 2016-02-05 | 2017-08-15 | 中国电信股份有限公司 | Forming method, the device and system of intelligent terminal control rule |
CN107153359A (en) * | 2016-03-03 | 2017-09-12 | 阿姆Ip有限公司 | What configuration was set accesses in limited time |
CN106527168A (en) * | 2016-12-27 | 2017-03-22 | 海尔优家智能科技(北京)有限公司 | AllJoyn linkage control conflict resolution method, trigger and server |
WO2019000980A1 (en) * | 2017-06-30 | 2019-01-03 | 华为技术有限公司 | Control method, related device, and system |
US11556100B2 (en) | 2017-06-30 | 2023-01-17 | Huawei Technologies Co., Ltd. | Control method, related device, and system |
CN107688329A (en) * | 2017-08-21 | 2018-02-13 | 杭州古北电子科技有限公司 | Intelligent home furnishing control method and intelligent home control system |
CN107564522A (en) * | 2017-09-18 | 2018-01-09 | 郑州云海信息技术有限公司 | A kind of intelligent control method and device |
CN108111378A (en) * | 2017-12-22 | 2018-06-01 | 南京物联传感技术有限公司 | The conflict coordination system and method for work of a kind of scene settings of smart home |
CN108919669B (en) * | 2018-09-11 | 2022-04-29 | 深圳和而泰数据资源与云技术有限公司 | Intelligent home dynamic decision method and device and service terminal |
CN108919669A (en) * | 2018-09-11 | 2018-11-30 | 深圳和而泰数据资源与云技术有限公司 | A kind of smart home dynamic decision method, apparatus and service terminal |
CN109324585A (en) * | 2018-11-20 | 2019-02-12 | 曹景荣 | A kind of monitoring system and method for smart home |
CN109361766A (en) * | 2018-12-04 | 2019-02-19 | 安徽信息工程学院 | A kind of internet of things net controller based on SDN |
CN109683715B (en) * | 2018-12-29 | 2020-12-18 | 北京诺亦腾科技有限公司 | VR equipment control method and device and computer readable storage medium |
CN109683715A (en) * | 2018-12-29 | 2019-04-26 | 北京诺亦腾科技有限公司 | A kind of VR apparatus control method, device and computer readable storage medium |
CN110888336A (en) * | 2019-12-02 | 2020-03-17 | 蚁蜂(广州)智能科技有限公司 | Intelligent household control system and control method |
CN112328325A (en) * | 2020-11-06 | 2021-02-05 | 深圳壹账通智能科技有限公司 | Execution method and device of model file, terminal equipment and storage medium |
CN114722050A (en) * | 2022-06-10 | 2022-07-08 | 辰星(天津)自动化设备有限公司 | Data synchronization method of robot system and robot system |
Also Published As
Publication number | Publication date |
---|---|
CN104579873B (en) | 2017-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104579873A (en) | Method and system for controlling intelligent home equipment | |
CN109829106B (en) | Automatic recommendation method and device, electronic equipment and storage medium | |
CN101632050B (en) | Dynamically extensible and automatically configurable building automation system and architecture | |
US9182755B2 (en) | Automated operator interface generation in a control system | |
CN109240111A (en) | Intelligent home furnishing control method, device, system and intelligent gateway | |
Dibowski et al. | Automated design of building automation systems | |
CN113064351A (en) | Digital twin model construction method and device, storage medium and electronic equipment | |
Lu | IoT-enabled adaptive context-aware and playful cyber-physical system for everyday energy savings | |
CN104486416A (en) | Comprehensive utilizing system and method of intelligent home service rule | |
CN107179701B (en) | Intelligent household equipment self-adaptive linkage rule generation method | |
CN110333850A (en) | Generation method, device, computer equipment and the storage medium of interface document | |
CN106170764A (en) | Method and apparatus for automatic equipment Program Generating | |
CN104898634A (en) | Information processing method and electronic equipment | |
CN103713612A (en) | Intelligent home system solution facilitating personalized customization and expansion | |
CN105467846A (en) | Intelligent household appliance control method and system | |
CN104460573A (en) | Device-operation management apparatus, method of controlling device-operation management apparatus, and remote operation system | |
US10574764B2 (en) | Automated learning universal gateway | |
CN109407538A (en) | Intelligent home furnishing control method and system | |
CN106292302A (en) | The control method of intelligent domestic system and device | |
Kerang et al. | Task management system according to changes in the situation based on IoT | |
CN115793481A (en) | Device control method, device, electronic device and storage medium | |
Sai et al. | Smart Home Messenger Notifications System using IoT | |
CN111431776A (en) | Information configuration method, device and system | |
CN205427589U (en) | Intelligent house control system based on believe platform a little | |
CN109218085A (en) | Equipment automatization and scene configuration management method, device and mobile terminal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |