WO2016013411A1 - 制御装置、制御方法、情報処理装置、情報処理方法、及び、プログラム - Google Patents
制御装置、制御方法、情報処理装置、情報処理方法、及び、プログラム Download PDFInfo
- Publication number
- WO2016013411A1 WO2016013411A1 PCT/JP2015/069828 JP2015069828W WO2016013411A1 WO 2016013411 A1 WO2016013411 A1 WO 2016013411A1 JP 2015069828 W JP2015069828 W JP 2015069828W WO 2016013411 A1 WO2016013411 A1 WO 2016013411A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- state
- label
- unit
- information
- operating state
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 35
- 238000003672 processing method Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 title abstract description 98
- 238000000926 separation method Methods 0.000 claims abstract description 184
- 238000002372 labelling Methods 0.000 claims abstract description 35
- 238000001514 detection method Methods 0.000 claims description 141
- 238000004891 communication Methods 0.000 claims description 78
- 230000007704 transition Effects 0.000 claims description 42
- 230000008859 change Effects 0.000 claims description 31
- 230000006870 function Effects 0.000 claims description 8
- 230000036541 health Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 16
- 239000003795 chemical substances by application Substances 0.000 description 121
- 230000008569 process Effects 0.000 description 87
- 238000012545 processing Methods 0.000 description 36
- 239000011159 matrix material Substances 0.000 description 22
- 238000010586 diagram Methods 0.000 description 21
- 230000009471 action Effects 0.000 description 19
- 238000006243 chemical reaction Methods 0.000 description 16
- 238000011156 evaluation Methods 0.000 description 12
- 239000004065 semiconductor Substances 0.000 description 12
- 239000006185 dispersion Substances 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 6
- 230000014509 gene expression Effects 0.000 description 6
- 238000005286 illumination Methods 0.000 description 6
- 238000005259 measurement Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- JLQFVGYYVXALAG-CFEVTAHFSA-N yasmin 28 Chemical compound OC1=CC=C2[C@H]3CC[C@](C)([C@](CC4)(O)C#C)[C@@H]4[C@@H]3CCC2=C1.C([C@]12[C@H]3C[C@H]3[C@H]3[C@H]4[C@@H]([C@]5(CCC(=O)C=C5[C@@H]5C[C@@H]54)C)CC[C@@]31C)CC(=O)O2 JLQFVGYYVXALAG-CFEVTAHFSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/25—Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
- G01R19/2513—Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R21/00—Arrangements for measuring electric power or power factor
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/041—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a variable is automatically adjusted to optimise the performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2642—Domotique, domestic, home control, automation, smart house
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2310/00—The network for supplying or distributing electric power characterised by its spatial reach or by the load
- H02J2310/10—The network having a local or delimited stationary reach
- H02J2310/12—The local stationary network supplying a household or a building
- H02J2310/14—The load or loads being home appliances
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2310/00—The network for supplying or distributing electric power characterised by its spatial reach or by the load
- H02J2310/70—Load identification
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02B90/20—Smart grids as enabling technology in buildings sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
Definitions
- the present technology relates to a control device, a control method, an information processing device, an information processing method, and a program.
- device separation for obtaining power consumption of each of a plurality of devices such as homes can be quickly performed.
- the present invention relates to a control device, a control method, an information processing device, an information processing method, and a program.
- NILM Non-Intrusive Load Monitoring
- FHMM Factor Hidden Markov Model
- the device In order to perform device separation to obtain the current consumption of each device from the current information measured by the distribution board, the device is turned on or off, for example, by changing the operating state of the device. It is necessary for the current consumption of the device to change as a result.
- the operating state of the device can be changed, for example, by the user's operation. However, depending on the user's operation, it takes time to separate the devices.
- This technology has been made in view of such a situation, and enables device separation to be performed quickly.
- the information processing apparatus or the first program of the present technology operates a device, recognizes the operation state of the device, and transmits a device label representing the device and an operation state label representing the operation state of the device.
- the device information acquisition unit that acquires the device label and the operation state label from the control device, and the total data regarding the total current consumed by a plurality of devices, the consumption of each operation state of each device
- a device that separates the current consumption of each device by updating the pattern information based on the possibility information representing the possibility that the current consumption represented by the pattern information is consumed, which is obtained for the pattern information representing the current.
- the device is operated, the operating state of the device is recognized, the device label representing the device, and the operating state of the device are represented.
- the device label and the operating state label are acquired from the control device that transmits the operating state label.
- the consumption current represented by the pattern information obtained for the pattern information representing the consumption current in each operating state of each device is consumed by using the sum data relating to the sum of the current consumed by multiple devices.
- the possibility information obtained in the device separation for separating the current consumption of each device is acquired by updating the pattern information based on the possibility information representing the property. Then, based on the possibility information, pattern information indicating current consumption consumed in the current operating state of the device represented by the device label is specified, and the device label and the operating state label are included in the pattern information. Corresponding labeling is performed.
- the control method of the present technology includes a step of operating a device, a step of recognizing the operation state of the device, and sum data relating to a sum of currents consumed by a plurality of devices, and each operation state of each device.
- the current consumption of each device is separated by updating the pattern information based on the possibility information representing the possibility that the current consumption represented by the pattern information is consumed, which is obtained for the pattern information representing the current consumption.
- a control method including a step of transmitting a device label representing the device and an operation state label representing an operation state of the device to a device separation apparatus that performs device separation.
- the device is operated and the operating state of the device is recognized. Then, using the sum data on the sum of the currents consumed by multiple devices, the current consumption represented by the pattern information obtained for the pattern information representing the current consumption in each operating state of each device may be consumed. Based on the possibility information that represents the device, by updating the pattern information, the device separation device for separating the current consumption of each device, the device label representing the device, and the operating state of the device An operating state label representing is transmitted.
- the information processing device and the control device may be independent devices or may be internal blocks constituting one device.
- the program can be provided by being transmitted through a transmission medium or by being recorded on a recording medium.
- FIG. 3 is a block diagram illustrating a configuration example of a device separation device 16. It is a figure explaining FHMM. It is a figure explaining the outline
- FIG. 4 is a perspective view showing an external configuration example of an agent 14.
- FIG. 4 is a block diagram illustrating an example of an internal configuration of an agent 14.
- 4 is a diagram illustrating an example of a device table stored in a semiconductor memory 115.
- FIG. It is a block diagram which shows the structural example of the label acquisition part 35 in case the agent 14 and the apparatus separation apparatus 16 operate
- 6 is a diagram illustrating an example of a correspondence table stored in a correspondence storage unit 204.
- FIG. 1 is a diagram illustrating a configuration example of an embodiment of a device separation system to which the present technology is applied.
- a distribution board 11 is attached to the user's home (home, company, etc.).
- the electric power provided from the electric power company is drawn into the distribution board 11 via the wattmeter 12, and is supplied from the distribution board 11 to devices such as home appliances (connected to an outlet or the like) at the user's house. .
- the current sensor 13 is attached to the distribution board 11, for example.
- the current sensor 13 measures the total current consumed by all the devices (one or more devices) in the user's home at one location at the root of the distribution board 11 that supplies power to the user's home. Then, the data is transmitted to the device separation device 16 configured on the cloud, for example, via the network 15 such as the Internet.
- the agent 14 is, for example, a movable pet-type robot imitating a dog, and functions as a control device that controls equipment at the user's home.
- the agent 14 moves the hand or foot to directly operate the buttons provided on the device or the remote controller of the device, thereby operating the device such as turning on / off the device or changing the operation mode. I do.
- the agent 14 operates (controls) the device by emitting infrared rays similar to the infrared rays emitted by the remote controller of the device, or by performing wireless or wired communication with the device via the home network. can do.
- the agent 14 can operate the device by holding a stick or the like by hand and operating a button provided on the device with the stick.
- the agent 14 recognizes the operating state (for example, ON or OFF state, mute state, etc.) of the device at the user's home, and supplies it to the device separation device 16 via the network 15 as device information related to the device. To do.
- the operating state for example, ON or OFF state, mute state, etc.
- the device separation device 16 receives the sum of currents transmitted from the current sensor 13 of the user's home via the network 15, and from the current sum series (current waveform series), for example, TV ( Equipment separation that separates current consumption and power consumption consumed by individual equipment such as television receivers), electric pots, refrigerators, and lighting.
- TV Equipment separation that separates current consumption and power consumption consumed by individual equipment such as television receivers
- electric pots refrigerators
- lighting lighting.
- the device separation device 16 uses the device information from the agent 14 to label the current consumption, etc. obtained as a result of the device separation, indicating in what operating state of which device.
- the device separation information 16 detects the operating state of the device at the user's home from the current consumption obtained as a result of the device separation and transmits it to the agent 14 via the network 15.
- FIG. 2 is a diagram for explaining an example of processing of the device separation system of FIG.
- the device separation device 16 receives the sum of currents transmitted from the current sensor 13 at the user's home, and calculates the current consumed by each device in the user's home from the current sum series (current waveform series). Equipment separation is performed.
- the agent 14 at the user's home moves within the user's home as necessary, and operates, for example, a button provided on the device or a device remote controller to turn on / off the device. Perform the operation.
- the sum of the current consumption of devices # 1 to # 4 when a device is turned off or the devices # 1 to # 4 when the power of other devices is turned on The total power consumption when the devices # 1 to # 4 are in various operating states, such as the total current consumption, is transmitted from the current sensor 13 to the device separation device 16.
- the device separation device 16 performs device separation using the sum of the power consumption when the devices # 1 to # 4 are in various operating states, whereby the consumption of each of the devices # 1 to # 4 is performed. The current is separated.
- the device separation device 16 detects the operating state of the device at the user's home from the current consumption of each of the devices # 1 to # 4 obtained as a result of the device separation, and transmits it to the agent 14 via the network 15.
- the agent 14 receives the operating state of the device at the user's home from the device separating apparatus 16, and performs a predetermined action according to the operating state.
- the agent 14 when the agent 14 receives from the device separation device 16 the operating state of the device indicating that the illumination of the entrance of the user's house is turned on, the illumination of the entrance of the agent 14 is not required even if the agent 14 is not at the entrance of the user's home. You can recognize that it is turned on.
- the agent 14 can notify the user near the current location by voice or the like that the front door lighting of the user's house has been turned on.
- a current sensor 13 is provided on the distribution board 12, and the current sensor 13 simply measures the total current consumed by the devices in the user's home, and the current sum series (current waveform series). ), Current consumption (power consumption) of individual devices in the user's home is separated.
- total data relating to the total current consumed by each device such as the total current consumed by each device can be employed.
- the sum data the sum of the values that can be added can be used.
- the total data in addition to the total current consumed by each device itself, for example, the sum of the power consumed by each device and the waveform of the current consumed by each device can be FFT.
- the sum of the frequency components obtained by (Fast Fourier Transform) or the like can be adopted.
- the sum of currents consumed by each device is adopted as the sum data, and in device separation, for example, from the waveform of the sum of currents as sum data, consumed by individual devices.
- the waveform of the current is separated.
- waveform separation learning is performed to obtain the current consumption waveform of each device from the total data.
- the characteristic waveform W (m) consumed by each device #m is obtained.
- FIG. 4 there are five devices # 1 to # 5 in the user's home, and devices # 1, # 2, # 4, and # 5 are turned on among the five devices # 1 to # 5. Is in a state (a state in which power is consumed), and device # 3 is in an off state (a state in which power is not consumed).
- FIG. 5 is a block diagram illustrating a configuration example of the device separation device 16 of FIG.
- the communication unit 30 communicates with the current sensor 13 and the agent 14 via the network 15.
- the communication unit 30 receives the time series (current time series) of the current waveform Y as the total data transmitted from the current sensor 13 via the network 15 and supplies it to the data acquisition unit 31.
- the communication unit 30 receives data transmitted from the agent 14 via the network 15 and supplies the data to the label acquisition unit 35 and the data output unit 36. Further, the communication unit 30 transmits data supplied from the label acquisition unit 35 and the data output unit 36 to the agent 14 via the network 15.
- the communication may be any of wireless communication, wired communication, and communication in which wireless and wired are mixed.
- the data acquisition unit 31 acquires a time series (current time series) of the current waveform Y as total data transmitted from the current sensor 13 by receiving via the communication unit 30, and the state estimation unit 32, model The data is supplied to the learning unit 34 and the data output unit 36.
- the data acquisition unit 31 acquires a time series (voltage time series) of a voltage waveform (voltage waveform) V corresponding to the current waveform Y as total data, a state estimation unit 32, a model learning unit 34, and The data is supplied to the data output unit 36.
- the voltage waveform V can be measured by the distribution board 12 and transmitted to the device separation device 16 via the network 15 in the same manner as the current waveform Y. Further, as the voltage waveform V, a sin wave having a predetermined value such as 100 V and an effective value of a predetermined frequency such as 50 Hz or 60 Hz, which approximates a voltage supplied from an electric power company, can be employed.
- the state estimation unit 32 uses the current waveform Y from the data acquisition unit 31 and the overall model (model parameter) ⁇ stored in the model storage unit 33 as a model of the entire device in the user home. The state estimation for estimating the state of the entire model corresponding to each device is performed. Then, the state estimation unit 32 supplies the state estimation result (state of the entire model) ⁇ of the state estimation to the model learning unit 34, the label acquisition unit 35, and the data output unit 36.
- the state estimation unit 32 includes an evaluation unit 41 and an estimation unit 42.
- the evaluation unit 41 is supplied from the data acquisition unit 31 (to the state estimation unit 32) in each combination of the states of the plurality of device models # 1 to #M constituting the overall model ⁇ stored in the model storage unit 33.
- An evaluation value E that evaluates the degree to which the current waveform Y is observed is obtained and supplied to the estimation unit 42.
- the estimation unit 42 uses the evaluation value E supplied from the evaluation unit 41 to estimate the state ⁇ of each of the plurality of device models # 1 to #M constituting the overall model ⁇ stored in the model storage unit 33, and the model The data is supplied to the learning unit 34, the label acquisition unit 35, and the data output unit 36.
- the overall model ⁇ is composed of a plurality of device models # 1 to #M which are models (representing current consumption) of devices.
- the device models # 1 to #M and the overall model ⁇ composed of the device models # 1 to #M are, for example, a probability generation model or a state transition model, and have a plurality of states.
- the parameter ⁇ of the overall model includes a current waveform parameter that represents current consumption for each state of the device model #m (operating state of the corresponding device).
- the model parameter ⁇ of the entire model stored in the model storage unit 33 is referred to as necessary by the evaluation unit 41 and the estimation unit 42 of the state estimation unit 32, the label acquisition unit 35, and the data output unit 36, and the model It is updated by a waveform separation learning unit 51, a distributed learning unit 52, and a state variation learning unit 53, which will be described later, of the learning unit.
- the model learning unit 34 includes the current waveform Y supplied from the data acquisition unit 31 and the state estimation result of state estimation supplied from the state estimation unit 32 (the estimation unit 42 thereof) (each device model constituting the overall model). Model learning is performed to update the model parameter ⁇ of the entire model stored in the model storage unit 33 using the state #m).
- the waveform separation learning unit 51 includes the current waveform Y supplied from the data acquisition unit 31 (to the model learning unit 34) and the state estimation result of each device model #m supplied from the state estimation unit 32 (the estimation unit 42 thereof).
- waveform separation learning for obtaining (updating) the current waveform parameter as the model parameter ⁇ is performed, and the current waveform parameter stored in the model storage unit 33 is determined by the current waveform parameter obtained by the waveform separation learning. Update.
- the distributed learning unit 52 includes the current waveform Y supplied from the data acquisition unit 31 (to the model learning unit 34) and the state estimation result ⁇ of each device model #m supplied from the state estimation unit 32 (the estimation unit 42 thereof). Is used to perform (learn) a dispersion parameter as the model parameter ⁇ , and the dispersion parameter stored in the model storage unit 33 is updated with the dispersion parameter obtained by the dispersion learning.
- the state variation learning unit 53 uses the state estimation result ⁇ of each device model #m supplied from the state estimation unit 32 (the estimation unit 42 thereof) to obtain the initial state parameter and the state variation parameter as the model parameter ⁇ .
- the obtained (updated) state variation learning is performed, and the initial state parameter and the state variation parameter stored in the model storage unit 33 are respectively updated with the initial state parameter and the state variation parameter obtained by the state variation learning. .
- the label acquisition unit 35 includes a state estimation result ⁇ of each device model #m supplied from the state estimation unit 32 (its estimation unit 42), the overall model ⁇ stored in the model storage unit 33, and a data output unit 36.
- the data output unit 36 stores the voltage waveform V supplied from the data acquisition unit 31, the state estimation result ⁇ of each device model #m supplied from the state estimation unit 32 (the estimation unit 42), and the model storage unit 33. Using the stored overall model, the power consumption U (m) of each device in the user's home represented by each device model #m (corresponding to each device model #m ) is obtained, and the device supplied from the label acquisition unit 35 Along with the label L (m) , it is provided to the user at the user's house.
- the data output unit 36 detects the operating state of each device in the user's house according to the state estimation result ⁇ of each device model #m, and sends an operating state label indicating the operating state from the communication unit 30 to the network 15. To the agent 14 via.
- the agent 14 can take a predetermined action according to the operating state (label) of each device at the user's home from the data output unit 36.
- FIG. 6 is a diagram for explaining the FHMM.
- the FHMM has a configuration having a plurality of HMMs.
- Each HMM that the FHMM has is called a factor.
- the m-th factor included in the FHMM is also referred to as factor #m.
- a combination of a plurality of states S (1) t to S (M) t at time t is a combination of the states of factor #m at time t (factor # 1 and factor # 2 State, ..., set of factor #M).
- FIG. 7 shows an FHMM with a factor number M of 3.
- one factor corresponds to one device (one factor corresponds to one device).
- factor #m corresponds to device #m.
- factor # 1 is in the state # 14 (indicated by a thick circle) among the four states # 11, # 12, # 13, and # 14. ),
- Factor # 2 is state # 21 (indicated by a bold circle) among the four states # 21, # 22, # 23, and # 24.
- factor # 3 is state # 33 (indicated by a thick circle) among the four states # 31, # 32, # 33, and # 34.
- the state of factor #m corresponds to the operating state of device #m corresponding to factor #m, for example.
- state # 11 corresponds to the off state of device # 1
- state # 14 corresponds to the so-called normal mode on state of device # 1.
- state # 12 corresponds to the so-called sleep mode on state of device # 1
- state # 13 represents the so-called energy saving mode on of device # 1. Corresponds to the state.
- W (2) # 21 is observed.
- FHMM a composite waveform that combines the unique waveforms observed in each factor is generated as an observed value observed in FHMM.
- the total (addition) of eigen waveforms can be employed.
- weighted addition of eigen waveforms or logical sum of eigen waveforms when the eigen waveform values are 0 and 1) can be used. The sum of is adopted.
- the device model #m constituting the overall model ⁇ corresponds to the factor #m.
- the number M of FHMM factors a value larger by a predetermined number as a margin than the maximum number of devices assumed to exist in the user's home is adopted.
- the FHMM as the overall model ⁇ , an FHMM in which each factor is in a state of 2 or 3 or more can be adopted.
- S t- 1) is the time t-1
- the state of the combination S t-1 Niite, at time t represents the transition probability of transition to a combination of state S t.
- S t ) can be calculated as follows.
- a column vector of rows (a matrix of K rows and 1 column).
- S t ⁇ 1 ) can be calculated according to equation (3).
- S t ) can be calculated according to equation (4).
- the dash (') represents transposition
- the superscript -1 represents the reciprocal (inverse matrix).
- represents the absolute value (determinant) of C (determinant operation).
- the current sensor 13 uses a timing at which a zero crossing occurs when the voltage changes from a negative value to a positive value as a timing at which the phase of the current is 0, at a predetermined sampling interval, for one cycle. (In Japan, 1/50 or 1/60 second) is sampled, and a column vector having the sampled value as a component is output as a current waveform Y t for one time.
- the current waveform Y t is a column vector of D rows.
- the observed value Y t follows a normal distribution with an average value (average vector) of ⁇ t and a variance (covariance matrix) of C.
- the average value ⁇ t is a D-row column vector similar to the current waveform Y t
- the variance C is a D-row D-column matrix (a diagonal component is a variance matrix).
- the average value ⁇ t is expressed by Expression (5) using the natural waveform W (m) described in FIG.
- eigen waveform of state #k with factor #m is expressed as W (m) k
- eigen waveform W (m) k of state #k with factor #m is, for example, current waveform Y t and It is a column vector of similar D rows.
- the characteristic waveform W (m) is the characteristic waveform W (m) 1 , W (m) 2 , ..., W (m ) A matrix of D rows and K columns having a column vector that is a collection of K and a natural waveform W (m) k of state #k with factor #m as a component of the kth column.
- S * (m) t represents the state of factor #m at time t, and hereinafter, S * (m) t is also referred to as the current state of factor #m at time t.
- Current state S * (m) t factors #m of time t, for example, as shown in equation (6), a component of only one row of the K line 1, the other rows component is 0 It is a column vector of K rows.
- the column vector S * (m) t of K rows as the current state S * (m) t of factor #m at time t is k rows Only the eye component is set to 1, and the other components are set to 0.
- the total sum of the characteristic waveforms W (m) k of the state #k of each factor #m at time t is obtained as the average value ⁇ t of the current waveform Y t at time t.
- the model parameters ⁇ of the FHMM are the initial state probability P (S (m) 1 ) in equation (2), the transition probability P (S (m) t
- S (m) t-1 ) in equation ( 3 ), 4) dispersion C and the characteristic waveform W (m) ( W (m) 1 , W (m) 2 ,..., W (m) K ) of equation (5), and the model of FIG.
- the learning unit 34 obtains model parameters ⁇ of these FHMMs.
- the waveform separation learning unit 51 obtains the unique waveform W (m) as a current waveform parameter by waveform separation learning.
- the distributed C is obtained as a distributed parameter by distributed learning.
- the state variation learning unit 53 obtains the initial state probability P (S (m) 1 ) and the transition probability P (S (m) t
- the HMM status is expressed when the operating status (combination) of 20 devices is expressed by a normal HMM.
- Is 2 20 1,048,576, and the number of transition probabilities is 1,099,511,627,776 of the square.
- the number of FHMM states total number
- FHMM learning that is, initial state probability P (S (m) 1 ), transition probability P (S (m) t
- the update of the waveform W (m) can be performed according to an EM (Expectation-Maximization) algorithm as described in Document A, for example.
- ⁇ ) of the conditional complete data log likelihood is a new model parameter when the complete data ⁇ S t , Y t ⁇ is observed under the model parameter ⁇ .
- ⁇ new it means the expected value of log likelihood log (P ( ⁇ S t , Y t ⁇
- ⁇ ) of the conditional complete data log likelihood of the equation (7) is obtained (the value corresponding to), and in the processing of the M step of the EM algorithm, A new model parameter ⁇ new is obtained that increases the expected value Q ( ⁇ new
- step S11 the model learning unit 34 has an initial state probability P (S (m) 1 ) and a transition probability P (S (m) t
- the k-th row component of the column vector of K rows as the initial state probability P (S (m) 1 ), that is, the k-th initial state probability ⁇ (m) k of the factor #m is, for example, 1 Initialized to / K.
- the i-th row and j-th column components (i, j 1,2,..., K) of the K-row and K-column matrix as the transition probability P (S (m) t
- S (m) t ⁇ 1 ), That is, in the factor #m, the transition probability P (m) i, j for transition from the i-th state #i to the j-th state #j is expressed by the expression P (m) i, 1 + P using a random number, for example. (m) i, 2 +... + P (m) Initialized to satisfy i, K 1.
- the matrix of D rows and D columns as the variance C is initialized to a diagonal matrix of D rows and D columns, for example, by setting a diagonal component using random numbers and setting other components (components) to 0.
- the column vector of the kth column of the matrix of D rows and K columns as the eigen waveform W (m) that is, each component of the column vector of the D row as the eigen waveform W (m) k of the state #k with factor #m
- initialization is performed using a random number.
- the voltage waveform from the data acquisition unit 31 is used for calculation of power consumption.
- step S13 the state estimation unit 32 performs an E step process using the measurement waveforms Y 1 to Y T from the data acquisition unit 31, and the process proceeds to step S14.
- step S13 the state estimation unit 32 uses the measured waveforms Y 1 to Y T from the data acquisition unit 31 to state probability of being in each state of each factor #m of the FHMM stored in the model storage unit 33. Etc., and the state estimation result of the state estimation is supplied to the model learning unit 34 and the data output unit 36.
- step S14 the model learning unit 34 performs an M step process using the measurement waveforms Y 1 to Y T from the data acquisition unit 31 and the state estimation result from the state estimation unit 32. Proceed to S15.
- step S14 the model learning unit 34 uses the measured waveforms Y 1 to Y T from the data acquisition unit 31 and the state estimation result from the state estimation unit 32 to store the FHMM stored in the model storage unit 33.
- step S15 If it is determined in step S15 that the convergence condition of the model parameter ⁇ is not satisfied, the process returns to step S13, and the same process is repeated thereafter.
- step S15 If it is determined in step S15 that the convergence condition of the model parameter ⁇ is satisfied, the learning process ends.
- steps S12 to S15 are repeated regularly or irregularly.
- FIG. 9 is a flowchart for explaining the processing of step E performed by the device separation apparatus 16 in FIG. 5 in step S13 in FIG.
- the estimation unit 42 includes the observation probability P (Y t
- j and initial state probability ⁇ (m) ) are used to observe the measured waveforms Y 1 , Y 2 ,..., Y t , and at time t, the state combination (state of factor # 1 at time t)
- the combination of the state of factor # 2 and the combination of the state of factor #M) determines the forward probability ⁇ t, z in z , and the process proceeds to step S23.
- the product of k is adopted.
- step S23 the estimation unit 42 determines the observation probability P (Y t
- the combination of the states z Niite then measured waveform Y t obtains a Y t + 1, ⁇ ⁇ ⁇ , backward probability observing Y T beta t, z, the process, step Proceed to S24.
- step S24 the estimation unit 42 is in the state combination z at time t in the FHMM as the overall model ⁇ according to the equation (8) using the forward probability ⁇ t, z and the backward probability ⁇ t, z .
- the posterior probability ⁇ t, z is obtained, and the process proceeds to step S25.
- ⁇ of the denominator of the right side of the equation (8) represents the summation of Nikki changing the w, all combinations S t of possible states at time t.
- the posterior probability ⁇ t, z is obtained by calculating the product ⁇ t, z ⁇ t, z of the forward probability ⁇ t, z and the backward probability ⁇ t, z as the product ⁇ t, w ⁇ t. , of w, obtained by normalizing the sum ⁇ t, w ⁇ t, w of the combination W ⁇ S t of possible states is FHMM.
- step S25 the estimation unit 42 uses the posterior probability ⁇ t, z and, at the factor #m, at the time t, the posterior probability ⁇ S (m) t > in the state S (m) t and the time t In factor #m, we are in state S (m) t , and in other factor #n, we find the posterior probability ⁇ S (m) t S (n) t '> in state S (n) t. The process proceeds to step S26.
- Equation (9) the posterior probability ⁇ S (m) t > is obtained according to Equation (9).
- Equation (9) the posterior probability ⁇ S (m) t > in state S (m) t at time t in factor #m is the posterior probability ⁇ t in state combination z at time t. , z is obtained by marginalizing a state combination z not including the state of factor #m.
- the posterior probability ⁇ S (m) t > is, for example, the state probability (posterior probability) of the kth state among the K states of factor #m at time t as the component of the kth row. It is a column vector of K rows.
- the posterior probability ⁇ S (m) t S (n) t ′> is, for example, the state probability (at time t) in the state #k of factor #m and the state #k ′ of other factor #n ( (A posteriori probability) is a matrix of K rows and K columns with the components of k rows and k 'columns.
- step S26 the estimation unit 42 calculates the forward probability ⁇ t, z , the backward probability ⁇ t, z , the transition probability P (z
- Equation (11) the posterior probability ⁇ S (m) t ⁇ 1 S (m) t ′> is obtained according to Equation (11).
- w) of transition from the state combination w to the state combination z is expressed by the equation (11) 3) According to 3), the transition probability P (1) i (1 ) from the state #i (1) of the factor # 1 constituting the state combination w to the state #j (1) of the factor # 1 constituting the state combination z ), j (1) , transition probability P (2) from state #i (2) of factor # 2 constituting state combination w to state #j (2) of factor # 2 constituting state combination z i (2), j (2 ), ⁇ , and factors #M constituting the combination z from one state factor #M constituting the combination w state #i (M) state #j (M ) Transition probability P (M) i (M), j (M) product P (1) i (1), j (1) ⁇ P (2) i (2), j (2) ⁇ XP (M) Obtained as i (M), j
- the posterior probability ⁇ S (m) t-1 S (m) t '> is, for example, the state probability of being in state #i at time t-1 and in state j at the next time t in factor #m.
- This is a matrix of K rows and K columns with (posterior probability) as a component of i row and j column.
- FIG. 10 illustrates the relationship between the FHMM forward probability ⁇ t, z and backward probability ⁇ t, z and the (normal) HMM forward probability ⁇ t, i and backward probability ⁇ t, j.
- FIG. 10 illustrates the relationship between the FHMM forward probability ⁇ t, z and backward probability ⁇ t, z and the (normal) HMM forward probability ⁇ t, i and backward probability ⁇ t, j.
- an HMM equivalent to that FHMM can be configured.
- An HMM equivalent to a certain FHMM has a state corresponding to a state combination z of each factor of the FHMM.
- the forward probability ⁇ t, z and the backward probability ⁇ t, z of the FHMM coincide with the forward probability ⁇ t, i and the backward probability ⁇ t, j of the HMM equivalent to the FHMM.
- 10A shows an FHMM composed of factors # 1 and # 2 having two states # 1 and # 2.
- FIG. 10B shows an HMM equivalent to the FHMM of FIG.
- the HMM in FIG. 10B corresponds to four combinations [1,1], [1,2], [2,1], and [2,2] of the states of the FHMM in FIG. It has two states # (1,1), # (1,2), # (2,1), and # (2,2).
- the backward probability ⁇ t, z ⁇ t, [1,1] , ⁇ t, [1,2] , ⁇ t, [2,1] , ⁇ t, [2 of the FHMM in FIG. , 2]
- FIG. 11 is a flowchart for explaining the process of the M step performed by the device separating apparatus 16 of FIG. 5 in step S14 of FIG.
- the waveform separation learning unit 51 obtains an updated value W (m) new of the natural waveform W (m) by calculating Expression (12) as waveform separation learning.
- W new is a D-row and K-column matrix
- the updated value W (m) new of the characteristic waveform W (m) of factor #m is arranged from left to right in the order of factor (index) #m. It is a matrix of D rows and K x M columns.
- the column vector of the (m-1) K + k columns of the eigen waveform (updated value) W new that is a matrix of D rows and K ⁇ M columns is the eigen waveform W (m) k ( Update value).
- ⁇ S t '> is a K ⁇ M column vector in which the posterior probability ⁇ S (m) t >, which is a column vector of K rows, is transposed from top to bottom in order of factor #m A row vector of columns.
- the (m-1) K + k column component of the posterior probability ⁇ S t '>, which is a row vector of K ⁇ M columns, has the state probability of being in state #k with factor #m at time t. .
- ⁇ S t S t '> is a matrix of K rows and K columns, the posterior probabilities ⁇ S (m) t S (n) t '> are arranged in the order of factor #m from top to bottom, and factor # A matrix of K ⁇ M rows and K ⁇ M columns arranged in order from left to right in n order.
- the component of the (m-1) K + k row (n-1) K + k 'column of the posterior probability ⁇ S t S t '> which is a matrix of K ⁇ M rows and K ⁇ M columns is at time t. It is the state probability of being in state #k with factor #m and state #k ′ with other factor #n.
- the superscript asterisk (*) represents an inverse matrix or a pseudo inverse matrix.
- the measurement waveform Y t is separated into the natural waveform W (m) .
- the distributed learning unit 52 calculates the updated value C new of the distributed C by calculating Expression (13) as distributed learning.
- step S33 the state variation learning unit 53 uses the posterior probabilities ⁇ S (m) t > and ⁇ S (m) t ⁇ 1 S (m) t ′> from the estimation unit 42 to perform state variation learning.
- the updated value P (m) i, j new of the transition probability P (m) i, j and the updated value ⁇ (m) new of the initial state probability ⁇ (m) and the updated value P (m) i, j new and ⁇ (m) new are used to update the transition probability P (m) i, j and initial state probability ⁇ (m) stored in the model storage unit 33, and M Return from step processing.
- the state fluctuation learning unit 53 calculates the expressions (14) and (15) as the state fluctuation learning, thereby obtaining the updated values P (m) i of the transition probabilities P (m) i and j , respectively. , j new and the updated value ⁇ (m) new of the initial state probability ⁇ (m) are obtained.
- ⁇ S (m) t-1, i S (m) t, j > is the i of the posterior probability ⁇ S (m) t-1 S (m) t '> which is a matrix of K rows and K columns. This is a component in the row j column, and represents the state probability of being in the state #i at the time t-1 and in the state #j at the next time t in the factor #m.
- ⁇ S (m) t-1, i > is a component of the i -th row of column vector posterior probability ⁇ S (m) t-1 > of K row, and at time t-1, state # of factor #m i represents the state probability.
- ⁇ (m) ( ⁇ (m) new ) is K with the initial state probability ⁇ (m) k (updated value ⁇ (m) k new ) of state #k with factor #m as the component in the k-th row A column vector of rows.
- FIG. 12 is a flowchart for explaining an example of an information presentation process for presenting information on the device #m performed by the device separation device 16 (FIG. 5).
- step S41 the data output unit 36 (voltage waveform corresponding to the current waveform Y t) the voltage waveform from the data acquisition unit 31 V t, the posterior probability of the state estimation result from the state estimation unit 32 ⁇ S (m)
- the power consumption U (m) of each factor #m is obtained using t > and the natural waveform W (m) stored in the model storage unit 33, and the process proceeds to step S42.
- the data output unit 36 using the voltage waveform V t at time t, the current consumption A t of instrument #m corresponding to factor #m time t, corresponding to a factor #m time t equipment Obtain the power consumption U (m) of #m.
- the current consumption At of the device #m corresponding to the factor #m at the time t is obtained as follows.
- the data output unit 36 uses the device # corresponding to the factor #m of the time t as the characteristic waveform W (m) of the state #k having the maximum posterior probability ⁇ S (m) t > in the factor #m. calculated as the current consumption a t of m.
- the data output unit 36 uses, for example, a factor that uses a state probability of each state of factor #m at time t as a weight, which is a component of a posteriori probability ⁇ S (m) t > that is a column vector of K rows.
- Current consumption of device #m corresponding to factor #m at time t with the weighted addition value of eigen waveform W (m) 1 , W (m) 2 ,..., W (m) K in each state of #m Calculate as At.
- the characteristic waveform W (m) of the state #k having the maximum posterior probability ⁇ S (m) t > and the state probability of each state of the factor #m at the time t are used as weights.
- the weighted addition values of the characteristic waveforms W (m) 1 , W (m) 2 ,..., W (m) K in each state of factor #m are substantially the same.
- step S42 the label acquisition unit 35 acquires a device label L (m) that identifies a device #m represented by each device model #m, that is, a device #m corresponding to each factor #m of the FHMM, and outputs the data.
- the process proceeds to step S43.
- the label obtaining unit 35 for example, consumption and current A t of instrument #m corresponding to each factor #m obtained in the data output unit 36, from the power consumption U (m), the power consumption U (m)
- the usage time zone of the recognized device #m can be transmitted to the user's home from the communication unit 30 via the network 15 and displayed on the display device at the user's home to be presented to the user.
- the label obtaining unit 35 for example, and the current consumption A t which is presented to the user, the power consumption U (m), the name of the device that applies the use time period, asked to enter a user, the network 15, and a communication
- the name of the device input by the user can be acquired as the device label L (m) .
- the label acquisition unit 35 prepares in advance a database in which attributes such as power consumption, current waveform (current consumption), usage time zone, and the like are registered in association with the names of devices for various devices. ; then, in the database, is recognized consumption and current a t of instrument #m corresponding to each factor #m sought data output unit 36, from the power consumption U (m), the power consumption U (m)
- the name of the device associated with the usage time zone of the device #m can be acquired as the device label L (m) .
- the label acquisition unit 35 has already acquired the device label L (m), and for the factor #m corresponding to the device #m supplied to the data output unit 36, the processing in step S42 may be skipped. it can.
- step S43 the data output unit 36 sends the power consumption U (m) of the device corresponding to each factor #m together with the device label L (m) of the factor #m from the communication unit 30 via the network 15.
- the information presenting process is completed by presenting it to the user by transmitting it to the home and displaying it on the display device at the user's home.
- FIG. 13 is a diagram showing a display example of the power consumption U (m) displayed on the display device at the user's home in the information presentation process of FIG.
- the time series of the power consumption U (m) of the device #m corresponding to each factor #m is displayed together with the device label L (m) such as the name of the device #m. Can be displayed.
- the device separation device 16 can determine the power consumption of each device in the user's home simply by measuring the total current consumed by each device in the user's home at one location such as the distribution board 12. Therefore, it is possible to easily “visualize” the power consumption of each device in the user's home from both the cost and labor aspects.
- the awareness of power saving at the user's home can be increased.
- the device separation device 16 collects the power consumption of each device in many user homes, and estimates the usage time zone of the device and thus the life pattern from the power consumption of the devices in each user home, which is useful for marketing and the like. be able to.
- the device separation device 16 by using the current waveform Y t measured by the current sensor 13, in performing the current consumption or the like device separating seeking of each device of the user's home, and change operational state of the device, the device It is necessary that the current consumption changes.
- the operating state of the device changes depending on, for example, a user's operation. However, since it depends only on the user's operation for changing the operating state of the device, it takes time to separate the devices.
- the device name input by the user is acquired as the device label L (m) , so that the user is given the name of the device to be the device label L (m). It is necessary to have the user input, and the user may feel annoyed.
- the agent 14 and the device separation device 16 operate in a coordinated manner, so that the operating state of the device is quickly changed, and the device separation is performed quickly. Yes.
- the agent 14 and the device separation device 16 operate in cooperation so that the user does not have to input the device label L (m) (name of the device to be).
- the device label L (m) of the device corresponding to the factor #m is acquired and associated with the factor #m.
- FIG. 14 is a perspective view showing an external configuration example of the agent 14 of FIG.
- the agent 14 is a pet-type robot imitating a dog, and is roughly divided into a body unit 1, leg units 102 ⁇ / b> A, 102 ⁇ / b> B, 102 ⁇ / b> C, 102 ⁇ / b> D, a head unit 103, and a tail. Part unit 104.
- Leg units 102A, 102B, 102C, 102D corresponding to the legs are connected to the front, rear, left and right of the body unit 101 corresponding to the body, respectively, and the front end portion and the rear end portion of the body unit 101 are respectively
- a head unit 103 corresponding to the head and a tail unit 104 corresponding to the tail are connected.
- a back sensor 101 ⁇ / b> A is provided on the upper surface of the trunk unit 101.
- the head unit 103 is provided with a head sensor 103A at the top and a chin sensor 103B at the bottom. Note that each of the back sensor 101A, the head sensor 103A, and the chin sensor 103B is configured by a pressure sensor, and detects pressure applied to the part.
- the tail unit 104 is attached to the body unit 101 so as to be swingable in the horizontal direction and the vertical direction.
- FIG. 15 is a block diagram showing an example of the internal configuration of the agent 14 of FIG.
- the body unit 101 includes a control unit 111, an A / D (Analog / Digital) conversion unit 112, a D / A conversion unit 113, a communication unit 114, a semiconductor memory 115, a back sensor 101A, and the like. Stored.
- a / D Analog / Digital
- the control unit 111 controls the entire agent 14 and performs various processes.
- the A / D conversion unit 112 converts the analog signals output from the microphone 121, the CCD cameras 122L and 122R, the back sensor 101A, the head sensor 103A, and the chin sensor 103B into digital signals, and sends them to the control unit 111. Supply.
- the D / A converter 113 converts the digital signal supplied from the controller 111 into an analog signal by D / A conversion, and supplies the analog signal to the speaker 123.
- the communication unit 114 communicates with the outside wirelessly or by wire. That is, the communication unit 114 receives data transmitted from the outside and supplies the data to the control unit 111. In addition, the communication unit 114 transmits data supplied from the control unit 111 to the outside.
- the semiconductor memory 115 can be configured to be detachable from a slot (not shown) provided in the body unit 101.
- the back sensor 101A is provided in a part corresponding to the back of the agent 14 in the torso unit 101.
- the back sensor 101A detects a pressure from the user applied thereto and outputs a pressure detection signal corresponding to the pressure to the A / D.
- the data is output to the control unit 111 via the conversion unit 112.
- the body unit 101 also stores, for example, a battery (not shown) serving as a power source for the agent 14, a circuit for detecting the remaining battery capacity, and the like.
- a battery (not shown) serving as a power source for the agent 14, a circuit for detecting the remaining battery capacity, and the like.
- the head sensor 103A and the chin sensor 103B corresponding to a tactile sensation that senses a pressure applied by touching the user or the like, for example, are provided at corresponding portions, respectively.
- a speaker 123 corresponding to the “mouth” of the agent 14 is installed, for example, in a corresponding part.
- the joint portions of the leg units 102A to 102D, the connection portions of the leg units 102A to 102D and the body unit 101, the connection portions of the head unit 103 and the body unit 101, and the tail unit 104 and the body Actuators are installed at the connecting parts of the unit units 101.
- the actuator operates each unit based on an instruction from the control unit 111. That is, for example, the leg units 102A to 102D are moved by the actuator, whereby the robot walks.
- the microphone 121 installed in the head unit 103 collects surrounding sounds (sounds) including utterances from the user, and sends the obtained audio signals to the control unit 111 via the A / D conversion unit 112. Output.
- the image sensors 122 ⁇ / b> L and 122 ⁇ / b> R capture the surrounding situation and output the obtained image signal to the control unit 111 via the A / D conversion unit 112.
- the head sensor 103A provided at the upper part of the head unit 103 and the chin sensor 103B provided at the lower part of the head unit 103 were received by physical actions such as “blow” and “slap” from the user, for example.
- the pressure is detected, and the detection result is output as a pressure detection signal to the control unit 111 via the A / D conversion unit 112.
- control unit 111 includes an action determination unit 131, a recognition unit 132, a position detection unit 133, an operation control unit 134, a notification control unit 135, and a table generation unit 136.
- the action determining unit 131 is, for example, a voice signal, an image signal, or a pressure given from the microphone 121, the image sensors 122L and 122R, the back sensor 101A, the head sensor 103A, and the chin sensor 103B via the A / D conversion unit 112. Based on the detection signal or the like, the surrounding situation, the instruction from the user, the presence / absence of the user's action, and the like are determined, and the action to be taken next by the agent 14 is determined based on the determination result. And the action determination part 131 drives a required actuator based on the determination result of action, and, thereby, the head unit 103 can be shaken up and down, right and left, and the tail part unit 104 can be moved.
- the action determining unit 131 drives the leg units 102A to 102D to take actions such as walking the agent 14 and operating devices at the user's home.
- the behavior determination unit 131 generates a synthesized sound based on the determination result of the behavior, and supplies the synthesized sound to the speaker 123 via the D / A conversion unit 113 for output, or the “eye” of the agent 14.
- the LED Light Emitting Diode (not shown) provided at the position “” is turned on, turned off, or blinked.
- the action determination unit 131 causes the agent 14 to autonomously take action based on the surrounding situation, the user who is in contact with the user, and the like.
- the recognition unit 132 recognizes the operation target device based on the image signal of the operation target device supplied from the image sensors 122L and 122R, and identifies the operation target device such as the model name and model number. Is acquired as the device label of the operation target device.
- the recognition unit 132 recognizes the operating state of the operation target device based on the image signal of the operation target device supplied from the image sensors 122L and 122R, for example, and acquires an operation state label indicating the operation state.
- the recognition unit 132 when the recognition unit 132 recognizes that the operation target device is powered off or on as the operating state of the operation target device, the recognition unit 132 indicates that the power is off or on, respectively. Is obtained by generating a character string “power off” or “power on” representing the operating state label.
- the recognition unit 132 recognizes that the operation mode of the operation target device is the power saving mode as the operation state of the operation target device, the character string “power saving mode” indicating the power saving mode is displayed. ”Is generated as an operating state label.
- the operation target device and the operation state of the operation target device can be recognized by, for example, listening to the user about the operation target device and the operation state of the operation target device.
- the agent 14 asks the user about the operation target device or the operation state of the operation target device by synthetic sound or the like, and recognizes the answer by the user's voice to the question.
- the operating state of the operation target device can be recognized.
- the voice recognition result of the answer by the user's voice can be adopted as the device label or the operating state label.
- the position detection unit 133 detects the position of the operation target device using, for example, GPS (Global Positioning System) and outputs position information indicating the position.
- GPS Global Positioning System
- an absolute three-dimensional coordinate system can be defined, and the coordinates of the three-dimensional coordinate system can be adopted as the position information of the operation target device.
- the agent 14 obtains a sketch of the user's house by creating a sketch of the user's house and the coordinates of the coordinate system with reference to a predetermined position on the sketch or a place on the sketch (for example, a living room, a bedroom, etc.) Etc.) can be employed as the position information of the operation target device.
- the operation control unit 134 controls the operation of the operation target device by the agent 14.
- the operation control unit 134 determines what operation is to be performed on the operation target device, and requests the action determination unit 131 to perform the operation.
- the behavior determination unit 131 determines the behavior of the agent 14 and drives a necessary actuator so that an operation according to the request of the operation control unit 134 is performed.
- the notification control unit 135 acquires an operating state label that is transmitted from the device separating apparatus 16 and indicates the operating state of the device at the user's home, and responds to the operating state label. To control the operation status of the device.
- the device separation device 16 transmits an operation state label indicating the operation state of the user's home device to the agent 14 via the network 15 as necessary.
- the notification control unit 135 transmits the operation state label from the device separation device 16 via the communication unit 114.
- the behavior determining unit 131 is requested to acquire and notify the operating state of the device according to the operating state label.
- the behavior determination unit 131 determines, for example, to output the operation state of the device by voice, generates a synthesized sound for informing the operation state of the device, and generates a D / A conversion unit 113. Through the speaker 123.
- the table generation unit 136 associates the device label of the device (operation target device) obtained by the recognition unit 132 with the position information of the device obtained by the position detection unit 133 and registers it in the device table stored in the semiconductor memory 115. To do.
- the agent 14 operates the operation target device under the control of the operation control unit 134, and the recognition unit 132 recognizes the operation state of the operation target device after the operation, and displays the operation state label of the operation state. If acquired, the operating state label is transmitted from the communication unit 114 to the device separating apparatus 16 via the network 15.
- the operation control unit 134 searches, for example, from the server on the Internet what operating state the operation target device can take as its operation state, and from among the operation states that the operation target device can take The operation state to be taken by the operation target device can be selected as the instruction operation state. In this case, the operation control unit 134 can control the operation of the operation target device so as to be in the instruction operation state. Furthermore, in this case, the recognition unit 132 can recognize the instruction operation state as the operation state of the operation target device.
- FIG. 16 is a diagram showing an example of a device table stored in the semiconductor memory 115 of FIG.
- the agent 14 autonomously moves in the user's home, and the recognition unit 132 receives the image signal supplied from the image sensors 122L and 122R to the control unit 111 via the A / D conversion unit 112 during the movement. Based on the above, the device is recognized, and the device is selected as a candidate device that is a candidate for the operation target device.
- the recognition unit 132 acquires the device label of the candidate device, and causes the position information detection unit 133 to detect the position information of the candidate device. Then, the recognizing unit 132 determines whether the set of the device label and the position information of the candidate device is registered in the device table.
- the recognition unit 132 When the set of the device label and the position information of the candidate device is registered in the device table, the recognition unit 132 recognizes that the candidate device has already been selected as the operation target device. Then, the recognizing unit 132 reselects another device as a candidate device, and thereafter repeats the same processing.
- the recognition unit 132 selects the candidate device as the operation target device and controls the table generation unit 136.
- the device label of the operation target device (which is a candidate device) is associated with the position information and registered in the device table stored in the semiconductor memory 115.
- FIG. 17 is a block diagram illustrating a configuration example of the label acquisition unit 35 in FIG. 5 when the agent 14 and the device separating apparatus 16 operate in cooperation.
- the label acquisition unit 35 includes acquisition units 201 and 202, a labeling unit 203, a correspondence storage unit 204, and a control unit 205.
- the device information of the operation target device is information related to the operation target device, and in this embodiment, the device information includes a device label of the operation target device, position information, and an operating state label.
- each operation of each device is performed using the current waveform Y t as the sum data regarding the sum of the currents consumed by all devices in the user's house.
- the current consumption of each device is separated (the inherent waveform W ( m) k is a current consumption in operating state of the apparatus #m corresponding to the state #k factors #m, is separated from the current waveform Y t), it can be said that.
- the updating of the specific waveform W (m) k as pattern information the specific waveform W (m) k, consumed by the device #m consumption current representing its own waveform W (m) k corresponds to the factor #m
- the state probability ⁇ S (m) t > as possibility information representing the possibility of being generated is obtained using the current waveform Y t and is based on the state probability ⁇ S (m) t > as the possibility information
- the characteristic waveform W (m) k as the pattern information is updated.
- the acquisition unit 202 performs state probability ⁇ S (m) t > as possibility information obtained by the state estimation unit 32 for the characteristic waveform W (m) k (having the state #k) in the device separation as described above. Is acquired from the state estimation unit 32.
- the labeling unit 203 registers the correspondence information obtained as a result of labeling in the correspondence table stored in the correspondence storage unit 204.
- the correspondence relationship storage unit 204 stores a correspondence relationship table.
- control unit 205 monitors the labeling unit 203 and, based on the monitoring result, communicates with the communication unit 30 (FIG. 5) and the agent 14 (the communication unit 114) via the network 15. , Exchange necessary messages.
- FIG. 18 is a diagram showing an example of a correspondence table stored in the correspondence storage unit 204 of FIG.
- the device label, the position information, the factor #m corresponding to the device (factor number #m as information specifying the device), and the state map are associated with the device at the user's home.
- the state #k of factor #m corresponding to device #m (state number #k as information for specifying) and the operation state label of the operation state of device #m are associated with each other.
- the device label of the association information and the factor #m it is possible to recognize the factor #m corresponding to the device represented by the device label and to recognize the device label of the device corresponding to the factor #m.
- the device recognizes the factor #m state #k corresponding to the operation state of the device indicated by the device label, and the device corresponding to the factor #m state #k ( The operating label of the operating state of the device corresponding to the factor #m can be recognized.
- the factor #m state #k corresponding to the device #m is associated with the operating state label of the operating state of the device #m. It is possible to recognize the correspondence with the operating state (the operating state label) of the device corresponding to the factor #m.
- the factor #m state #k is because it has a unique waveform W (m) k as pattern information, a state map, the specific waveform W (m) k as the pattern information, the inherent waveform W (m ) It can be said that the operating state label of the operating state of the device #m that consumes the current represented by k is labeled.
- the state map in which the operation state label as described above is labeled with the unique waveform W (m) k (having the state #k) as the pattern information is associated with the device label. Therefore, the characteristic waveform W (m) k as the pattern information is labeled with the device label of the device #m that consumes the current represented by the specific waveform W (m) k. .
- the device waveform of the device that consumes the current represented by the waveform W (m) k and the operation status label of the operation status of the device are included in the waveform W (m) k as the pattern information. Can be said to be labeled.
- the device label of the device corresponding to the factor #m can be recognized. Furthermore, according to the association information, it is possible to recognize the operating state label of the operating state (of the device #m) corresponding to the state #k of the factor #m.
- FIG. 19 is a flowchart showing an example of processing performed by the agent 14 in FIG. 15 as labeling processing for registering the association information in the correspondence table.
- step S101 the recognition unit 132 detects, for example, a device reflected in the visual sense of the agent 14 based on an image signal supplied from the image sensors 122L and 122R to the control unit 111 via the A / D conversion unit 112 ( And any one of the devices is selected as a candidate device, and the process proceeds to step S102.
- step S102 the recognition unit 132 generates (acquires) the device label of the candidate device, the position information detection unit 133 detects the position information of the candidate device, and the process proceeds to step S103.
- step S103 the recognition unit 132 determines whether the candidate device is not registered in the device table (FIG. 16) based on the set of the device label and the position information of the candidate device.
- step S102 If it is determined in step S102 that the candidate device is not registered, that is, if the set of the device label and the position information of the candidate device is already registered in the device table, the process returns to step S101. Thereafter, the same processing is repeated.
- step S102 If it is determined in step S102 that the candidate device is not registered, that is, if the set of the device label and the position information of the candidate device is not registered in the device table, the process proceeds to step S104.
- the recognition unit 132 selects the candidate device as the operation target device and controls the table generation unit 136 to associate the device label of the operation target device with the position information, and stores the device label in the semiconductor memory 115. Register in the device table.
- step S104 the recognizing unit 132 controls the communication unit 114 to transmit the device label and position information of the operation target device to the device separating apparatus 16, and the process proceeds to step S105.
- the device label and the position information of the operation target device transmitted from the communication unit 114 of the agent 14 are received by the communication unit 30 (FIG. 5) of the device separation apparatus 16 via the network 15, and the label acquisition unit 35. (FIG. 17).
- step S105 the operation control unit 134 waits for the RESULT: READY message to be transmitted from the device separating apparatus 16, and then causes the agent 14 to operate the device to be operated in a predetermined operating state (for example, an operation in which the power is on).
- the operation target device is operated so as to be in the state), and the process proceeds to step S106.
- step S106 the recognition unit 132 recognizes the operation state of the operation target device, generates (acquires) an operation state label indicating the operation state, and the process proceeds to step S107.
- the operating state label of the operation target device transmitted from the communication unit 114 of the agent 14 is received by the communication unit 30 (FIG. 5) of the device separating apparatus 16 via the network 15, and the label acquisition unit 35 (FIG. 5). 17).
- step S108 the operation control unit 134 determines whether a RESULT: FINISHED message has been received from the device separation device 16.
- the RESULT: FINISHED message is a message indicating that the label acquisition unit 35 (FIG. 17) of the device separating apparatus 16 has registered the association information of the operation target device in the correspondence table (FIG. 18).
- the data is transmitted from the control unit 205 of the acquisition unit 35 to the agent 14 via the communication unit 30 (FIG. 5) and the network 15.
- step S108 it is determined whether the RESULT: FINISHED message as described above is transmitted from the device separation device 16 and received by the communication unit 114 (FIG. 15).
- the RESULT: MORE message is a message requesting to change the operation state of the operation target device to another operation state, and is transmitted from the control unit 205 of the label acquisition unit 35 (FIG. 17) to the communication unit 30 (FIG. 5) and sent to the agent 14 via the network 15.
- step S109 it is determined whether or not the RESULT: MORE message as described above is transmitted from the device separation device 16 and received by the communication unit 114 (FIG. 15).
- step S109 If it is determined in step S109 that the RESULT: MORE message has not been received, the process returns to step S108, and the same process is repeated thereafter.
- step S109 determines whether the RESULT: MORE message has been received. If it is determined in step S109 that the RESULT: MORE message has been received, the process proceeds to step S110, and the operation control unit 134 causes the agent 14 to change the operation target device to an operation state different from the current operation state. For example, the operation target device is operated so that the power source is switched from the on-operation state to the off-operation state, and the process proceeds to step S111.
- step S111 the recognition unit 132 recognizes the operation state of the operation target device after the operation by the agent in step S110, generates (acquires) an operation state label indicating the operation state, Proceed to S112.
- step S112 the recognizing unit 132 controls the communication unit 114 to cause the device separating apparatus 16 to transmit the operation state label of the operation state of the operation target device after the operation by the agent in step S110 is performed. Returns to step S108, and the same processing is repeated thereafter.
- step S108 When it is determined in step S108 that the RESULT: FINISHED message has been received, that is, in the label acquisition unit 35 of the device separating apparatus 16, the association information of the operation target device is registered in the correspondence table (FIG. 18). In the case where it is found, the process returns to step S101, and the same process is repeated thereafter.
- FIG. 20 is a flowchart showing an example of processing performed by the label acquisition unit 35 (FIG. 17) of the device separation apparatus 16 as labeling processing for registering the association information in the correspondence table.
- step S121 the acquisition unit 201 waits for the device label and position information of the operation target device to be transmitted from the agent 14, and acquires the device label and position information.
- the agent 14 transmits the device label and position information of the operation target device in step S104 of FIG. Since the device label and position information of the operation target device from the agent 14 are received by the communication unit 30 of the device separation device 16 (FIG. 5), the acquisition unit 201 receives the device of the operation target device received by the communication unit 30. The label and position information are acquired and supplied to the labeling unit 203.
- step S122 the acquisition unit 201 waits for the operation state label of the operation state of the operation target device to be transmitted from the agent 14, and acquires the operation state label.
- the agent 14 transmits an operation state label indicating the operation state of the operation target device in steps S107 and S122 of FIG. Since the operation state label of the operation state of the operation target device from the agent 14 is received by the communication unit 30 of the device separation device 16 (FIG. 5), the acquisition unit 201 receives the operation target device received by the communication unit 30. The operating state label of the operating state is acquired and supplied to the labeling unit 203.
- step S122 the acquisition unit 202 obtains each factor obtained by device separation using the current waveform Y t of the user's home when the operation target device is in the current operating state.
- State probabilities (posterior probabilities) ⁇ S (m) t > as possibility information about the characteristic waveform W (m) k as pattern information of each state #k of #m are obtained from the state estimation unit 32 (FIG. 5). The information is acquired and supplied to the labeling unit 203, and the process proceeds to step S124.
- step S124 the labeling unit 203, based on the state probability ⁇ S (m) t > from the acquisition unit 202, of the characteristic waveform W (m) k as the pattern information of each state #k of each factor #m.
- the target specific waveform W (m) k representing the current consumption consumed in the current operating state of the operation target device is identified from the inside, and the process proceeds to step S125.
- step S124 for example, the eigen waveforms W (1) 1 , W (1) 2 ,..., W (1 ) possessed by all states # 1 to #k of all factors # 1 to #M, respectively.
- the natural waveform can be identified as the target natural waveform W (m) k .
- the specific waveform with the lowest state probability ⁇ S (m) t > is selected as the target specific waveform W (m) k Can be specified.
- target-specific waveform W (m) k of state probabilities ⁇ S (m) t> i.e., the factor #m with object inherent waveform W (m) k states #k state
- the probability that the target specific waveform W (m) k represents the current consumption in the current operating state of the operation target device is determined.
- the target is unique waveform W (m) k state probability of ⁇ S (m) t>, if it is less than 1.0 the probability of more than a threshold value close to 1.0, certainly the object inherent waveform W (m) k It is determined to likely target specific waveform W (m) k state probability of ⁇ S (m) t>, if not a probability more than a threshold value, it is determined that the target-specific waveform W (m) k is not like certain.
- step S125 If it is determined in step S125 that the target specific waveform W (m) k is not accurate, the process proceeds to step S126, and the control unit 205 controls the operation target device to change the current consumption of the operation target device.
- a RESULT: MORE message for requesting to change the operating state to another operating state is generated and transmitted to the agent 14 from the communication unit 30 (FIG. 5). And a process returns to step S122 from step S126, and the same process is repeated hereafter.
- step S125 if it is determined in step S125 that the target specific waveform W (m) k is likely, the process proceeds to step S127, and the labeling unit 203 acquires the target specific waveform W (m) k in the acquisition unit 201.
- the association information (FIG. 18) that associates the device label and the position information of the operation target device with the operation status label is generated.
- the labeling unit 203 sets the factor #m having the state #k as the target specific waveform W (m) k as the corresponding factor #m corresponding to the operation target device, and the corresponding factor #m (factor number #m). ) And the device label and position information of the operation target device are associated with each other and included in the association information of the operation target device.
- the labeling unit 203 sets the corresponding factor #m state #k having the target specific waveform W (m) k as the corresponding state #k corresponding to the current operating state of the operation target device, and In the attachment information state map (FIG. 18), the correspondence state #k (the state number #k) and the operation state label of the current operation state of the operation target device are registered in association with each other.
- step S128 the control unit 205 generates a RESULT: FINISHED message indicating that the association information of the operation target device has been registered in the correspondence table, and transmits the RESULT: FINISHED message to the agent 14 from the communication unit 30 (FIG. 5). And a process returns to step S121 from step S128, and the same process is repeated hereafter.
- the agent 14 at the user's home operates the operation target device and recognizes the operating state of the operation target device. Then, the device label representing the operation target device and the operation state label representing the operation state of the operation target device are transmitted from the agent 14 to the device separation apparatus 16.
- the device separation device 16 obtains the operation target device label and the operation state label from the agent 14, and the state probability as possibility information obtained in device separation using the current waveform Y of the user's house ⁇ S (m) t > is obtained. Then, in the device separation device 16, the unique waveform as the pattern information indicating the current consumption consumed in the current operating state of the operation target device based on the state probability ⁇ S (m) t > as the possibility information. W (m) k is specified, and the device label and the operating state label of the operation target device are associated with the specific waveform W (m) k as the pattern information.
- the agent 14 transmits a device label representing the device and an operation state label representing the operation state of the device to the device separation device 16, and the device separation device 16 uses the unique waveform W (m) k as pattern information. based on the state probability of potentially information indicating the possibility of being consumed current consumption represents ⁇ S (m) t>, the specific waveform W (m) k as pattern information, equipment label and health labels Therefore, it is possible to present which device in the user's home is in what operating state by the device label and the operating state label.
- a person recognizes the device that consumes the current consumption represented by the pattern information associated with the device label and the operating state label, and the operating state of the device.
- the operating state of the operation target device can be changed autonomously by the agent 14 in addition to being changed according to the RESULT: MORE message transmitted from the device separating apparatus 16 to the agent 14. That is, the agent 14 can take an action of operating the operation target device autonomously so as to change the operating state of the operation target device.
- the data output unit 36 includes an acquisition unit 211, a detection target storage unit 212, and an operating state detection unit 213.
- the operating state detection request message transmitted by the agent 14 includes the device label of the detection target device.
- the operating state detection unit 213 selects the association information including the device label stored in the detection target storage unit 212 from the correspondence table as the target association information, and the factor # included in the target association information A state probability ⁇ S (m) t > as possibility information of each state #k of m is acquired from the state estimation unit 32.
- the operating state detection unit 213 displays the operating state label associated with the state #k having the highest state probability ⁇ S (m) t > in the state map included in the target association information.
- the target operating state label representing the current operating state is transmitted from the communication unit 30 (FIG. 5) to the agent 14 together with the device label of the detection target device.
- FIG. 22 is a flowchart illustrating an example of a process performed by the agent 14 of FIG. 15 as an operation state notification process for notifying the user of the operation state of the user's home device.
- step S141 the notification control unit 135 determines a detection target device (device to be detected) from devices whose device labels are registered in the device table (FIG. 16) stored in the semiconductor memory 115.
- the agent 14 cannot recognize the operating state from the current location (for example, from the current location, the image sensor).
- One or more of the devices that cannot be photographed by 122L and 122R can be determined as detection target devices.
- the notification control unit 135 After determining the detection target device, the notification control unit 135 includes, from the communication unit 114, a device separation device that includes an operation state detection request message that includes the device label of the detection target device and requests detection of the operation state of the detection target device. 16 to transmit.
- the MONITOR: TV # 1 message that includes the device label and requests detection of the operation state of the detection target device is sent as an operation state detection request message Is done.
- the notification control unit 135 After transmitting the operating state detection request message, the notification control unit 135 waits for a RESULT: FINISHED message indicating that the reception of the request for detecting the operating state of the detection target device has been completed from the device separation device 16. Then, the RESULT: FINISHED message is acquired, and the process proceeds from step S141 to step S142.
- the device separating apparatus 16 when the device separating apparatus 16 receives the operation state detection request message from the agent 14, it transmits a RESULT: FINISHED message to the agent 14.
- the communication unit 114 receives the RESULT: FINISHED message from the device separation device 16 and supplies it to the notification control unit 135. Therefore, the notification control unit 135 is supplied from the communication unit 114 as described above.
- the RESULT: FINISHED message from the agent 14 is acquired.
- step S142 the notification control unit 135 waits for the device label and the operation state label of the detection target device to be transmitted from the device separation device 16, and acquires the device label and the operation state label. The process proceeds to step S143.
- the device separation device 16 detects the operating state of the detection target device and transmits the operating state label of the operating state together with the device label of the detection target device.
- the communication unit 114 receives the device label and the operating state label of the detection target device from the device separation device 16 and supplies them to the notification control unit 135. The device label and the operating state label of the detection target device supplied from 114 are acquired.
- step S143 the notification control unit 135 notifies the user of the operating state of the device to be detected of the device label acquired in step S142 according to the operating state label acquired in step S142, and the process proceeds to step S144. move on.
- the notification control unit 135 causes the agent 14 to output a message indicating that the device to be detected of the device label is in the operating state represented by the operating state label by the synthesized sound.
- the notification control unit 135 communicates with a display device among the devices at the user's home via the communication unit 114, and the device targeted for detection of the device label is in the operation state represented by the operation state label. Can be output by voice or displayed on the screen.
- the notification control unit 135 communicates via the communication unit 114 so that the user has a message indicating that the device label detection target device is in the operating state indicated by the operating state label.
- a mobile terminal such as a smartphone, it is possible to notify the user of the operating state of the detection target device.
- step S144 the notification control unit 135 displays a RESULT: CHANGE message indicating that the operation state of the detection target device has changed, the device label of the detection target device, and the operation state before and after the change from the device separation device 16. Waiting for the operating status label to be transmitted, the RESULT: CHANGE message, the device label and the operating status label are acquired, and the process proceeds to step S145.
- the device separation device 16 detects the operating state of the detection target device, and when the operating state changes, along with the RESULT: CHANGE message, the device label of the detection target device and the operating state indicating the operating state before and after the change.
- Send a label the agent 14, the communication unit 114 receives the RESULT: CHANGE message from the device separation device 16, the device label of the detection target device, and the operation state label indicating the operation state before and after the change, and the notification control unit 135. Therefore, the notification control unit 135 supplies the RESULT: CHANGE message, the device label of the detection target device, and the operation state label indicating the operation state before and after the change, which are supplied from the communication unit 114 in this manner. To get.
- step S145 the notification control unit 135 notifies the user of the operation state (change) of the detection target device of the device label acquired in step S144 according to the operation state label acquired in step S144. finish.
- the device label of the detection target device is “TV # 1” representing a certain TV
- the operation state label indicating the operation state before the change indicates that the power is turned off
- the agent 14 If the operating state label indicating the operating state indicates that the power is on, the agent 14 generates a message such as “TV # 1 has been turned on”, for example. Then, by outputting as described in step S143, the operating state (change) of the detection target device can be notified to the user.
- the device separation device 16 can also transmit the position information of the detection target device together with the device label and the operation state label of the detection target device to the agent 14.
- the agent 14 can generate a message notifying the user of the operating state (change) of the detection target device using the position information of the detection target device.
- the device label of the detection target device is “TV # 1” representing a certain TV
- the operation state label indicating the operation state before the change indicates that the power is turned off
- the location information of the detection target device recognizes that the detection target device is installed in the living room of the user's house
- the agent 14 can generate a message such as “Someone has started watching TV # 1 in the living room” using the position information of the device to be detected.
- the device label of the detection target device is a label representing a certain illumination
- the operation state label representing the operation state before the change represents that the illumination is turned off, and represents the operation state after the change. Recognizing that the lighting as the detection target device is installed at the entrance of the user's house from the position information of the detection target device when the operating state label indicates that the lighting is on.
- the agent 14 can generate a message such as “Entrance electricity (lighting) has been turned on” using the position information of the detection target device.
- FIG. 23 is a flowchart showing an example of processing performed by the data output unit 36 (FIG. 21) of the device separation apparatus 16 as the operation state notification processing for notifying the user of the operation state of the device at the user's home.
- step S151 the acquisition unit 211 waits for an operating state detection request message for requesting detection of the operating state of the detection target device in step S141 of FIG.
- the request message is acquired, and the process proceeds to step S152.
- the acquisition unit 211 receives the operation state detection request message received by the communication unit 30. get.
- step S152 the acquisition unit 211 supplies the device label of the detection target device included in the operation state detection request message to the detection target storage unit 212 and registers it in the detection list stored in the detection target storage unit 212.
- the acquisition unit 211 generates a RESULT: FINISHED message indicating that the reception of the request for detecting the operating state of the detection target device is completed, and transmits the message to the agent 14 from the communication unit 30 (FIG. 5).
- the process proceeds from step S152 to step S153.
- step S153 the operating state detection unit 213 is obtained by device separation using the current waveform Y t of the user's home including the current consumption of the detection target device whose device label is registered in the detection list of the detection target storage unit 212.
- State probabilities posterior probabilities
- the process proceeds to step S154.
- step S154 the operating state detection unit 213 refers to the correspondence relationship table (FIG. 18) stored in the correspondence relationship storage unit 204, and calculates the factor #m associated with the device label of the detection target device. Recognized as factor #m corresponding to.
- the operating state detection unit 213 refers to the correspondence table (FIG. 18) stored in the correspondence storage unit 204, and among the state probabilities ⁇ S (m) t > acquired in step S153, the detection target device. Is detected based on the state probability ⁇ S (m) t > of each state #k of factor #m corresponding to.
- the operating state detection unit 213 uses the factor #m state #k corresponding to the detection target device in the state map associated with the device label of the detection target device in the correspondence table stored in the correspondence storage unit 204.
- the operation state indicated by the operation state label associated with the state #k having the highest state probability ⁇ S (m) t > is detected as the operation state of the detection target device.
- step S154 proceeds from step S154 to step S155, and the operation state detection unit 213 receives the device label of the detection target device and the operation state label (the operation state label of the operation state detected in step S154) as the communication unit 30.
- the agent 14 is transmitted from (FIG. 5), and the process proceeds to step S156.
- step S155 the position information of the detection target device registered in the correspondence table (FIG. 18) in association with the device label of the detection target device is also provided together with the device label of the detection target device and the operation state label. , Can be sent to the agent 14.
- step S156 the operating state detection unit 213 waits for the device separation using the current waveform Y t at the next time t in the user's house, including the current consumption of the detection target device, for example.
- the state probability (posterior probability) ⁇ S (m) t > of each state #k of each factor #m obtained by the separation is acquired from the state estimation unit 32 (FIG. 5), and the process proceeds to step S157.
- step S157 the operation state detection unit 213 detects the current operation state of the detection target device based on the state probability ⁇ S (m) t > in the same manner as in step S154, and the process proceeds to step S158.
- step S158 If it is determined in step S158 that the operating state of the detection target device has not changed, the process returns to step S156, and the same process is repeated thereafter.
- step S158 If it is determined in step S158 that the operation state of the detection target device has changed, the process proceeds to step S159, and the operation state detection unit 213 indicates that the operation state of the detection target device has changed.
- the message, the device label of the detection target device, and the operation state label indicating the operation state before and after the change of the detection target device are transmitted from the communication unit 30 (FIG. 5) to the agent 14, and the process proceeds to step S160. move on.
- step S160 the operating state detection unit 213 deletes the device label of the detection target device that has transmitted the device label to the agent 14 from the detection list in the detection target storage unit 212, and the process ends.
- step S153 to step S160 is performed when the plurality of device labels (of the detection target device) are registered in the detection list of the detection target storage unit 212. Done for each.
- the device separation device 16 detects the operating state of the detection target device based on the state probability ⁇ S (m) t > as the possibility information, and determines the device label and the operating state label of the detection target device.
- the agent 14 acquires the device label and the operation state label of the detection target device from the device separation apparatus 16 and transmits the operation state of the detection target device indicated by the device label according to the operation state label. For example, the agent 14 notifies the operating status of a device that cannot recognize the operating status from the current location (for example, a device that cannot be imaged by the image sensors 122L and 122R from the current location). It can be recognized on the spot and notified to the user.
- the agent 14 determines all devices in the user's home (all devices whose device labels are registered in the device table (FIG. 16)) as detection target devices, so that the agent 14 The operating state of the device can be recognized in real time.
- the agent 14 is notified of the operating state of the detection target device according to the operating state label of the detection target device obtained from the device separation device 16.
- an action corresponding to the operation state of the detection target device can be taken.
- the agent 14 recognizes that the user has returned home and welcomes the user to the entrance. You can take action to go to.
- the movable agent 14 is used as a control device for controlling the device at the user's home.
- a control device for controlling the device in addition to a movable robot such as the agent 14, A device (for example, a home network server) that can control the power on / off of the device and the setting (change) of the operation mode by wireless or wired communication via a home network or the like can be employed. .
- the inherent waveform W (m) k obtained for the inherent waveform W (m) k representing the current consumption in each operating state of each device.
- the current consumption of each device is separated by obtaining (updating ) the natural waveform W (m) k
- equipment separation was adopted, other equipment separation using any possibility information that represents the possibility that the current consumption represented by the pattern information other than the state probability ⁇ S (m) t > is consumed is adopted. be able to.
- an error between the predicted value of the total data (current waveform Y t ) obtained using the current consumption represented by the pattern information and the actual total data is adopted, and based on the error, Device separation that separates the current consumption of each device can be employed by updating the pattern information so as to reduce the error.
- the series of processing of the agent 14 and the device separation device 16 described above can be performed by hardware or can be performed by software.
- a program constituting the software is installed in a computer or the like.
- FIG. 24 shows a configuration example of an embodiment of a computer in which a program for executing the series of processes described above is installed.
- the program can be recorded in advance in a hard disk 305 or ROM 303 as a recording medium built in the computer.
- the program can be stored (recorded) in a removable recording medium 311.
- a removable recording medium 311 can be provided as so-called package software.
- examples of the removable recording medium 311 include a flexible disk, a CD-ROM (Compact Disc Read Only Memory), an MO (Magneto Optical) disc, a DVD (Digital Versatile Disc), a magnetic disc, a semiconductor memory, and the like.
- the computer includes a CPU (Central Processing Unit) 302, and an input / output interface 310 is connected to the CPU 302 via the bus 301.
- a CPU Central Processing Unit
- an input / output interface 310 is connected to the CPU 302 via the bus 301.
- the CPU 302 executes a program stored in a ROM (Read Only Memory) 303 accordingly. .
- the CPU 302 loads a program stored in the hard disk 305 to a RAM (Random Access Memory) 304 and executes it.
- the CPU 302 performs processing according to the flowchart described above or processing performed by the configuration of the block diagram described above. Then, the CPU 302 causes the processing result to be output from the output unit 306 or transmitted from the communication unit 308 via the input / output interface 310, or recorded on the hard disk 305, for example, as necessary.
- the input unit 307 includes a keyboard, a mouse, a microphone, and the like.
- the output unit 306 includes an LCD (Liquid Crystal Display), a speaker, and the like.
- the processing performed by the computer according to the program does not necessarily have to be performed in chronological order in the order described as the flowchart. That is, the processing performed by the computer according to the program includes processing executed in parallel or individually (for example, parallel processing or object processing).
- the program may be processed by one computer (processor), or may be distributedly processed by a plurality of computers. Furthermore, the program may be transferred to a remote computer and executed.
- the system means a set of a plurality of components (devices, modules (parts), etc.), and it does not matter whether all the components are in the same housing. Accordingly, a plurality of devices housed in separate housings and connected via a network and a single device housing a plurality of modules in one housing are all systems. .
- the present technology can take a cloud computing configuration in which one function is shared by a plurality of devices via a network and is jointly processed.
- each step described in the above flowchart can be executed by one device or can be shared by a plurality of devices.
- the plurality of processes included in the one step can be executed by being shared by a plurality of apparatuses in addition to being executed by one apparatus.
- this technique can take the following structures.
- a possibility information acquisition unit for acquiring the possibility information obtained in device separation for separating the current consumption of each device; Based on the possibility information, the pattern information indicating the current consumption consumed in the current operating state of the device represented by the device label is specified, and the device label and the operating state label are associated with the pattern information
- An information processing apparatus comprising: a labeling unit that performs labeling. ⁇ 2> Based on the possibility information, the pattern information determines the probability of representing the current consumption of the current operating state of the device represented by the device label, and if the pattern information is not certain, the control device The information processing apparatus according to ⁇ 1>, wherein the operation state of the device represented by the device label is requested to be changed to another operation state.
- ⁇ 3> Based on the possibility information, further comprising an operating state detection unit that detects the current operating state of the device, The information according to ⁇ 1> or ⁇ 2>, wherein the device label and the operation state label associated with the pattern information representing the current consumption consumed in the current operation state of the device are transmitted to the control device. Processing equipment. ⁇ 4> The device information acquisition unit also acquires position information indicating the position of the device, The labeling unit associates the position information with the pattern information together with the device label and the operation state label, The position information is also transmitted to the control device together with the device label and the operation state label associated with the pattern information indicating the current consumption consumed in the current operation state of the device. Information processing device.
- the FHMM is A unique waveform unique to each state of each factor, used to determine the average of the observed values of the summation data observed in the combination of states of each factor; The variance of the observed value of the sum data observed in the combination of states of each factor; The initial state probability that the state of each factor is the initial state, and The transition probability that the state of each factor transitions is as a model parameter, In the FHMM learning, As the pattern information, waveform separation learning for obtaining the unique waveform; Distributed learning to determine the variance; The information processing device according to ⁇ 5>, wherein the state variation learning for obtaining the initial state probability and the transition probability is performed.
- the pattern information that represents the current consumption in each operating state of each device, and that the current consumption represented by the pattern information may be consumed.
- An information processing method comprising the steps of: ⁇ 9> Operate the device, Recognizing the operating state of the device, From the control device that transmits the device label that represents the device and the operation state label that represents the operation state of the device, the device information acquisition unit that acquires the device label and the operation state label, Using the sum data on the sum of the current consumed by multiple devices, it is possible to obtain the pattern information that represents the current consumption in each operating state of each device, and that the current consumption represented by the pattern information may be
- a possibility information acquisition unit for acquiring the possibility information obtained in device separation for separating the current consumption of each device; Based on the possibility information, the pattern information indicating the current consumption consumed in the current operating state of the device represented by the device label is specified, and the device label and the operating state label are associated with the pattern information
- the device separation device that performs device separation that separates the current consumption of each device, the device label representing the device, and the operating state of the device are represented.
- a control device comprising: a communication unit that transmits an operating state label. ⁇ 11> The control device according to ⁇ 10>, wherein the control device is a movable agent. ⁇ 12> The control device according to ⁇ 11>, wherein the operation control unit controls the operation of the device so as to change the operation state of the device to another operation state in response to a request from the device separation device.
- an operation state label representing an operation state of the device and a device label representing the device obtained based on the possibility information are acquired, and the device label is obtained according to the operation state label.
- the control apparatus in any one of ⁇ 10> thru
- the recognition unit also recognizes the position of the device,
- the communication unit also transmits position information indicating the position of the device to the device separating apparatus, The control device according to ⁇ 13>, wherein the notification control unit controls notification of an operation state of the device represented by the device label according to the operation state label and the position information.
- ⁇ 15> The control unit according to any one of ⁇ 10> to ⁇ 14>, wherein the recognition unit recognizes the device and an operating state of the device by listening to a user.
- ⁇ 16> A step of operating the device; Recognizing the operating state of the device; Using the sum data on the sum of the current consumed by multiple devices, it is possible to obtain the pattern information that represents the current consumption in each operating state of each device, and that the current consumption represented by the pattern information may be consumed.
- the device separation device that performs device separation that separates the current consumption of each device, the device label representing the device, and the operating state of the device are represented. Sending a health label.
- An operation control unit for controlling the operation of the device;
- a recognition unit for recognizing the operating state of the device;
- Using the sum data on the sum of the current consumed by multiple devices it is possible to obtain the pattern information that represents the current consumption in each operating state of each device, and that the current consumption represented by the pattern information may be consumed.
- the device separation device that performs device separation that separates the current consumption of each device, the device label representing the device, and the operating state of the device are represented.
- a program that causes a computer to function as a communication unit that transmits operating status labels.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Automation & Control Theory (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Selective Calling Equipment (AREA)
Abstract
Description
機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得する機器情報取得部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得する可能性情報取得部と、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うラベル付け部と
を備える情報処理装置。
<2>
前記可能性情報に基づいて、前記パターン情報が、前記機器ラベルが表す機器の現在の稼働状態の消費電流を表している確からしさを判定し、前記パターン情報が確からしくない場合、前記制御装置に、前記機器ラベルが表す機器の稼働状態を、他の稼働状態に変更することを要求する
<1>に記載の情報処理装置。
<3>
前記可能性情報に基づいて、前記機器の現在の稼働状態を検知する稼働状態検知部をさらに備え、
前記機器の現在の稼働状態において消費されている消費電流を表すパターン情報に対応付けられた前記機器ラベル及び前記稼働状態ラベルを、前記制御装置に送信する
<1>又は<2>に記載の情報処理装置。
<4>
前記機器情報取得部は、前記機器の位置を表す位置情報も取得し、
前記ラベル付け部は、前記パターン情報に、前記機器ラベル及び前記稼働状態ラベルとともに、前記位置情報も対応付け、
前記機器の現在の稼働状態において消費されている消費電流を表すパターン情報に対応付けられた前記機器ラベル及び前記稼働状態ラベルとともに、前記位置情報も、前記制御装置に送信する
<3>に記載の情報処理装置。
<5>
前記機器分離では、
前記総和データを用いて、FHMM(Factorial Hidden Markov Model)の各ファクタの状態にいる状態確率を、前記可能性情報として求める状態推定を行い、
前記状態確率を用いて、前記FHMMの学習を行う
<1>ないし<4>のいずれかに記載の情報処理装置。
<6>
前記FHMMは、
各ファクタの状態の組み合わせにおいて観測される前記総和データの観測値の平均値を求めるのに用いられる、各ファクタの各状態に固有の固有波形と、
各ファクタの状態の組み合わせにおいて観測される前記総和データの観測値の分散と、
各ファクタの状態が初期状態である初期状態確率と、
各ファクタの状態が遷移する遷移確率と
を、モデルパラメータとして有し、
前記FHMMの学習では、
前記パターン情報として、前記固有波形を求める波形分離学習と、
前記分散を求める分散学習と、
前記初期状態確率と、前記遷移確率とを求める状態変動学習と
を行う
<5>に記載の情報処理装置。
<7>
前記状態推定では、
前記平均値、及び、前記分散を用いて、各ファクタの状態の組み合わせにおいて、前記総和データが観測される観測確率を求め、
前記観測確率、及び、前記遷移確率を用いて、前記総和データの系列Y1,Y2,・・・,YTについて、前記総和データY1,Y2,・・・,Ytを観測し、時刻tに、各ファクタの状態の組み合わせzにいる前向き確率αt,zと、時刻tに、各ファクタの状態の組み合わせzにいて、その後、前記総和データYt,Yt+1,・・・,YTを観測する後ろ向き確率βt,zとを求め、
前記前向き確率αt,z、及び、前記後ろ向き確率βt,zを用いて、時刻tに、各ファクタの状態の組み合わせzにいる事後確率γt,zを求め、
前記事後確率γt,zを周辺化することにより、前記状態確率を求める
<6>に記載の情報処理装置。
<8>
機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得するステップと、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得するステップと、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うステップと
を含む情報処理方法。
<9>
機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得する機器情報取得部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得する可能性情報取得部と、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うラベル付け部と
して、コンピュータを機能させるためのプログラム。
<10>
機器の操作を制御する操作制御部と、
前記機器の稼働状態を認識する認識部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する通信部と
を備える制御装置。
<11>
移動可能なエージェントである
<10>に記載の制御装置。
<12>
前記操作制御部は、前記機器分離装置からの要求に応じて、前記機器の稼働状態を、他の稼働状態に変更するように、前記機器の操作を制御する
<11>に記載の制御装置。
<13>
前記機器分離装置において、前記可能性情報に基づいて得られる、前記機器の稼働状態を表す稼働状態ラベル、及び、前記機器を表す機器ラベルを取得し、その稼働状態ラベルに応じて、前記機器ラベルが表す前記機器の稼働状態の報知を制御する報知制御部をさらに備える
<10>ないし<12>のいずれかに記載の制御装置。
<14>
前記認識部は、前記機器の位置も認識し、
前記通信部は、前記機器の位置を表す位置情報も、前記機器分離装置に送信し、
前記報知制御部は、前記稼働状態ラベル、及び、前記位置情報に応じて、前記機器ラベルが表す前記機器の稼働状態の報知を制御する
<13>に記載の制御装置。
<15>
前記認識部は、前記機器、及び、前記機器の稼働状態を、ユーザに聞くことで認識する
<10>ないし<14>のいずれかに記載の制御装置。
<16>
機器を操作するステップと、
前記機器の稼働状態を認識するステップと、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信するステップと
を含む制御方法。
<17>
機器の操作を制御する操作制御部と、
前記機器の稼働状態を認識する認識部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する通信部と
して、コンピュータを機能させるためのプログラム。
Claims (17)
- 機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得する機器情報取得部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得する可能性情報取得部と、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うラベル付け部と
を備える情報処理装置。 - 前記可能性情報に基づいて、前記パターン情報が、前記機器ラベルが表す機器の現在の稼働状態の消費電流を表している確からしさを判定し、前記パターン情報が確からしくない場合、前記制御装置に、前記機器ラベルが表す機器の稼働状態を、他の稼働状態に変更することを要求する
請求項1に記載の情報処理装置。 - 前記可能性情報に基づいて、前記機器の現在の稼働状態を検知する稼働状態検知部をさらに備え、
前記機器の現在の稼働状態において消費されている消費電流を表すパターン情報に対応付けられた前記機器ラベル及び前記稼働状態ラベルを、前記制御装置に送信する
請求項2に記載の情報処理装置。 - 前記機器情報取得部は、前記機器の位置を表す位置情報も取得し、
前記ラベル付け部は、前記パターン情報に、前記機器ラベル及び前記稼働状態ラベルとともに、前記位置情報も対応付け、
前記機器の現在の稼働状態において消費されている消費電流を表すパターン情報に対応付けられた前記機器ラベル及び前記稼働状態ラベルとともに、前記位置情報も、前記制御装置に送信する
請求項3に記載の情報処理装置。 - 前記機器分離では、
前記総和データを用いて、FHMM(Factorial Hidden Markov Model)の各ファクタの状態にいる状態確率を、前記可能性情報として求める状態推定を行い、
前記状態確率を用いて、前記FHMMの学習を行う
請求項3に記載の情報処理装置。 - 前記FHMMは、
各ファクタの状態の組み合わせにおいて観測される前記総和データの観測値の平均値を求めるのに用いられる、各ファクタの各状態に固有の固有波形と、
各ファクタの状態の組み合わせにおいて観測される前記総和データの観測値の分散と、
各ファクタの状態が初期状態である初期状態確率と、
各ファクタの状態が遷移する遷移確率と
を、モデルパラメータとして有し、
前記FHMMの学習では、
前記パターン情報として、前記固有波形を求める波形分離学習と、
前記分散を求める分散学習と、
前記初期状態確率と、前記遷移確率とを求める状態変動学習と
を行う
請求項5に記載の情報処理装置。 - 前記状態推定では、
前記平均値、及び、前記分散を用いて、各ファクタの状態の組み合わせにおいて、前記総和データが観測される観測確率を求め、
前記観測確率、及び、前記遷移確率を用いて、前記総和データの系列Y1,Y2,・・・,YTについて、前記総和データY1,Y2,・・・,Ytを観測し、時刻tに、各ファクタの状態の組み合わせzにいる前向き確率αt,zと、時刻tに、各ファクタの状態の組み合わせzにいて、その後、前記総和データYt,Yt+1,・・・,YTを観測する後ろ向き確率βt,zとを求め、
前記前向き確率αt,z、及び、前記後ろ向き確率βt,zを用いて、時刻tに、各ファクタの状態の組み合わせzにいる事後確率γt,zを求め、
前記事後確率γt,zを周辺化することにより、前記状態確率を求める
請求項6に記載の情報処理装置。 - 機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得するステップと、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得するステップと、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うステップと
を含む情報処理方法。 - 機器を操作し、
前記機器の稼働状態を認識し、
前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する
制御装置から、前記機器ラベル、及び、前記稼働状態ラベルを取得する機器情報取得部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離において得られる前記可能性情報を取得する可能性情報取得部と、
前記可能性情報に基づいて、前記機器ラベルが表す機器の現在の稼働状態において消費されている消費電流を表すパターン情報を特定し、そのパターン情報に、前記機器ラベルと前記稼働状態ラベルを対応付けるラベル付けを行うラベル付け部と
して、コンピュータを機能させるためのプログラム。 - 機器の操作を制御する操作制御部と、
前記機器の稼働状態を認識する認識部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する通信部と
を備える制御装置。 - 移動可能なエージェントである
請求項10に記載の制御装置。 - 前記操作制御部は、前記機器分離装置からの要求に応じて、前記機器の稼働状態を、他の稼働状態に変更するように、前記機器の操作を制御する
請求項11に記載の制御装置。 - 前記機器分離装置において、前記可能性情報に基づいて得られる、前記機器の稼働状態を表す稼働状態ラベル、及び、前記機器を表す機器ラベルを取得し、その稼働状態ラベルに応じて、前記機器ラベルが表す前記機器の稼働状態の報知を制御する報知制御部をさらに備える
請求項12に記載の制御装置。 - 前記認識部は、前記機器の位置も認識し、
前記通信部は、前記機器の位置を表す位置情報も、前記機器分離装置に送信し、
前記報知制御部は、前記稼働状態ラベル、及び、前記位置情報に応じて、前記機器ラベルが表す前記機器の稼働状態の報知を制御する
請求項13に記載の制御装置。 - 前記認識部は、前記機器、及び、前記機器の稼働状態を、ユーザに聞くことで認識する
請求項12に記載の制御装置。 - 機器を操作するステップと、
前記機器の稼働状態を認識するステップと、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信するステップと
を含む制御方法。 - 機器の操作を制御する操作制御部と、
前記機器の稼働状態を認識する認識部と、
複数の機器で消費されている電流の総和に関する総和データを用いて、各機器の各稼働状態の消費電流を表すパターン情報について得られる、そのパターン情報が表す消費電流が消費されている可能性を表す可能性情報に基づいて、前記パターン情報を更新することにより、各機器の消費電流を分離する機器分離を行う機器分離装置に、前記機器を表す機器ラベル、及び、前記機器の稼働状態を表す稼働状態ラベルを送信する通信部と
して、コンピュータを機能させるためのプログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016535873A JP6597612B2 (ja) | 2014-07-25 | 2015-07-10 | 制御装置、制御方法、情報処理装置、情報処理方法、及び、プログラム |
CN201580039393.XA CN106537369A (zh) | 2014-07-25 | 2015-07-10 | 控制装置、控制方法、信息处理装置、信息处理方法以及程序 |
EP15824919.3A EP3174004A4 (en) | 2014-07-25 | 2015-07-10 | Control device, control method, information processing device, information processing method, and program |
US15/327,173 US20170176967A1 (en) | 2014-07-25 | 2015-07-10 | Control apparatus, control method, information processing apparatus, information processing method, and program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014152407 | 2014-07-25 | ||
JP2014-152407 | 2014-07-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016013411A1 true WO2016013411A1 (ja) | 2016-01-28 |
Family
ID=55162940
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/069828 WO2016013411A1 (ja) | 2014-07-25 | 2015-07-10 | 制御装置、制御方法、情報処理装置、情報処理方法、及び、プログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US20170176967A1 (ja) |
EP (1) | EP3174004A4 (ja) |
JP (1) | JP6597612B2 (ja) |
CN (1) | CN106537369A (ja) |
WO (1) | WO2016013411A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018056890A (ja) * | 2016-09-30 | 2018-04-05 | Kddi株式会社 | 自動学習装置、方法、プログラム、自動学習システムおよび自動モニタ装置 |
JP2020514931A (ja) * | 2016-12-15 | 2020-05-21 | ビジェリー・インコーポレーテッド | 低頻度エネルギー分解技法の改善 |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013163460A1 (en) | 2012-04-25 | 2013-10-31 | Myenersave, Inc. | Energy disaggregation techniques for low resolution whole-house energy consumption data |
US11435772B2 (en) | 2014-09-04 | 2022-09-06 | Bidgely, Inc. | Systems and methods for optimizing energy usage using energy disaggregation data and time of use information |
US11126525B2 (en) * | 2015-09-09 | 2021-09-21 | Arris Enterprises Llc | In-home legacy device onboarding and privacy enhanced monitoring |
US10657110B2 (en) | 2017-01-23 | 2020-05-19 | Bidgely Inc. | Systems and methods for disaggregating appliance loads |
US20190050430A1 (en) * | 2017-01-23 | 2019-02-14 | Bidgely Inc. | Systems and Methods for Disaggregating Appliance Loads |
KR101976424B1 (ko) * | 2017-01-25 | 2019-05-09 | 엘지전자 주식회사 | 이동 로봇 |
US20200072887A1 (en) * | 2017-03-30 | 2020-03-05 | Mitsubishi Electric Corporation | Phase group estimation device, phase group estimation method, and recording medium |
JP7187397B2 (ja) * | 2019-07-18 | 2022-12-12 | オークマ株式会社 | 工作機械における診断モデルの再学習要否判定方法及び再学習要否判定装置、再学習要否判定プログラム |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070268509A1 (en) * | 2006-05-18 | 2007-11-22 | Xerox Corporation | Soft failure detection in a network of devices |
JP2012064023A (ja) * | 2010-09-16 | 2012-03-29 | Sony Corp | データ処理装置、データ処理方法、およびプログラム |
JP2012138123A (ja) * | 2012-04-16 | 2012-07-19 | Mitsubishi Electric Corp | 省エネ支援装置、省エネ支援システム |
JP2013044736A (ja) * | 2011-08-26 | 2013-03-04 | Mitsubishi Electric Corp | 稼働状況判別装置、稼働状況判別プログラム、稼働状況判別方法、波形パターン学習装置、波形パターン学習プログラム、及び波形パターン学習方法 |
JP2013045197A (ja) * | 2011-08-23 | 2013-03-04 | Sony Corp | 情報処理装置、情報処理方法、およびプログラム |
JP2013109742A (ja) * | 2011-11-21 | 2013-06-06 | Jiaotong Univ | ユーザーフィードバック情報に基づく電気器具検出方法とシステム |
WO2013094687A1 (ja) * | 2011-12-22 | 2013-06-27 | 大和ハウス工業株式会社 | エネルギー管理システム及び管理装置 |
JP2013213825A (ja) * | 2012-03-30 | 2013-10-17 | Infometis Co Ltd | 電気機器をモニタするための方法、及び、モニタ装置 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001058649A1 (fr) * | 2000-02-14 | 2001-08-16 | Sony Corporation | Systeme robotique, dispositif robotise et procede de controle d'un tel systeme et dispositif et procede de traitement de donnees |
WO2004106009A1 (ja) * | 2003-06-02 | 2004-12-09 | Matsushita Electric Industrial Co., Ltd. | 物品操作システムおよび方法、並びに物品管理システムおよび方法 |
JP2006155329A (ja) * | 2004-11-30 | 2006-06-15 | Toshiba Corp | 機器の制御方法及び装置 |
US9523718B2 (en) * | 2010-09-22 | 2016-12-20 | Philips Lighting Holding B.V. | Disaggregation apparatus for identifying an appliance in an electrical network |
US8812427B2 (en) * | 2011-04-27 | 2014-08-19 | Hewlett-Packard Development Company, L.P. | System and method for disaggregating power load |
MX353420B (es) * | 2012-03-30 | 2018-01-12 | Sony Corp | Aparato de procesamiento de datos, metodo y programa de procesamiento de datos. |
JP5670498B2 (ja) * | 2013-03-08 | 2015-02-18 | シャープ株式会社 | 機能管理装置、遠隔制御システム、機能管理装置の制御方法、制御プログラムおよび記録媒体 |
-
2015
- 2015-07-10 US US15/327,173 patent/US20170176967A1/en not_active Abandoned
- 2015-07-10 WO PCT/JP2015/069828 patent/WO2016013411A1/ja active Application Filing
- 2015-07-10 EP EP15824919.3A patent/EP3174004A4/en not_active Withdrawn
- 2015-07-10 CN CN201580039393.XA patent/CN106537369A/zh active Pending
- 2015-07-10 JP JP2016535873A patent/JP6597612B2/ja active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070268509A1 (en) * | 2006-05-18 | 2007-11-22 | Xerox Corporation | Soft failure detection in a network of devices |
JP2012064023A (ja) * | 2010-09-16 | 2012-03-29 | Sony Corp | データ処理装置、データ処理方法、およびプログラム |
JP2013218715A (ja) * | 2010-09-16 | 2013-10-24 | Infometis Co Ltd | 電気機器推定装置並びにその方法およびプログラム |
JP2013045197A (ja) * | 2011-08-23 | 2013-03-04 | Sony Corp | 情報処理装置、情報処理方法、およびプログラム |
JP2013044736A (ja) * | 2011-08-26 | 2013-03-04 | Mitsubishi Electric Corp | 稼働状況判別装置、稼働状況判別プログラム、稼働状況判別方法、波形パターン学習装置、波形パターン学習プログラム、及び波形パターン学習方法 |
JP2013109742A (ja) * | 2011-11-21 | 2013-06-06 | Jiaotong Univ | ユーザーフィードバック情報に基づく電気器具検出方法とシステム |
WO2013094687A1 (ja) * | 2011-12-22 | 2013-06-27 | 大和ハウス工業株式会社 | エネルギー管理システム及び管理装置 |
JP2013213825A (ja) * | 2012-03-30 | 2013-10-17 | Infometis Co Ltd | 電気機器をモニタするための方法、及び、モニタ装置 |
JP2012138123A (ja) * | 2012-04-16 | 2012-07-19 | Mitsubishi Electric Corp | 省エネ支援装置、省エネ支援システム |
Non-Patent Citations (1)
Title |
---|
See also references of EP3174004A4 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018056890A (ja) * | 2016-09-30 | 2018-04-05 | Kddi株式会社 | 自動学習装置、方法、プログラム、自動学習システムおよび自動モニタ装置 |
JP2020514931A (ja) * | 2016-12-15 | 2020-05-21 | ビジェリー・インコーポレーテッド | 低頻度エネルギー分解技法の改善 |
Also Published As
Publication number | Publication date |
---|---|
JP6597612B2 (ja) | 2019-10-30 |
US20170176967A1 (en) | 2017-06-22 |
EP3174004A1 (en) | 2017-05-31 |
EP3174004A4 (en) | 2018-04-25 |
CN106537369A (zh) | 2017-03-22 |
JPWO2016013411A1 (ja) | 2017-04-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6597612B2 (ja) | 制御装置、制御方法、情報処理装置、情報処理方法、及び、プログラム | |
JP5873864B2 (ja) | オブジェクト追跡及び認識方法及び装置 | |
US10803407B2 (en) | Method for selecting learned model corresponding to sensing data and provisioning selected learned model, and learned model provision device | |
US20180233145A1 (en) | Position calibration for intelligent assistant computing device | |
CN106923839B (zh) | 运动辅助装置、运动辅助方法以及记录介质 | |
KR102065801B1 (ko) | 데이터 처리 장치, 데이터 처리 방법, 및 프로그램 | |
JP6020880B2 (ja) | データ処理装置、データ処理方法、及び、プログラム | |
CA3137309A1 (en) | Systems and methods for predictive environmental fall risk identification | |
JP5720491B2 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
JP2013206273A (ja) | 情報処理装置、情報処理方法、および情報処理システム | |
EP3866685B1 (en) | Systems and methods for micro impulse radar detection of physiological information | |
JP2016170012A (ja) | 測位装置、測位方法、測位プログラム、および、測位システム | |
JP2020027370A (ja) | 最適化装置、シミュレーションシステム及び最適化方法 | |
JP6591232B2 (ja) | 連想メモリによって分類されたフレームを使用して位置に対する計量値を取得すること | |
JP2008269215A (ja) | 特異パターン検出システム、モデル学習装置、特異パターン検出方法、及び、コンピュータプログラム | |
US10382219B2 (en) | Method, apparatus, and non-transitory computer-readable recording medium storing program for estimating operator | |
Toris et al. | Temporal persistence modeling for object search | |
CN104977038B (zh) | 使用与关联存储器耦合的运动感测设备识别移动 | |
Singh et al. | Decision tree based mobile crowdsourcing for agriculture advisory system | |
JP7409486B2 (ja) | データ処理装置、システム、データ処理方法、およびプログラム | |
JP6000767B2 (ja) | 学習装置、判別装置、行動状態判別システム、および判別方法 | |
WO2015104804A1 (ja) | 信号処理システム、信号処理方法、及び信号処理プログラム | |
JP6045014B2 (ja) | 情報処理装置およびその方法、サーバ装置およびその制御方法、並びにプログラム | |
US20220040844A1 (en) | Information processing apparatus, information processing method and computer program | |
US20230030053A1 (en) | Information processing device, communication system, and generation method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15824919 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016535873 Country of ref document: JP Kind code of ref document: A |
|
REEP | Request for entry into the european phase |
Ref document number: 2015824919 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15327173 Country of ref document: US Ref document number: 2015824919 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |