Intelligent light control method and system for intelligent nursing home

Abstract

The invention relates to the technical field of intelligent light control, and particularly discloses an intelligent light control method and system for a smart nursing home, and the method comprises the steps: collecting sensing data in the smart nursing home; performing one or more rounds of iterative updating on a preset knowledge graph based on the sensing data to obtain a target knowledge graph; inputting the target knowledge graph into a graph processing model to generate a light control instruction; and adjusting the light setting of the target area based on the light control instruction. According to the intelligent light control method for the smart nursing home, light in the smart nursing home is controlled in combination with the knowledge graph, health conditions and living habits of different old people can be met, and light control management considering individuals and groups is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and in particular to an intelligent lighting control method and system for a smart nursing home. Background Art

[0002] As the aging of society intensifies, the problem of elderly care becomes increasingly prominent. As a new model of elderly care, smart nursing homes have greatly improved the quality and efficiency of elderly care services by introducing advanced technologies such as the Internet of Things and artificial intelligence. As an important part of the nursing home environment, the intelligent transformation of lighting control systems is of great significance to improving the quality of life of the elderly.

[0003] Therefore, the present invention provides an intelligent lighting control method and system for a smart nursing home to better control the lighting in the smart nursing home. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent lighting control method for a smart nursing home to solve the following technical problems: the existing lighting control systems mostly rely on manual operation or simple sensor control, and cannot be dynamically adjusted according to the individual needs of the elderly and environmental changes, and have problems such as inconvenience in use and high energy consumption.

[0005] The purpose of the present invention can be achieved through the following technical solutions: An intelligent lighting control method for a smart nursing home, comprising: collecting sensor data in the smart nursing home; the sensor data comprises light sensor data, sound sensor data, and motion sensor data; Based on the sensor data, the preset knowledge graph is iteratively updated in one or more rounds to obtain a target knowledge graph; the nodes of the preset knowledge graph include installed lights, activity areas, and rest areas; the edges of the knowledge graph include action paths between nodes; the attribute characteristics of the nodes include lamp type, lamp installation location, light intensity, and light intensity; the attribute characteristics of the edges include path attributes and path usage frequency; Inputting the target knowledge graph into a graph processing model to generate lighting control instructions; Adjusting the lighting settings of the target area based on the lighting control instructions; Among them, one round of iteration includes: Determine a light intensity map of the smart nursing home based on the light sensing data; Determine a person distribution map in the smart nursing home based on the sound sensor data and the motion sensor data; Determine a lighting area map based on the light intensity map and the character distribution map; Determining a motion trend distribution map based on the motion sensing data; The following formula is used to determine whether the overall coverage of the light area map to the motion trend distribution map meets the judgment condition:

[0006] Among them, C represents the lighting coverage, I i represents the light intensity of the ith area, A i represents the area or weight of the ith region, n is the number of regions for parameter calculation; D represents the coverage of the movement trend, T j represents the motion trend value of the jth motion area, W j represents the area or weight of the jth motion area, and E represents the overall coverage of the light area distribution to the motion trend; If yes, extract data from the preset knowledge graph based on the light area graph and construct a subgraph as the target knowledge graph; If not, based on the motion trend graph, the character distribution graph is updated, and based on the updated character distribution graph, the lighting area graph is re-determined and the next round of iteration is performed.

[0007] As a further solution of the present invention: extracting data from the preset knowledge graph based on the light area graph and constructing a subgraph includes: Based on the light area map, determining a target location where a light needs to be lit; Obtaining the installation locations of installed lamps in the preset knowledge graph; respectively calculating the distance between the installation position of the installed lamp and the target position where the lamp needs to be lit; Extract the installed lights corresponding to the installation positions whose distance is less than the preset value as nodes of the subgraph; Based on the nodes of the subgraph, edges between nodes corresponding to the nodes of the subgraph are extracted from the preset knowledge graph as edges of the subgraph.

[0008] As a further solution of the present invention: the updating of the character distribution map based on the motion trend map includes: Based on the exercise trend graph, determining a target population with an exercise trend; Acquire the portraits of the target group through an image sensor; Identify the person object in the person portrait and obtain the attribute characteristics of the person object; Determine a motion vector of each character object based on the attribute feature and the current action behavior of the character object; Based on the motion vector of each of the character objects, determining the motion end point position of each of the character objects; Based on the endpoint positions of the various character objects, an updated character distribution map is determined.

[0009] As a further solution of the present invention: determining the motion vector of each character object based on the attribute feature and the current action behavior of the character object includes: Determining the movement speed of the person object based on the attribute feature; Determine a movement direction based on a current action of the character object; The motion vector is determined based on the motion speed and the motion direction.

[0010] As a further solution of the present invention: the step of inputting the target knowledge graph into a graph processing model to generate a lighting control instruction includes: Acquire the living habits and physical condition data of the corresponding person object based on the target knowledge graph; Converting the life habits of the character object and the body state data into a first feature vector; The feature vector and the target knowledge graph are input into the graph processing model to generate the lighting control instruction.

[0011] As a further solution of the present invention: the sensing data further includes biometric sensing data, and the biometric sensing data includes a facial image; the method further includes: Based on the facial image of the person object, determining the facial micro-expression of the person object by image recognition; Converting the facial micro-expression into a second feature vector; The feature vector is used as input data of the image processing model to generate the light control instruction.

[0012] As a further solution of the present invention: the method further comprises: Based on the life habits and physical condition data of the character object, a weight coefficient is assigned to each character object; the weight coefficient reflects the acceptance degree of the character object to the light intensity; The weight coefficient is used as input data of the image processing model to generate the light control instruction.

[0013] An intelligent lighting control system for a smart nursing home, the system comprising: A data acquisition module, used to collect sensor data in the smart nursing home; the sensor data includes light sensor data, sound sensor data, and motion sensor data; An iteration module, configured to perform one or more rounds of iterative updates on a preset knowledge graph based on the sensor data to obtain a target knowledge graph; the nodes of the preset knowledge graph include installed lights, activity areas, and rest areas; the edges of the knowledge graph include action paths between nodes; the attribute characteristics of the nodes include lamp type, lamp installation location, light intensity, and light intensity; the attribute characteristics of the edges include path attributes and path usage frequency; An instruction generation module, used for inputting the target knowledge graph into a graph processing model to generate a lighting control instruction; A lighting control module, used to adjust the lighting settings of the target area based on the lighting control instructions; Among them, one round of iteration includes: Determine a light intensity map of the smart nursing home based on the light sensing data; Determine a person distribution map in the smart nursing home based on the sound sensor data and the motion sensor data; Determine a lighting area map based on the light intensity map and the character distribution map; Determining a motion trend distribution map based on the motion sensing data; The following formula is used to determine whether the overall coverage of the light area map to the motion trend distribution map meets the judgment condition:

[0014] Among them, C represents the lighting coverage, I i represents the light intensity of the ith area, A i represents the area or weight of the ith region, n is the number of regions for parameter calculation; D represents the coverage of the movement trend, T j represents the motion trend value of the jth motion area, W j represents the area or weight of the jth motion area, and E represents the overall coverage of the light area distribution to the motion trend; If yes, extract data from the preset knowledge graph based on the light area graph and construct a subgraph as the target knowledge graph; If not, based on the motion trend graph, the character distribution graph is updated, and based on the updated character distribution graph, the lighting area graph is re-determined and the next round of iteration is performed.

[0015] Beneficial effects of the present invention: By introducing a variety of sensors and intelligent algorithms, the knowledge graph is used to achieve close integration with the installation of lamps in the nursing home and the activities of the elderly, and the automatic adjustment of the lights in the smart nursing home is realized. The automatic adjustment of the lights can take into account the activities of individuals and multiple people, and achieve the purpose of improving the comfort and safety of the elderly's living environment. In addition, the present invention can intelligently adjust the lighting settings in the nursing home through continuous iteration and optimization, improve the comfort and safety of living, and achieve energy-saving effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a flow chart of the intelligent lighting control method of the present invention for a smart nursing home. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, the present invention is an intelligent lighting control method for a smart nursing home. Figure 1 1 is a flow chart of the intelligent lighting control method for a smart nursing home of the present invention. In some embodiments, process 100 may be executed by a processor or a lighting intelligent control system. Process 100 may include the following operations.

[0020] Step 101, collecting sensor data in the smart nursing home.

[0021] Among them, the sensor data includes light sensor data, sound sensor data, and motion sensor data.

[0022] Light sensing data refers to data obtained by detecting the light intensity in a room through a light sensor.

[0023] Sound sensor data refers to data obtained by detecting sound signals in a room through a sound sensor.

[0024] Motion sensing data refers to the data obtained by detecting the activities of people in a room through a motion sensor.

[0025] In some embodiments, the sensor data in the smart nursing home may also include other sensor data, such as temperature and humidity sensor data, infrared sensor data, biometric sensor data, and image sensor data.

[0026] Step 102: perform one or more rounds of iterative updates on the preset knowledge graph based on the sensor data to obtain a target knowledge graph.

[0027] A preset knowledge graph refers to a pre-built knowledge graph. A knowledge graph is a semantic network that reveals the relationships between entities (or objects). Nodes in the graph represent entities. Nodes can be of multiple types, called node types, which are used to indicate various entities. Edges in the graph represent relationships, and edges can also be of multiple types, called edge types, which are used to indicate various relationships. Entities can refer to things in the real world, such as people, geographic locations (for example, a location in a room), devices (for example, lights), and so on. Relationships can be used to express connections between different entities, for example, a person moves from one location in a room to another; for example, a path of action between two locations in a room, and so on.

[0028] In some embodiments, the nodes of the preset knowledge graph include installed lights, activity areas, and rest areas. The edges of the knowledge graph include action paths between nodes. The attribute characteristics of the nodes include lamp type, lamp installation location, light intensity, and light intensity. The attribute characteristics of the edges include path attributes and path usage frequency.

[0029] Updating refers to updating the nodes and edges of the preset knowledge graph, including adding, deleting or changing node attributes.

[0030] Among them, one round of iteration may include the following operations.

[0031] S10, determining a light intensity map of the smart nursing home based on the light sensing data.

[0032] The light sensing data may include the light intensity of various locations within the smart nursing home, and a light intensity diagram may be added based on the floor plan of the entire smart nursing home to obtain a light intensity map.

[0033] S11, determining a distribution map of people in the smart nursing home based on the sound sensor data and the motion sensor data.

[0034] In some embodiments, data preprocessing may be performed on the sound sensor data and the motion sensor data, including operations such as data cleaning and noise filtering, to ensure the accuracy and reliability of the data.

[0035] Afterwards, useful features can be extracted from the processed sound and motion data, such as the frequency, intensity, duration of the sound, the speed and direction of the movement, etc. These features can help better identify and understand the activity status of the person.

[0036] Then, the extracted features can be analyzed and learned through machine learning and artificial intelligence algorithms to identify different people. And according to the activity trajectory of people in the sound and motion data, combined with the layout of the nursing home, the specific location of each person in the nursing home can be determined. For example, the precise positioning of people can be achieved by matching sensor data with the building structure of the nursing home.

[0037] Finally, a character distribution map is generated by displaying the location information of each character in a graphical manner.

[0038] S12: Determine a lighting area map based on the light intensity map and the character distribution map.

[0039] When determining the lighting area map, if the current time is daytime, you can consider using ambient light to provide overall brightness, using lights as key lighting to highlight important areas, and auxiliary lighting to create a specific atmosphere; if the current time is night, lights are needed to focus on highlighting important areas, and auxiliary lighting to supplement areas with insufficient light.

[0040] In some embodiments, the area corresponding to the person in the person distribution map may be regarded as an important area, other areas, such as the motion path area, may be regarded as secondary areas, and other areas may be regarded as general areas.

[0041] S13, determining a motion trend distribution diagram based on the motion sensing data.

[0042] Motion sensing data can include information such as the position, speed, acceleration, etc. of a person. This information can reflect the object's motion trajectory, speed changes, acceleration changes, etc.

[0043] Based on this information, a motion trend distribution map can be generated.

[0044] S14, determining whether the overall coverage of the light area map to the motion trend distribution map meets the judgment condition by the following formula:

[0045] Among them, C represents the lighting coverage, I i represents the light intensity of the ith area, A i represents the area or weight of the ith region, n is the number of regions for parameter calculation; D represents the coverage of the movement trend, T j represents the motion trend value of the jth motion area, W j It represents the area or weight of the jth motion area, and E represents the overall coverage of the light area distribution to the motion trend.

[0046] C measures the average coverage intensity of the lighting area, taking into account the actual lighting intensity and importance of each area. D measures the overall importance or intensity of the motion trend distribution. E provides a ratio that represents the actual coverage of the lighting area to the motion trend distribution. The higher the value, the better the coverage effect. When the overall coverage of E is greater than 90% or 95%, it can be considered that the overall coverage meets the judgment condition.

[0047] In this embodiment, by effectively combining light intensity and motion trend, a comprehensive evaluation of the light area coverage is achieved, which can improve the rationality of intelligent light control and save resources as much as possible while meeting the light coverage requirements.

[0048] When the overall coverage level meets the judgment condition, execute S15, otherwise execute S16.

[0049] S15: If yes, extract data from the preset knowledge graph based on the light area graph and construct a subgraph as the target knowledge graph.

[0050] In some embodiments, extracting data from the preset knowledge graph based on the light area graph and constructing a subgraph includes: Based on the light area map, a target position where a light needs to be turned on is determined; for example, the target position may be an important area, a secondary area, a general area, etc. in the light area map.

[0051] Obtaining the installation locations of installed lamps in the preset knowledge graph; The distances between the installation positions of the installed lamps and the target positions where the lamps need to be lit are calculated respectively; the calculation method may be a pairwise calculation, that is, the distance between the installation position of each installed lamp and each target position where the lamps need to be lit is calculated.

[0052] The installed lights corresponding to the installation positions whose distance is less than a preset value are extracted as nodes of the subgraph; for example, the distance is less than 3m, 5m, etc.

[0053] Based on the nodes of the subgraph, edges between nodes corresponding to the nodes of the subgraph are extracted from the preset knowledge graph as edges of the subgraph.

[0054] Through the above steps, a subgraph of the preset knowledge graph can be constructed.

[0055] S16: If not, based on the motion trend graph, update the character distribution graph, and based on the updated character distribution graph, re-determine the lighting area graph and execute the next round of iteration.

[0056] In some embodiments, updating the character distribution map based on the motion trend map includes: Based on the motion trend graph, a target population with a motion trend is determined; the target population may be a part of the people in the motion trend graph or may be all of them.

[0057] The image sensor is used to obtain the portrait of the target group, which is an image including the face image of the target group, such as a photo.

[0058] Identify the person object in the person portrait and obtain the attribute characteristics of the person object. The attribute characteristics of the person object include the person's age, gender, physical health, eyesight, sensitivity to light, etc. This information can be obtained in advance through detection and diagnosis of each person object and stored in the database in advance, and the relevant data can be directly called when needed.

[0059] Based on the attribute features and the current action behavior of the character object, a motion vector of each character object is determined. The motion vector includes a motion direction and a motion distance of the character object.

[0060] Based on the motion vector of each of the character objects, the motion end position of each of the character objects is determined, wherein the vector end position of the motion vector corresponding to each character object is also the motion end position of the character object.

[0061] Based on the terminal positions of each character object, an updated character distribution map is determined. The updated character distribution map reflects the location areas to which the characters in the smart nursing home may move in the short term. In this embodiment, the character distribution map is updated by pre-predicting the motion vectors of each character, so that the lighting can be controlled in advance, ensuring that the lighting intensity within the sight range of the elderly during activities will not change suddenly, thereby improving the comfort of the lighting intensity in the smart nursing home.

[0062] In some embodiments, determining the motion vector of each character object based on the attribute feature and the current action behavior of the character object includes: Determining the movement speed of the person object based on the attribute feature; Determine a movement direction based on a current action of the character object; The motion vector is determined based on the motion speed and the motion direction.

[0063] Step 103: input the target knowledge graph into a graph processing model to generate lighting control instructions.

[0064] In some embodiments, inputting the target knowledge graph into a graph processing model to generate a lighting control instruction includes: Based on the target knowledge graph, the living habits and physical condition data of the corresponding character object are obtained. The living habits include the character's exercise habits, light-on and light-off habits, etc., and the physical condition data includes the character's exercise ability, body function, whether there is an eye disease (for example, presbyopia, glaucoma, etc.), etc.

[0065] The life habits and the body state data of the character object are converted into a first feature vector. Specifically, the life habits and the body state data can be converted into the first feature vector by feature extraction.

[0066] The feature vector and the target knowledge graph are input into the graph processing model to generate the lighting control instruction.

[0067] The graph processing model can process the input feature vector and target knowledge graph, and output lighting control instructions after processing.

[0068] The graph processing model can be pre-trained based on big data, and the graph processing model can be a graph neural network model, etc.

[0069] In this embodiment, by considering the living habits and physical condition data of the character object, the lighting control can be more in line with the character's needs. For example, if a character object included in the target knowledge graph has presbyopia and is accustomed to reading at night, the graph processing model can generate a lighting control instruction to adjust the light brightness to suit the visual needs of the character while ensuring that sufficient light is provided when reading.

[0070] In some embodiments, in order to enable the graph processing model to output more accurate or more suitable lighting control instructions for the current smart nursing home environment, the sensor data may also include biometric sensor data, and the biometric sensor data may include a facial image; the method may further include: Based on the facial image of the person object, the facial micro-expression of the person object is determined by image recognition. The facial micro-expression of the person object can reflect whether the current light intensity makes it feel comfortable. For example, if the light is too strong, the person's eyes will have a slightly squinting expression.

[0071] Converting the facial micro-expression into a second feature vector; The feature vector is used as input data of the image processing model to generate the light control instruction.

[0072] In this embodiment, in order to make the graph processing model output more accurate or more suitable lighting control instructions for the current smart nursing home environment, biometric sensor data including facial images is introduced. The purpose of this method is to determine whether the current light intensity makes them feel comfortable by analyzing the facial micro-expressions of the characters, thereby adjusting the lighting control instructions and improving the comfort of the lighting environment of the nursing home.

[0073] In some embodiments, the method further comprises: Based on the living habits and physical condition data of the character object, a weight coefficient is assigned to each character object; the weight coefficient reflects the acceptance degree of the character object to the light intensity; The weight coefficient is used as input data of the image processing model to generate the light control instruction.

[0074] In this embodiment, by assigning a weight coefficient to each character object and using an image processing model to generate a lighting control instruction, a more personalized and comfortable lighting environment can be provided for each character object.

[0075] Step 104: Adjust the lighting settings of the target area based on the lighting control instruction.

[0076] In some embodiments, the lighting control of the target area based on the lighting control instructions can be achieved in various ways, for example, a PLC-based control method.

[0077] The working principle of the present invention is as follows: by introducing a variety of sensors and intelligent algorithms, combined with the knowledge graph, the installation of lamps in the nursing home, the activities of the elderly, and the physical health status of the elderly are comprehensively considered, and the automatic adjustment of the lights in the smart nursing home is realized. The automatic adjustment of the lights can take into account the activities of individuals and multiple people, and achieve the purpose of improving the comfort and safety of the elderly's living environment. At the same time, it can meet the health conditions and living habits of different elderly people, and realize the lighting control management that takes into account both individuals and groups. In addition, the present invention can intelligently adjust the lighting settings in the nursing home through continuous iteration and optimization, improve the comfort and safety of living, and achieve energy-saving effects.

[0078] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent lighting control method for a smart nursing home, characterized in that: The method comprises: Collecting sensor data in the smart nursing home; the sensor data includes light sensor data, sound sensor data, and motion sensor data; Based on the sensor data, the preset knowledge graph is iteratively updated in one or more rounds to obtain a target knowledge graph; the nodes of the preset knowledge graph include installed lights, activity areas, and rest areas; the edges of the knowledge graph include action paths between nodes; the attribute characteristics of the nodes include lamp type, lamp installation location, light intensity, and light intensity; the attribute characteristics of the edges include path attributes and path usage frequency; Inputting the target knowledge graph into a graph processing model to generate lighting control instructions; Adjusting the lighting settings of the target area based on the lighting control instructions; Among them, one round of iteration includes: Determine a light intensity map of the smart nursing home based on the light sensing data; Determine a person distribution map in the smart nursing home based on the sound sensor data and the motion sensor data; Determine a lighting area map based on the light intensity map and the character distribution map; Determining a motion trend distribution map based on the motion sensing data; The following formula is used to determine whether the overall coverage of the light area map to the motion trend distribution map meets the judgment condition: Among them, C represents the lighting coverage, I i represents the light intensity of the ith area, A i represents the area or weight of the ith region, n is the number of regions for parameter calculation; D represents the coverage of the movement trend, T j represents the motion trend value of the jth motion area, W j represents the area or weight of the jth motion area, and E represents the overall coverage of the light area distribution to the motion trend; If yes, extract data from the preset knowledge graph based on the light area graph and construct a subgraph as the target knowledge graph; If not, based on the motion trend graph, the character distribution graph is updated, and based on the updated character distribution graph, the lighting area graph is re-determined and the next round of iteration is performed.

2. The intelligent lighting control method for a smart nursing home according to claim 1, characterized in that: The step of extracting data from the preset knowledge graph based on the light area graph and constructing a subgraph includes: Based on the light area map, determining a target location where a light needs to be lit; Obtaining the installation locations of installed lamps in the preset knowledge graph; respectively calculating the distance between the installation position of the installed lamp and the target position where the lamp needs to be lit; Extract the installed lights corresponding to the installation positions whose distance is less than the preset value as nodes of the subgraph; Based on the nodes of the subgraph, edges between nodes corresponding to the nodes of the subgraph are extracted from the preset knowledge graph as edges of the subgraph.

3. The intelligent lighting control method for a smart nursing home according to claim 1, characterized in that: The updating of the character distribution map based on the motion trend map includes: Based on the exercise trend graph, determining a target population with an exercise trend; Acquire the portrait of the target group through an image sensor; Identify the person object in the person portrait and obtain the attribute characteristics of the person object; Determine a motion vector of each character object based on the attribute feature and the current action behavior of the character object; Based on the motion vector of each of the character objects, determining the motion end point position of each of the character objects; Based on the endpoint positions of the various character objects, an updated character distribution map is determined.

4. The intelligent lighting control method for a smart nursing home according to claim 3 is characterized in that: The step of determining the motion vector of each character object based on the attribute feature and the current action behavior of the character object comprises: Determining the movement speed of the person object based on the attribute feature; Determine a movement direction based on a current action of the character object; The motion vector is determined based on the motion speed and the motion direction.

5. The intelligent lighting control method for a smart nursing home according to claim 1 is characterized in that: The step of inputting the target knowledge graph into a graph processing model to generate a lighting control instruction includes: Acquire the living habits and physical condition data of the corresponding person object based on the target knowledge graph; Converting the life habits of the character object and the body state data into a first feature vector; The feature vector and the target knowledge graph are input into the graph processing model to generate the lighting control instruction.

6. The intelligent lighting control method for a smart nursing home according to claim 5, characterized in that: The sensor data further includes biometric sensor data, and the biometric sensor data includes a facial image; the method further includes: Based on the facial image of the person object, determining the facial micro-expression of the person object by image recognition; Converting the facial micro-expression into a second feature vector; The feature vector is used as input data of the image processing model to generate the light control instruction.

7. The intelligent lighting control method for a smart nursing home according to claim 5, characterized in that: The method further comprises: Based on the life habits and physical condition data of the character object, a weight coefficient is assigned to each character object; the weight coefficient reflects the acceptance degree of the character object to the light intensity; The weight coefficient is used as input data of the image processing model to generate the light control instruction.

8. The intelligent lighting control system for a smart nursing home according to claim 1 is characterized in that: The system comprises: A data acquisition module, used to collect sensor data in the smart nursing home; the sensor data includes light sensor data, sound sensor data, and motion sensor data; An iteration module, configured to perform one or more rounds of iterative updates on a preset knowledge graph based on the sensor data to obtain a target knowledge graph; the nodes of the preset knowledge graph include installed lights, activity areas, and rest areas; the edges of the knowledge graph include action paths between nodes; the attribute characteristics of the nodes include lamp type, lamp installation location, light intensity, and light intensity; the attribute characteristics of the edges include path attributes and path usage frequency; An instruction generation module, used for inputting the target knowledge graph into a graph processing model to generate a lighting control instruction; A lighting control module, used to adjust the lighting settings of the target area based on the lighting control instructions; Among them, one round of iteration includes: Determine a light intensity map of the smart nursing home based on the light sensing data; Determine a person distribution map in the smart nursing home based on the sound sensor data and the motion sensor data; Determine a lighting area map based on the light intensity map and the character distribution map; Determining a motion trend distribution map based on the motion sensing data; The following formula is used to determine whether the overall coverage of the light area map to the motion trend distribution map meets the judgment condition: Among them, C represents the lighting coverage, I i represents the light intensity of the ith area, A i represents the area or weight of the ith region, n is the number of regions for parameter calculation; D represents the coverage of the movement trend, T j represents the motion trend value of the jth motion area, W j represents the area or weight of the jth motion area, and E represents the overall coverage of the light area distribution to the motion trend; If yes, extract data from the preset knowledge graph based on the light area graph and construct a subgraph as the target knowledge graph; If not, based on the motion trend graph, the character distribution graph is updated, and based on the updated character distribution graph, the lighting area graph is re-determined and the next round of iteration is performed.

Images