Lighting management device, lighting management system, learning device and reasoning device

Abstract

The invention comprises: an illumination storage unit (12) for storing illumination collected by a moving body (2) and position information corresponding to the illumination, wherein the moving body (2) comprises an illumination detection unit (21) and moves in a room (4); a lighting control unit (14) for adjusting the output of a lighting device (3) installed in a facility according to the illumination; and a judgment unit (13) for judging whether the illumination is measured in one state or in another state, and when the judgment unit (13) judges that the illumination is measured in another state, the lighting control unit (14) adjusts the output of the lighting device (3) corresponding to the position information according to the illumination and the position information corresponding to the illumination, and when the judgment unit (13) judges that the illumination is measured in one state, the lighting control unit (14) does not adjust the output of the lighting device (3) corresponding to the position information according to the illumination and the position information corresponding to the illumination. Thus, illumination data in which the illumination detection unit (21) is blocked by obstacles or the like is excluded as inappropriate data, so that the illumination control of the lighting device (3) can be performed more appropriately.

Description

Technical Field

[0001] The invention relates to a lighting management device, a lighting management system, a learning device and an inference device. Background Art

[0002] As a conventional lighting management device, for example, in the robot device for environmental control disclosed in Patent Document 1, position information detected by the position detection device and measurement data of illumination measured by the sensor of the sensing unit are collected and recorded in a memory. The operation processing unit compares the minimum value with the reference value, and if the minimum value exceeds the reference value, it is determined that the illumination in the room is sufficiently ensured. That is, the operation processing unit determines that there is room for energy-saving operation, for example, sets an energy-saving mode to turn off the fluorescent lamp of the lighting device, and stores the setting information in the address. On the other hand, if the minimum value is not less than the reference value, the operation processing unit cancels the energy-saving mode, for example, by turning on the fluorescent lamp of the lighting device in order to maintain comfort.

[0003] On the other hand, the calculation processing unit obtains the surrounding illumination distribution information of the occupant based on the indoor environment information stored in the memory. The calculation processing unit obtains the central value of the surrounding illumination distribution based on the surrounding illumination distribution information. The calculation processing unit compares the central value with the reference value, and if the central value exceeds the reference value, it is determined that the surrounding illumination of the occupant is sufficiently ensured. That is, the calculation processing unit determines that there is a margin for energy-saving operation, for example, sets an energy-saving mode for turning off the fluorescent lamp of the lighting equipment, and stores the setting information in the address. That is, by setting the central value of the illumination as the comparison object, the influence of abnormal data that may become a human body shadow is reduced when measuring the illumination.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2007-183032 Summary of the invention

[0007] Problem that the invention aims to solve

[0008] In the robot device for environmental control disclosed in Patent Document 1, the central value of the surrounding illumination distribution information is set as a comparison object, thereby reducing the influence of abnormal data that may be a human body shadow when measuring illumination. That is, the abnormal data is not made a comparison object for comparison with the reference value. However, in this case, if the surrounding illumination distribution information contains a lot of data affected by obstacles such as human bodies, for example, if more than half of the data in the surrounding illumination distribution information is the above data, the central value of the surrounding illumination distribution information will be affected by the abnormal data, and there is a problem that appropriate illumination control cannot be performed.

[0009] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a lighting management device and a lighting management system that are less susceptible to obstacles such as human bodies and can perform more appropriate illumination control.

[0010] Means used to solve problems

[0011] The lighting management device of the present invention comprises: an illumination storage unit that stores illumination collected by a moving body and position information corresponding to the illumination, the moving body having an illumination detection unit and moving within a facility;

[0012] a lighting control unit that adjusts the output of lighting equipment installed in the facility according to the illumination; and

[0013] an inference device which determines whether the illuminance is measured in one state or in another state,

[0014] When the inference device determines that the illuminance is measured in another state, the lighting control unit adjusts the output of the lighting fixture corresponding to the position information based on the illuminance and the position information corresponding to the illuminance. When the inference device determines that the illuminance is measured in one state, the lighting control unit does not adjust the output of the lighting fixture corresponding to the position information based on the illuminance and the position information corresponding to the illuminance.

[0015] In addition, the lighting management system in the present invention comprises: a mobile body, which has an illuminance detection unit and moves within the facility; and a lighting management device, which has: an illuminance storage unit, which stores the illuminance collected by the mobile body and position information corresponding to the illuminance; a lighting control unit, which performs lighting control of lighting equipment installed in the facility based on the illuminance and the position information corresponding to the illuminance; and an inference device, which determines whether the illuminance is measured in one state or in another state.

[0016] When the inference device determines that the illuminance is measured in another state, the lighting control unit adjusts the output of the lighting fixture corresponding to the position information based on the illuminance and the position information corresponding to the illuminance. When the inference device determines that the illuminance is measured in one state, the lighting control unit does not adjust the output of the lighting fixture corresponding to the position information based on the illuminance and the position information corresponding to the illuminance.

[0017] In addition, the learning device in the present invention comprises: a data acquisition unit, which acquires learning data, the learning data including the illuminance collected by a moving body and position information corresponding to the illuminance, and whether the illuminance is measured in one state or in another state, the moving body having an illuminance detection unit and moving within a facility; and a model generation unit, which uses the learning data to generate a learned model, the learned model being used to infer whether the illuminance is measured in one state or in another state based on the illuminance and the position information corresponding to the illuminance.

[0018] In addition, the inference device in the present invention comprises: a data acquisition unit, which acquires the illuminance collected by a moving body and position information corresponding to the illuminance, and the moving body has an illuminance detection unit and moves within the facility; and an inference unit, which uses a learned model to output whether the illuminance is measured in one state or in another state based on the illuminance input from the data acquisition unit, and the learned model is used to infer whether the illuminance corresponding to the position information is measured in one state or in another state based on the illuminance and the position information corresponding to the illuminance.

[0019] Effects of the Invention

[0020] The lighting management device of the present invention excludes illuminance data detected when the illuminance sensor is blocked by an obstacle or the like as inappropriate data, and thus can more appropriately adjust the illuminance of the lighting fixture. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram of the lighting management system according to Embodiment 1 of the present invention.

[0022] Figure 2 This is a functional configuration diagram of the lighting management system according to Embodiment 1 of the present invention.

[0023] Figure 3 This is a graph showing the threshold values ​​of illumination in the lighting management system according to Embodiment 1 of the present invention.

[0024] Figure 4 This is a flowchart showing the operation of the lighting management system according to Embodiment 1 of the present invention.

[0025] Figure 5 This is a flowchart showing the operation of the lighting management system according to Embodiment 1 of the present invention.

[0026] Figure 6 This is a functional configuration diagram of a lighting management system according to Embodiment 2 of the present invention.

[0027] Figure 7 This is a flowchart showing the operation of the lighting management system according to Embodiment 2 of the present invention.

[0028] Figure 8 This is a functional configuration diagram of a lighting management system according to Embodiment 3 of the present invention.

[0029] Fig. 9 This is a flowchart showing the operation of the lighting management system according to Embodiment 3 of the present invention.

[0030] Fig.10 This is a functional configuration diagram of a lighting management system according to a fourth embodiment of the present invention.

[0031] Fig.11 This is a functional configuration diagram of a learning device of a lighting management device according to a fourth embodiment of the present invention.

[0032] Fig.12 This is a flowchart showing the operation of the learning device of the lighting management device according to Embodiment 4 of the present invention.

[0033] Fig.13 This is a functional configuration diagram of an inference device of a lighting management device according to a fourth embodiment of the present invention.

[0034] Fig.14 This is a flowchart showing the operation of the inference device of the lighting management device according to the fourth embodiment of the present invention.

[0035] Fig.15 This is a diagram showing a configuration example of a model generation unit of an inference device of a lighting management device according to Embodiment 4 of the present invention.

[0036] Fig.16 This is a hardware configuration diagram of the lighting management device according to Embodiments 1 to 4 of the present invention. DETAILED DESCRIPTION

[0037] Implementation Method 1

[0038] Below, use Figure 1 to Figure 5 The configuration and operation of the lighting management system according to this embodiment will be described. Figure 1 : This is a schematic diagram of the lighting management system of this embodiment. A window 5 that transmits sunlight is provided in the room 4. In the room 4, the brightness is maintained by the sunlight that transmits through the window 5 during a clear day. On the other hand, from dusk to night or in bad weather, there is no sunlight or the sunlight is insufficient. Therefore, in order to maintain the brightness in the room 4, it is necessary to illuminate the room 4 through the lighting device 3. The lighting device 3 is connected to the lighting management system, and the lighting control is performed by the lighting management system.

[0039] The lighting management system is composed of a lighting management device 1 and a moving body 2. The moving body 2 is, for example, a robot that moves autonomously. The moving body 2 is, for example, a mobile device that moves autonomously. The moving body 2 moves autonomously in a room 4. The moving body 2 has an illuminance detection unit 24 and an antenna 21 as illuminance sensors. The lighting management device 1 is connected to the antenna 11. The lighting management device 1 and the moving body 2 communicate wirelessly via the antenna 11 and the antenna 21. The moving body 2 collects illuminance data at various locations in the room 4 by autonomously moving in the facility, i.e., the room 4. The lighting management device 1 controls the illuminance of the lighting device 3 based on the illuminance data and the position information where the illuminance data is collected, so that the room 4 maintains an appropriate illuminance. Here, lighting control refers to turning on / off the power of the lighting device 3 and adjusting the intensity of the illuminance of the lighting device 3.

[0040] For example, lighting control using daylight is described. In room 4, the ideal desktop illumination is set to 800 lux. If the current desktop illumination is 700 lux, the desktop illumination is insufficient. In this case, the lighting management device 1 performs lighting control to increase the illumination by 100 lux. At this time, based on the position information corresponding to the illumination, only the illumination of the lighting device 3 at the required position is increased. If the current desktop illumination is 800 lux or more, for example, 900 lux, under the effect of daylight, the desktop illumination is sufficient. In this case, based on the position information corresponding to the illumination, lighting control is performed in a manner of reducing the illumination by 100 lux (dimming or turning off the lighting). Energy saving can be achieved by utilizing daylight for lighting.

[0041] For example, an obstacle 6 is set in the room 4. For example, the obstacle 6 is a corrugated cardboard box. For example, the obstacle 6 is a user. Figure 1 As shown in (a), a shadow 7 is created by an obstacle 6. Figure 1 (a) shows that the moving object 2 is outside the shadow 7. On the other hand, Figure 1 (b) shows that the moving body 2 is located inside the shadow 7. That is, the sunlight from the window 5 and the illumination light from the lighting 3 that should be irradiated to the moving body 2 are blocked by the obstacle 6. Figure 1 As shown in (b), the state in which the sunlight from the window 5 and the illumination light from the illumination 3 to be irradiated to the moving body 2 are blocked by the obstacle 6 is referred to as "one state". Figure 1 As shown in (a), the state in which the sunlight from the window 5 and the illumination light from the illumination 3 irradiating the moving body 2 are not blocked by the obstacle 6 is referred to as "another state".

[0042] Figure 2 FIG. 1 is a functional structure diagram of a lighting management system according to Embodiment 1 of the present invention. Figure 2As shown, the moving body 2 includes an illumination detection unit 24, a moving body control unit 22, a communication unit 23, and an antenna 21. The illumination detection unit 24 is, for example, disposed on the upper portion of the moving body 2. The illumination detection unit 24 detects the illumination in the room 4 according to the position of the moving body 2 that moves autonomously. The moving body control unit 22 controls the driving and stopping of the moving body 2. The moving body control unit 22 grasps the current position and the current time of the moving body 2. The communication unit 23 transmits the illumination detected by the illumination detection unit 24, the time grasped by the moving body control unit 22 when the illumination is detected, and the position information of the current position grasped by the moving body control unit 22 when the illumination is detected to the lighting management device 1 via the antenna 21.

[0043] like Figure 2 As shown, the lighting management device 1 includes an antenna 11, a communication unit 19, a determination unit 13 as an inference device, a lighting control unit 14, an illuminance storage unit 12, and a threshold storage unit 15. The communication unit 19 receives the illuminance, the time and position information corresponding to the illuminance, transmitted from the moving body 2 via the antenna 11. The illuminance storage unit 12 stores the illuminance received by the communication unit 19, and the time and position information corresponding to the illuminance.

[0044] The determination unit 13 determines whether the illuminance stored in the illuminance storage unit 12 is measured in one state or another state. The determination unit 13 determines whether each illuminance is measured in one state or another state based on the threshold value stored in the threshold value storage unit 15, for example.

[0045] The lighting control unit 14 performs lighting control of the lighting device 3 based on the illuminance stored in the illuminance storage unit 12, and the position information and time corresponding to the illuminance. The lighting control unit 14 does not use the illuminance determined by the determination unit 13 as measured in one state, and the position information and time corresponding to the illuminance for lighting control of the lighting fixture. The lighting control unit 14 uses the illuminance determined by the determination unit 13 as measured in another state, and the position information and time corresponding to the illuminance for lighting control of the lighting fixture.

[0046] Figure 3 This is a graph showing the threshold values ​​of illumination in the lighting management system according to the present embodiment. Figure 3 The vertical axis of the graph represents the illuminance detected by the illuminance detection unit 24 . Figure 3 The horizontal axis of the graph represents the time when the illuminance detection unit 24 of the autonomously moving moving object 2 detects the illuminance. Figure 3 B lux in the room represents the illuminance in the room 4 when there is no lighting by the lighting device 3 . Figure 3 A+B lux in the room 4 represents the illuminance in the room 4 when the lighting device 3 is illuminated. That is, the illuminance is increased by A lux due to the lighting of the lighting device 3. The threshold value C is a threshold value stored in the threshold value storage unit 15.

[0047] Figure 3 The illuminance at time Ta in the graph of is greater than the threshold value C. In this case, the determination unit 13 determines that the illuminance is measured in another state. As a result, the lighting control unit 14 uses the illuminance at time Ta and the position information and time corresponding to the illuminance for lighting control of the lighting fixture. On the other hand, Figure 3 The illuminance at time Tb in the graph of is smaller than the threshold value C. In this case, the determination unit 13 determines that the illuminance is measured in one state. As a result, the lighting control unit 14 does not use the illuminance at time Tb and the position information corresponding to the illuminance for lighting control of the lighting fixture.

[0048] Threshold C can be set appropriately. The manager of the lighting management system can measure the illuminance in the room 4 when the light from the lighting device 3 and the window 5 is blocked by the shadow 7 of the obstacle 6 and the illuminance when it is not blocked, and set the value between the illuminances as the threshold C. At this time, the illuminance corresponding to the position where the light from the lighting device 3 or the window 5 is blocked by the shadow 7 can be judged as the illuminance measured in one state. The illuminance corresponding to the position where the light from the lighting device 3 and the window 5 is not blocked by the shadow 7 can be judged as the illuminance measured in another state. Threshold C can also be determined by the illuminance of another state and the illuminance of one state, respectively. Threshold C can also be determined by the average of multiple illuminances of one state and the average of illuminances of another state, respectively.

[0049] Next, through Figure 4 and Figure 5 The operation of the lighting management system is described. Figure 4 This is a flowchart showing the operation of the lighting management system according to this embodiment.

[0050] First, it is determined whether the communication unit 19 has received the illuminance detected by the illuminance detection unit 24 of the mobile body 2 and the position information and time when the illuminance is detected (S101). If the communication unit 19 has received the illuminance and the position information and time when the illuminance is detected (Yes in S101), the process proceeds to S102. If the communication unit 19 has not received the illuminance and the position information and time when the illuminance is detected (No in S101), the process of S101 is repeated.

[0051] Next, the illuminance storage unit 12 stores the illuminance received by the communication unit 19 and the position information and time when the illuminance was detected (S102). Thereafter, the process proceeds to S103.

[0052] Next, the determination unit 13 determines whether it is a prescribed timing (S103). The prescribed timing refers to a fixed period such as every 10 minutes. In addition, for example, it is a fixed time that can be set by the administrator. It is possible to use a common clock and timer to determine whether it is a prescribed timing. In the case of the prescribed timing (S101 is yes), enter the processing of S104. In the case of not the prescribed timing (S101 is no), return to the processing of S101.

[0053] Next, the determination unit 13 determines whether to exclude the illuminance based on the illuminance stored in the illuminance storage unit 12 and the position information and time when the illuminance is detected (S104). Then, the process proceeds to S105. The detailed operation of S104 will be described later.

[0054] Next, the lighting control of the lighting device 3 is performed based on the illuminance stored in the illuminance storage unit 12 and the position information and time corresponding to the illuminance (S105). As described above, the lighting control unit 14 does not use the illuminance determined by the determination unit 13 as measured in one state and the position information and time corresponding to the illuminance for the lighting control of the lighting device. The lighting control unit 14 uses the illuminance determined by the determination unit 13 as measured in another state and the position information and time corresponding to the illuminance for the lighting control of the lighting device. Then, the process returns to S101. By repeating Figure 4 The action of the flowchart is performed to appropriately adjust the illumination of the lighting device 3 installed in the room 4.

[0055] Figure 5 104. First, the determination unit 13 collects the illuminance stored in the illuminance storage unit 12 within a specified period (for example, a fixed period of every 10 minutes, which can be set by the administrator) and the position information and time corresponding to the illuminance (S201). Then, the process proceeds to S202.

[0056] Next, for each illuminance, it is determined whether it is below the threshold value C stored in the threshold value storage unit 15 (S202). If the illuminance is below the threshold value C (Yes in S202), the process proceeds to S203. If the illuminance is not below the threshold value C (No in S202), the process of S203 is skipped and the process proceeds to S204.

[0057] Next, the illuminance determined to be below the threshold C and the position information and time information corresponding to the illuminance are excluded from the objects of lighting control (S203). Specifically, for example, a flag that can identify the illuminance and the position information corresponding to the illuminance is set to identify the excluded data. Then, the process proceeds to S204.

[0058] Next, the determination unit 13 determines whether the determination of S202 has been performed for all the illuminances collected in S202 (S204). If the determination of S202 has been performed for all the illuminances (Yes in S204), the process proceeds to S205. If the determination of S202 has not been performed for all the illuminances (No in S204), the process returns to S202 and the processes of S202 to S204 are repeated.

[0059] Next, the determination unit 13 transmits the illuminance not excluded in S203 and the position information and time corresponding to the illuminance to the lighting control unit 14 (S205). The operation of S104 is now completed.

[0060] Next, the effects of the invention in this embodiment will be described.

[0061] The lighting management system of the present invention includes a determination unit 13 that determines whether the illuminance collected by the mobile body 2 is measured in one state or in another state. When the determination unit 13 determines that the illuminance is measured in one state, the lighting control unit 14 does not adjust the output of the lighting device based on the illuminance and the position information and time corresponding to the illuminance.

[0062] When the illuminance is less than the threshold value C, the lighting control unit 14 determines that the illuminance is measured in a state where the moving body 2 is affected by the shadow 7 generated by the obstacle 6 set in the room 4. In this case, the lighting control unit considers that the illuminance hinders the appropriate lighting control in the room 4 and does not use the illuminance. As a result, the following effect is achieved: it is not easily affected by the influence of the obstacle 6 such as a human body in the room 4 and more appropriate illumination control can be performed.

[0063] In addition, the determination unit 13 of the lighting management system in the present invention determines whether the illuminance is measured in one state or in another state by using the threshold value C. For example, in the case where the central value is used for lighting control as in the prior art 1 mentioned above as the prior art, when there is only one illuminance data in the surrounding illuminance distribution information, if the illuminance data is abnormal data affected by the shadow of an obstacle, etc., lighting control will be performed using the abnormal data. In addition, in the case where more than half of the illuminance data in the surrounding illuminance distribution information is abnormal data affected by the shadow of an obstacle, etc., lighting control will be performed using the abnormal data. On the other hand, the determination unit 13 of the lighting management system in the present invention only needs to compare one illuminance with the preset threshold value C. That is, as a result, the following effect is achieved: it is not easily affected by the influence of obstacles 6 such as human bodies located in the room 4, and more appropriate illumination control can be performed.

[0064] In addition, the determination unit 13 of the lighting management system of the present invention determines whether the illuminance is measured in one state or in another state by using the threshold value C. For example, when the central value is used for lighting control as in the prior art 1 mentioned above as the prior art, when the surrounding illuminance distribution information has only one illuminance data, the above-mentioned inconvenience occurs, so it is desirable to use a plurality of illuminance data to calculate the central value. On the other hand, the determination unit 13 of the lighting management system of the present invention only needs to compare one illuminance with the preset threshold value C. That is, it has a further effect that appropriate lighting control can be performed in less calculation time.

[0065] Implementation Method 2

[0066] Below, use Figure 6 and Figure 7 The configuration and operation of the lighting management system according to this embodiment will be described. Figure 6 FIG. 1 is a functional structure diagram of the lighting management system of this embodiment. Figure 6 As shown, the lighting management device 1 in this embodiment is newly equipped with a threshold value calculation unit 16 compared to the lighting management device 1 in Embodiment 1. The other points are the same configuration as the lighting management device 1 in Embodiment 1, and detailed description thereof will be omitted.

[0067] The threshold calculation unit 16 calculates the threshold C based on the illuminance collected in advance by the mobile body 2 and stored in the threshold storage unit 15, and the position information and time corresponding to the illuminance. For example, the threshold C is calculated by the following formula (1). Here, A(t) is the average value of the illuminance in another state (when the illuminance detection unit 24 is not blocked by the shadow 7 generated by the obstacle 6) at a certain time t (for example, the same time on different days). w is an arbitrary coefficient in the range of 0<w<1.

[0068] C=A(t)×w···(1)

[0069] For example, w=0.5. When 0.5<w<1, low illuminance is excluded from the objects of lighting control. That is, energy saving tends to be achieved. When 0<w<0.5, low illuminance is not excluded from the objects of lighting control. That is, energy saving tends to be not achieved.

[0070] In addition, for example, the threshold value C can also be calculated by the following formula (2). Here, B(t) is the average value of the illuminance for a state at a certain time t (a case where the illuminance detection unit 24 is blocked by the shadow 7 generated by the obstacle 6). The following formula (2) is a value exactly halfway between the average value A(t) of the illuminance when the illuminance detection unit 24 is not blocked by the shadow 7 generated by the obstacle 6 and the average value B(t) of the illuminance when the illuminance detection unit 24 is blocked by the shadow 7 generated by the obstacle 6.

[0071] C=B(t)+{(A(t)-B(t)) / 2}···(2)

[0072] The values ​​of A(t) and B(t) may be obtained by using the illuminance collected by the mobile body 2 through autonomous movement in the room 4 and stored in the illuminance storage unit 12, as well as the position information and time corresponding to the illuminance. In this case, the manager intentionally places the obstacle 6 in the room 4 at a certain position at a certain time, sets the illuminance corresponding to the time and position information as the illuminance measured in one state, and causes the threshold calculation unit 16 to calculate the value of B(t). The illuminance not corresponding to the time and position information may be set as the illuminance measured in another state, and causes the threshold calculation unit 16 to calculate the value of A(t).

[0073] Alternatively, for example, the threshold C may be calculated based on the distribution of all illuminances collected by the moving body 2 during a certain fixed period. Specifically, the threshold calculation unit 16 may calculate the standard deviation σ of all illuminances during a certain fixed period and set it as the average value of all illuminances - n·σ (n is an arbitrary value, such as 3).

[0074] Next, through Figure 7 The operation of the lighting management system is described. Figure 7 is a flowchart showing the operation of the lighting management system of this embodiment. Figure 5 In the present embodiment, only the details of the processing of S104 of the flowchart will be described. Figure 7 The detailed operation of S104 in this embodiment is shown.

[0075] like Figure 7 As shown, the embodiment 1 Figure 5 The difference of the flowchart is that the process of S206 is added between the processes of S201 and S202. Therefore, only the process of S206 will be described. Other aspects are the same as those of the first embodiment, and detailed descriptions are omitted.

[0076] First, after the process of S201 is completed, the threshold value calculation unit 16 calculates the threshold value C by the method described above (S206). Thereafter, the process proceeds to S202.

[0077] Next, the effects of the invention in this embodiment will be described.

[0078] The lighting management system in the present invention has the same effect as that of Embodiment 1. In addition, the threshold value C is calculated by the threshold value calculation unit 16 based on the illuminance collected by the mobile body 2 and the position information corresponding to the illuminance and the time. Therefore, it is possible to use more illuminance and the position information corresponding to the illuminance to calculate the threshold value C. That is, a more appropriate threshold value C can be set, and therefore, it is also possible to achieve an effect that more appropriate lighting control can be performed compared to Embodiment 1.

[0079] In addition, the threshold value C can be calculated using the illuminance automatically collected by the moving body 2 and the position information and time corresponding to the illuminance. In order to calculate the threshold value C, the manager can deliberately place the obstacle 6 in the room 4 at a certain position at a certain time, classify the illuminance corresponding to the time and position information as the illuminance measured in one state, and classify the illuminance not corresponding to the time and position information as the illuminance measured in another state. That is, there is a further effect of reducing the burden on the manager.

[0080] In addition, the threshold calculation unit 16 calculates the threshold C based on the illuminance collected in advance by the mobile body 2 and stored in the threshold storage unit 15, as well as the position information and time corresponding to the illuminance. However, the time is not necessarily required. For example, the threshold C calculated using the illuminance data measured at 10 am can also be used at 2 pm.

[0081] Implementation 3

[0082] Below, use Figure 8 and Fig. 9 The configuration and operation of the lighting management system according to this embodiment will be described. Figure 8 FIG. 1 is a functional structure diagram of the lighting management system of this embodiment. Figure 8 As shown, the lighting management device 1 in this embodiment is newly equipped with an external environment input unit 17 compared to the lighting management device 1 in Embodiment 2. The other points are the same configuration as the lighting management device 1 in Embodiment 2, and detailed descriptions thereof are omitted.

[0083] The external environment input unit 17 sends an offset (offset) to be added to the threshold value when the threshold value is calculated by the threshold value calculation unit 16 based on the input external environment information. Here, the external environment information refers to weather information such as sunny, cloudy, and rainy days. In addition, the external environment information is, for example, the opening and closing state of the blinds (not shown) provided on the window 5.

[0084] When the external environment information is the open / closed state of a blind (not shown) provided on the window 5, the external environment information input to the external environment input unit 17 is input manually, for example. In addition, when, for example, the blind (not shown) has a function of automatically detecting the open / closed state, the external environment information input to the external environment input unit 17 may be automatically input from the blind.

[0085] For example, when the threshold value C is based on the condition that the blinds not shown are in the open state, when the blinds not shown are in the closed state, an offset corresponding to the closed state can be added. The offset added to the threshold value C can be arbitrarily determined. The illuminance when the blinds are in the open state is compared with the illuminance when the blinds not shown are in the closed state, and the difference in illuminance is set to an offset corresponding to the closed state of the blinds not shown. The illuminance for each state of the open and closed state of the blinds not shown can be measured manually, or the movable body 2 can be moved autonomously in the room 4 to measure automatically.

[0086] When the external environment information is weather information, the external environment information input to the external environment input unit 17 is input manually, for example. The external environment information input to the external environment input unit 17 is automatically input from a server provided by a company or organization that handles weather information, for example.

[0087] For example, when the threshold value C is based on a sunny day, an offset corresponding to each weather condition may be added to the threshold value C on a cloudy day or a rainy day. The offset added to the threshold value C can be arbitrarily determined. The illuminance on a sunny day is compared with the illuminance on a cloudy day or a rainy day, and the difference in illuminance is set as an offset corresponding to each weather condition. The illuminance for each weather condition may be measured manually or automatically by making the moving body 2 move autonomously in the room 4.

[0088] Next, through Fig. 9 The operation of the lighting management system is described. Figure 4 In the present embodiment, only the details of the processing of S104 of the flowchart will be described. Fig. 9 This is a flowchart showing the operation of the lighting management system according to this embodiment. Fig. 9 The detailed operation of S104 in this embodiment is shown.

[0089] like Fig. 9 As shown, the embodiment 2 Figure 7 The difference from the flowchart of is that the processing of S207 and S208 is added instead of S206. Therefore, only the processing of S207 and S208 will be described. The other points are the same as those of the second embodiment, and the detailed description is omitted.

[0090] First, after the process of S201 is completed, external environment information is input to the external environment input unit 17. The external environment input unit 17 sends the offset corresponding to the input external environment information to the threshold value calculation unit 16 (S207). Then, the process proceeds to S208.

[0091] Next, after the process of S207 is completed, the threshold value calculation unit 16 calculates the threshold value C based on the received offset (S208). Then, the process proceeds to S202.

[0092] Next, the effects of the invention in this embodiment will be described.

[0093] The lighting management system in the present invention has the same effect as that of Embodiments 1 and 2. On this basis, external environmental information can be considered, so the influence of external disturbances such as changes in weather or the open and closed state of the blinds installed on the window 5 can be considered and reflected in the threshold C. That is, the threshold C can be set more appropriately, so there is also an effect that more appropriate lighting control can be performed compared with Embodiments 1 and 2.

[0094] The external environment input unit 17 sends the offset amount according to the input external environment information to the threshold calculation unit 16. However, instead of the offset amount added to the threshold C, a coefficient multiplied by the threshold C may be used. This coefficient can also be arbitrarily set according to each external environment information.

[0095] Implementation 4

[0096] Below, use Figure 10 to Figure 15 , the structure and operation of the lighting management system of this embodiment are explained. Fig.10 FIG. 1 is a functional structure diagram of the lighting management system of this embodiment. Figure 8 As shown, the lighting management device 1 in this embodiment does not include the threshold storage unit 15 and the determination unit 13 compared to the lighting management device 1 in the first embodiment. Figure 8 As shown, the lighting management device 1 in this embodiment is newly equipped with a learning model storage unit 18, a learning device 8, and an inference device 9 compared to the lighting management device 1 in the first embodiment. Other aspects are the same as the lighting management device 1 in the second embodiment, and detailed descriptions are omitted.

[0097] Fig.11 1 is a block diagram of a learning device related to the lighting management system of the present embodiment. The learning device 8 includes a data acquisition unit 81 , a model generation unit 82 , and a learned model storage unit 83 .

[0098] The data acquisition unit 81 acquires the illuminance, the position information and time corresponding to the illuminance, and one state / another state as correct answer data as learning data.

[0099] The model generation unit 82 learns the optimal one state / another state for the corresponding illuminance based on the learning data generated based on the illuminance output from the data acquisition unit 81, the position information and time corresponding to the illuminance, and the combination of one state and another state. That is, the following learning model is generated, which infers the optimal one state / another state for the corresponding illuminance based on the illuminance and the position information corresponding to the illuminance in the lighting management system, and one state / another state. Here, the learning data is data obtained by associating the illuminance and the position information corresponding to the illuminance, and one state / another state.

[0100] In addition, the learning device 8 and the inference device 9 are used to learn the optimal state / state of the lighting management device 1 for the corresponding illuminance, but they can also be connected to the lighting management device 1 via a network, for example, and can be used as devices separate from the lighting management device 1. In addition, the learning device 8 and the inference device 9 can also be built into the lighting management device 1. In addition, the learning device 8 and the inference device 9 can also exist on a cloud server.

[0101] The learning algorithm used by the model generation unit 82 can use a known algorithm such as supervised learning, unsupervised learning, and reinforcement learning. As an example, the case where a neural network is applied is described. The model generation unit 82 learns the optimal state / state for the corresponding illumination through so-called supervised learning, for example, according to the neural network model. Here, supervised learning refers to the following method: by providing a set of input and result (label) data to the learning device 8, the features in these learning data are learned, and the result is inferred based on the input.

[0102] A neural network includes an input layer composed of multiple neurons, an intermediate layer (hidden layer) composed of multiple neurons, and an output layer composed of multiple neurons. The intermediate layer may be one layer or more than two layers.

[0103] For example, if Fig.15 In the three-layer neural network shown, when multiple inputs are input to the input layer (X1-X3), the weight W1 (w11-w16) is assigned to the value and input to the intermediate layer (Y1-Y2). As a result, the weight W2 (w21-w26) is further assigned and output from the output layer (Z1-Z3). The output result changes according to the values ​​of the weights W1 and W2.

[0104] In the present application, the neural network learns one state / another state for the optimal corresponding illuminance through so-called supervised learning according to learning data generated based on the illuminance obtained by the data acquisition unit 81 and the position information corresponding to the illuminance, and the combination of one state / another state.

[0105] That is, the neural network learns by inputting illuminance and position information corresponding to the illuminance to the input layer and adjusting weights W1 and W2 so that the result output from the output layer approaches one state / another state.

[0106] The model generation unit 82 generates a learned model by executing the above-described learning and outputs the model.

[0107] The learned model storage unit 83 stores the learned model output from the model generation unit 82 .

[0108] Next, use Fig.12 , the learning process performed by the learning device 8 is explained. Fig.12 This is a flowchart related to the learning process of the learning device 8.

[0109] In step S301, the data acquisition unit 81 acquires the illuminance, the position information and time corresponding to the illuminance, and one state / another state. In addition, the illuminance, the position information and time corresponding to the illuminance, and one state / another state are acquired at the same time, but as long as the illuminance, the position information and time corresponding to the illuminance, and one state / another state can be input in association, the illuminance, the position information and time corresponding to the illuminance, and one state / another state data can also be acquired at different times.

[0110] In step S302, the model generation unit 82 learns one state / another state for the corresponding illumination through so-called supervised learning according to the learning data generated based on the illumination acquired by the data acquisition unit 81 and the position information and time corresponding to the illumination, and the combination of one state / another state, and generates a learned model.

[0111] In step S303 , the learned model storage unit 83 stores the learned model generated by the model generation unit 82 .

[0112] Fig.13 1 is a block diagram of an inference device 9 related to a lighting management system. The inference device 9 includes a data acquisition unit 91 and an inference unit 92 .

[0113] The data acquisition unit 91 acquires the illuminance, and the position information and time corresponding to the illuminance from the moving object 2 via the antenna 11 .

[0114] The inference unit 92 infers a state / another state for the corresponding illuminance obtained using the learned model. That is, by inputting the illuminance obtained by the data acquisition unit 91 and the position information and time corresponding to the illuminance into the learned model, the state / another state for the corresponding illuminance inferred based on the illuminance and the position information and time corresponding to the illuminance can be output.

[0115] In addition, in this embodiment, it is described that a state / another state for corresponding illumination is output using a learned model learned in the learning device 8 of the lighting management device 1, but a learned model can also be obtained from other lighting management devices or the like, and based on the learned model, a state / another state for corresponding illumination is output.

[0116] Next, use Fig.14 , a process for obtaining one state / another state for corresponding illuminance using the inference device 9 is described.

[0117] In step S401 , the data acquisition unit 91 acquires illuminance and position information and time corresponding to the illuminance.

[0118] In step S402 , the inference unit 92 inputs the illuminance and the position information and time corresponding to the illuminance to the learned model stored in the learned model storage unit, and obtains one state / another state for the corresponding illuminance.

[0119] In step S403 , the inference unit 92 outputs the one state / the other state for the corresponding illuminance obtained by the learned model to the lighting control unit 14 .

[0120] In step S404, the lighting control unit 14 performs lighting control using the illuminance and the position information and time corresponding to the illuminance according to the one state / another state output from the inference device 9 for the corresponding illuminance. Specifically, the lighting control unit 14 does not use the illuminance inferred by the inference device 9 as one state in the lighting control. The lighting control unit 14 uses the illuminance inferred by the inference device 9 as another state in the lighting control. As a result, the lighting control of the lighting fixture can be performed more appropriately.

[0121] As another example, the following method may be used: The model generation unit 82 directly stores the input illuminance, the position information and time corresponding to the illuminance, and one state / another state as training data in the learned model storage unit 83 as shown in Table 1 below.

[0122] [Table 1]

[0123]

[0124] On the other hand, in the inference device 9 , for example, the following measurement data is input to the data acquisition unit 91 .

[0125] Time: 10:05, Position: (x, y) = (20, 20), Illuminance: 550 lux

[0126] The inference device 9 compares the above-mentioned illuminance and the position information and time corresponding to the illuminance with the illuminances and the position information and time corresponding to the illuminances stored in the learned model storage unit 83, and selects the illuminance and the position information and time corresponding to the illuminance with the highest similarity.

[0127] The inference unit 92 determines the similarity and sorts the illumination in each position information at each time stored in the learned model storage unit 83 as shown in the following Table 2. For example, the similarity can be determined based on the difference between the time, each position information and illumination input to the data acquisition unit 91 and the time, each position information and illumination stored in the learned model storage unit 83.

[0128] like Figure 2 As shown, the inference unit 92 calculates the average similarity based on the similarity of each time, each position information and illumination. The data with the smallest average similarity is set as the data with the highest similarity. Figure 2 As shown, the average similarity of the data No. 2 is (1 bit + 1 bit + 2 bits) / 3 = 1.3, which is the smallest among the data No. 1 to No. 10.

[0129] [Table 2]

[0130]

[0131] Therefore, the data of time: 10:05, position: (x, y) = (20, 20), illuminance: 550 lux input to the data acquisition unit 91 is judged to be most similar to the data No. 2, i.e., time: 10:00, position: (x, y) = (10, 10), illuminance: 600 lux.

[0132] The inference unit 92 determines whether the illuminance input from the data acquisition unit 91 is measured in one state or in another state, based on the illuminance included in the learned model determined to be the most similar, and the position information and time corresponding to the illuminance. In the above example, the data No. 2 is determined to be in another state, that is, data not blocked by the shadow 7 of the obstacle 6. Therefore, the data such as time: 10:05, position: (x, y) = (20, 20), and illuminance: 550 lux input to the data acquisition unit 91 are not excluded by the lighting control unit 14 and are used as lighting control data. The data excluded as lighting control data by the lighting control unit 14 can also be inferred in the same way.

[0133] In addition, in the present embodiment, supervised learning is applied to the learning algorithm used by the model generation unit 82, but the present invention is not limited thereto. In addition to supervised learning, reinforcement learning, unsupervised learning, or semi-supervised learning can also be applied to the learning algorithm.

[0134] Furthermore, the model generation unit 82 may also learn one state / another state for corresponding illumination according to the learning data generated for a plurality of lighting management devices 1. Furthermore, the model generation unit 82 may also obtain the learning data from a plurality of lighting management devices 1 used in the same area, and may also learn one state / another state for corresponding illumination using the learning data collected from a plurality of lighting management devices 1 operating independently in different areas. Furthermore, it is also possible to add or remove the lighting management system 1 that collects the learning data to or from the object midway. Furthermore, it is also possible that the learning device 8 that has learned one state / another state for corresponding illumination with respect to a certain lighting management device 1 is applied to a different lighting management system 1, and the one state / another state for corresponding illumination is relearned and updated with respect to the different lighting management device 1.

[0135] In addition, as a learning algorithm for the model generation unit 82, deep learning that learns the extraction of feature quantities themselves can also be used, and machine learning can also be performed according to other well-known methods such as genetic programming, functional logic programming, support vector machines, etc.

[0136] Furthermore, the lighting management device 1 of the present embodiment may also include the external environment input unit 17 shown in Embodiment 3. The external environment information input from the external environment input unit 17 may be acquired from the data acquisition unit 81 of the learning device 8. The model generation unit 82 may further generate a learned model using the external environment information. Furthermore, the inference device 9 may acquire the external environment information input from the external environment input unit 17 from the data acquisition unit 91 of the inference device 9. The inference unit 92 may determine whether the corresponding illuminance is measured in one state or in another state based on the learned model generated by further using the external environment information.

[0137] In the lighting control systems of Embodiments 1 to 4, it is described that the lighting control unit 14 performs lighting control of the lighting device 3 based on the illuminance stored in the illuminance storage unit 12 and the position information and time corresponding to the illuminance. However, the time is not necessarily required. For example, the illuminance data 10 minutes ago can be fully used for lighting control at the current time.

[0138] Fig.16 1 is a hardware configuration diagram of the lighting management device of Embodiments 1 to 4. The respective units indicated by reference numerals 12 to 29, 81, 82, 91, and 92 in Embodiments 1 to 4 represent functions of the lighting management device 1. The lighting management device 1 includes, for example, a processing circuit including a processor 1001 and a memory 1002 as hardware resources. The functions of the illuminance storage unit 12, the threshold storage unit 15, and the learning model storage unit 18 are implemented by the memory 1002. The lighting management device 1 implements the functions of the determination unit 13, the lighting control unit 14, the threshold calculation unit 16, the external environment input unit 17, the communication unit 19, the data acquisition unit 81, the model generation unit 82, the data acquisition unit 91, and the inference unit 92 by executing the program stored in the memory 1002 by the processor 1001. Alternatively, the program stored in the external storage medium is installed in the memory 1002 mounted in the computer, and the program is executed by the processor 1001, thereby implementing the above functions by the above memory 1002 and the processor 1001.

[0139] The processor 1001 may also be referred to as a CPU (Central Processing Unit), a central processing device, a processing device, an arithmetic device, a microprocessor, a microcomputer or a DSP. As the memory 1002, a semiconductor memory, a magnetic disk, a floppy disk, an optical disk, a high-density disk, a mini disk or a DVD may also be used. The semiconductor memory that can be used includes RAM, ROM, flash memory, EPROM and EEPROM.

[0140] Some or all of the functions of the lighting management device 1 may be implemented by hardware. Hardware that implements the functions of the lighting management device 1 may include a single circuit, a decoding circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0141] The present invention is not limited to Embodiments 1 to 4 and their combinations, and may be partially changed or omitted within the scope of the technical concept of the present invention.

[0142] Description of Reference Numerals

[0143] 1 Lighting management device, 11 antenna, 12 illuminance storage unit, 13 determination unit (inference device), 14 lighting control unit, 15 threshold storage unit, 16 threshold calculation unit, 17 external environment input unit, 18 learning model storage unit, 19 communication unit, 2 moving body, 21 antenna, 22 moving body control unit, 23 communication unit, 24 illuminance detection unit, 3 lighting device, 4 room, 5 window, 6 obstacle, 7 shadow, 8 learning device, 81 data acquisition unit, 82 model generation unit, 9 inference device, 91 data acquisition unit, 92 inference unit.

Claims

1. A lighting management device, wherein: The lighting management device comprises: an illumination storage unit that stores illumination collected by a moving body and position information corresponding to the illumination, the moving body having an illumination detection unit and moving within the facility; a lighting control unit that adjusts the output of lighting equipment installed in the facility according to the illumination; as well as an inference device that determines that the illuminance is measured in one state when the illuminance is less than a predetermined threshold value, and determines that the illuminance is measured in another state when the illuminance is greater than the predetermined threshold value, When the inference device determines that the illuminance is measured in the other state, the lighting control unit uses the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information. When the inference device determines that the illuminance is measured in the one state, the lighting control unit does not use the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information.

2. The lighting management device according to claim 1, wherein: The lighting management device includes a threshold value calculation unit that calculates the threshold value based on the illuminance collected in advance by the moving object and the position information corresponding to the illuminance.

3. The lighting management device according to claim 2, wherein: The threshold value calculation unit further calculates the threshold value based on a time corresponding to the illuminance.

4. The lighting management device according to claim 3, wherein: When the average value of the illuminance measured at time t in the other state is assumed to be A(t) and w is assumed to be a coefficient of 0<w<1, the threshold value calculation unit sets the threshold value to A(t)×w.

5. The lighting management device according to claim 3, wherein: When the average value of the illuminance measured at time t in the other state is A(t) and the illuminance measured at time t in the one state is B(t), the threshold calculation unit sets the threshold to B(t)+{(A(t)-B(t)) / 2}.

6. A lighting management device, wherein: The lighting management device comprises: an illumination storage unit that stores illumination collected by a moving body and position information corresponding to the illumination, the moving body having an illumination detection unit and moving within the facility; a lighting control unit that adjusts the output of lighting equipment installed in the facility according to the illumination; A learning device having a model generating unit that generates a learned model using learning data, the learning data including whether the illuminance is measured in one state or in another state, the learned model being used to infer whether the illuminance corresponding to the position information is measured in the one state or in the other state based on the illuminance corresponding to the position information; as well as an inference device that uses a learned model for inferring whether the illuminance is measured in the one state or the other state, and determines whether the illuminance is measured in the one state or the other state based on the illuminance input from the data acquisition unit, When the inference device determines that the illuminance is measured in the other state, the lighting control unit uses the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information. When the inference device determines that the illuminance is measured in the one state, the lighting control unit does not use the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information.

7. The lighting management device according to claim 6, wherein: The inference device compares the illuminance and the position information and time corresponding to the illuminance input from the data acquisition unit with each illuminance and the position information and time corresponding to the illuminance contained in the completed model, and determines whether the illuminance input from the data acquisition unit is measured in the one state or the other state based on the illuminance with the highest similarity contained in the completed model and the position information and time corresponding to the illuminance.

8. The lighting management device according to any one of claims 1 to 7, wherein: The lighting management device includes an external environment input unit for inputting external environment information. The inference device further determines, based on the external environment information, whether the illuminance is measured in the one state or in the other state.

9. A lighting management system, wherein: The lighting management system comprises: a mobile object including an illumination detection unit and moving within the facility; and Lighting management devices, The lighting management device has: an illumination storage unit for storing the illumination collected by the moving object and position information corresponding to the illumination; a lighting control unit that performs lighting control of lighting equipment installed in the facility based on the illuminance and position information corresponding to the illuminance; as well as an inference device that determines that the illuminance is measured in one state when the illuminance is less than a predetermined threshold value, and determines that the illuminance is measured in another state when the illuminance is greater than the predetermined threshold value, When the inference device determines that the illuminance is measured in the other state, the lighting control unit uses the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information. When the inference device determines that the illuminance is measured in the one state, the lighting control unit does not use the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information.

10. A lighting management system, wherein: The lighting management system comprises: a mobile object including an illumination detection unit and moving within the facility; and Lighting management devices, The lighting management device has: an illumination storage unit for storing the illumination collected by the moving object and position information corresponding to the illumination; a lighting control unit that performs lighting control of lighting equipment installed in the facility based on the illuminance and position information corresponding to the illuminance; A learning device having a model generating unit that generates a learned model using learning data, the learning data including whether the illuminance is measured in one state or in another state, the learned model being used to infer whether the illuminance corresponding to the position information is measured in the one state or in the other state based on the illuminance corresponding to the position information; as well as an inference device that uses a learned model for inferring whether the illuminance is measured in one state or in another state, and determines whether the illuminance is measured in the one state or in the other state based on the illuminance input from the data acquisition unit, When the inference device determines that the illuminance is measured in the other state, the lighting control unit uses the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information. When the inference device determines that the illuminance is measured in the one state, the lighting control unit does not use the illuminance and the position information corresponding to the illuminance for adjusting the output of the lighting fixture corresponding to the position information.

11. A learning device, wherein: The learning device comprises: a data acquisition unit that acquires learning data, the learning data including illuminance collected by a moving body and position information corresponding to the illuminance, and one state or another state as correct answer data determined based on a predetermined threshold value as whether the illuminance is measured in one state or in another state, the moving body having an illuminance detection unit and moving within a facility; and A model generating unit generates a learned model using the learning data, the learned model being used to infer whether the illuminance is measured in the one state or the other state based on the illuminance and position information corresponding to the illuminance.

12. An inference device, wherein: The inference device comprises: a data acquisition unit that acquires illuminance collected by a moving object and position information corresponding to the illuminance, the moving object having an illuminance detection unit and moving within the facility; and An inference unit, which uses a learned model generated by the learning device according to claim 11 for inferring whether the illuminance corresponding to the position information is measured in one state or in another state based on the illuminance and the position information corresponding to the illuminance, and outputs whether the illuminance is measured in the one state or in the other state based on the illuminance input from the data acquisition unit.

Images