A device linkage control method and system for a smart home scene

Abstract

The application is suitable for the technical field of linkage control, and particularly relates to a device linkage control method and system for smart home scenes, which comprises the following steps: constructing a use scene of a smart home device, determining a living subject, setting a living behavior of each living subject, selecting a preposed action, collecting control parameters of the smart home device under each living behavior, summarizing and generating parameter groups, establishing a corresponding relationship among the living subject, the living behavior and the parameter groups, and cutting the parameter groups into a pre-adjustment layer and a core layer. The application can adaptively adjust the linkage control of the smart home device by selecting a candidate path and a target path, enhance the adaptability to complex behaviors, multi-user differences and sudden interruption, improve the behavior prediction accuracy, realize the change from traditional rule driving to active perception, and comprehensively improve the intelligent level of the smart home device control and the efficiency of the device linkage control.

Description

Technical Field

[0001] This invention relates to the field of linkage control technology, and in particular to a device linkage control method and system for smart home scenarios. Background Technology

[0002] A smart home scenario refers to a device management method that connects smart terminal devices such as lights, air conditioners, door locks, and sensors in the home environment to a unified smart home platform via a wireless network.

[0003] In existing technologies, the management of smart home devices is generally based on passive response or manual control. This means that devices such as human body sensors, door and window sensors, and temperature and humidity sensors detect changes in the actual state and then execute corresponding actions according to pre-defined rules. However, they lack the ability to detect changes in advance. For example, after a user opens the bathroom door, the human body sensor detects someone entering and then activates the lighting, ventilation system, and smart toilet. But by this time, the user has already entered the bathroom, causing a delay in device activation and affecting the user experience.

[0004] Therefore, "how to control smart home devices in advance" is the technical problem that this invention aims to solve. Summary of the Invention

[0005] The purpose of this invention is to provide a device linkage control method and system for smart home scenarios, so as to solve the problem of "how to control smart home devices in advance" mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A device linkage control method for smart home scenarios, the method comprising: To construct usage scenarios for smart home devices, identify the residents, set the daily behaviors of each resident, select pre-actions, collect the control parameters of smart home devices under each daily behavior, summarize and generate parameter groups, establish the correspondence between residents, daily behaviors and parameter groups, and divide the parameter groups into a pre-adjustment layer and a core layer. By using video surveillance equipment pre-installed in the usage scenario, the real-time behavioral characteristics of each resident are collected. Through posture recognition, the resident corresponding to the real-time behavioral characteristics is determined. When the real-time behavioral characteristics are the same as the previous actions, the pre-adjustment layer in the corresponding parameter group is extracted, and the smart home devices are adjusted. Collect feedback information from the residents after adjustment, wherein the feedback information includes at least: voice feedback and motion feedback, determine whether to continue activating the core layer, and if not, revert the control parameters of the pre-adjusted layer; Integrating real-time behavioral features and feedback information, a behavioral chain is generated, with each resident subject corresponding to at least one behavioral chain. Using the resident subject as the parent node and the behavioral chain as the child node, a control binary tree is constructed. The application rules corresponding to each path are edited, and the feedback information during the rollback process is defined as an interruption feature. In the control binary tree, paths containing interruption features are searched to obtain candidate paths. Attribute data of the usage scenario at the current moment is collected, and the attribute data includes at least time and historical behavioral data. The candidate path with the highest similarity between the application rules and the attribute data is traversed to obtain the target path, and the parameter group corresponding to the target path is activated.

[0007] Furthermore, the steps of setting the daily behaviors of each resident, selecting pre-actions, collecting control parameters of smart home devices under each daily behavior, and summarizing to generate parameter groups include: Using the video surveillance equipment, video data in the usage scenario is collected, and segments containing the resident subjects are selected; According to a preset time interval, several snapshots are extracted from the segment to determine the preceding action.

[0008] Furthermore, the step of establishing the correspondence between the resident subject, living behavior, and parameter group, and dividing the parameter group into a pre-adjustment layer and a core layer, includes: Using sensing devices pre-deployed in the usage scenario, multi-source data is collected, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data; Integrate parameter groups and multi-source data to generate linkage control logs, summarize the log sets, and embed a rollback mechanism.

[0009] Furthermore, the step of extracting the pre-tuning layer from the corresponding parameter group and adjusting the smart home device when the real-time behavioral characteristics are the same as the previous actions includes: Record the adjustment time corresponding to the pre-adjustment layer, and simultaneously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident; Semantic analysis is performed on the audio data to generate analysis conclusions, and the control parameters are corrected, wherein the correction includes maintaining and reverting.

[0010] Furthermore, the step of collecting and adjusting the feedback information of the resident subject, wherein the feedback information includes at least: voice feedback and motion feedback, includes: Configure the risk level of the feedback information, wherein the risk level includes at least: high, medium and low; When the risk level of the feedback information is high, a takeover reminder for smart home devices will be sent to the resident.

[0011] Furthermore, the step of constructing a control binary tree with the resident entity as the parent node and the behavior chain as the child nodes includes: Establish a mapping between the inhabitants and the control binary trees, integrate the control binary trees of all inhabitants, and generate a control forest; Identify conflict items in the control forest, distribute them to the user terminals corresponding to the resident entities, and grant correction permissions.

[0012] Furthermore, the system includes: The segmentation module is used to construct the usage scenarios of smart home devices, determine the residents, set the living behaviors of each resident, select the pre-action, collect the control parameters of smart home devices under each living behavior, summarize and generate parameter groups, establish the correspondence between residents, living behaviors and parameter groups, and segment the parameter groups into a pre-adjustment layer and a core layer. The adjustment module is used to collect the real-time behavioral characteristics of each resident using video surveillance equipment pre-installed in the usage scenario. Through posture recognition, it determines the resident corresponding to the real-time behavioral characteristics. When the real-time behavioral characteristics are the same as the previous actions, it extracts the pre-adjustment layer from the corresponding parameter group and adjusts the smart home devices. The judgment module is used to collect feedback information from the residents after the adjustment, wherein the feedback information includes at least: voice feedback and action feedback, and to determine whether to continue activating the core layer. If not, the control parameters of the pre-adjustment layer are rolled back. The activation module integrates real-time behavioral features and feedback information to generate behavioral chains, where each resident entity corresponds to at least one behavioral chain. A control binary tree is constructed with the resident entity as the parent node and the behavioral chain as the child node. The application rules corresponding to each path are edited, and feedback information during the rollback process is defined as interruption features. Paths containing interruption features are found in the control binary tree to obtain candidate paths. Attribute data of the usage scenario at the current moment is collected, including at least time and historical behavioral data. The candidate paths with the highest similarity between the application rules and the attribute data are traversed to obtain the target path, and the parameter group corresponding to the target path is activated.

[0013] Furthermore, the segmentation module includes: The selection unit is used to collect video data in the usage scenario using the video surveillance equipment and select segments containing the resident subjects; The segmentation unit is used to extract several snapshots from the segment at preset time intervals to determine the preceding actions; The data acquisition unit is used to collect multi-source data using sensing devices pre-deployed in the usage scenario, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data. The aggregation unit is used to integrate parameter groups and multi-source data, generate linkage control logs, aggregate the log sets, and embed a rollback mechanism.

[0014] Furthermore, the adjustment module includes: The recording unit is used to record the adjustment time corresponding to the pre-adjustment layer and synchronously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident. The correction unit is used to perform semantic analysis on the audio data, generate analysis conclusions, and correct the control parameters, wherein the correction includes: maintaining and reverting.

[0015] Furthermore, the determination module includes: A configuration unit is used to configure the risk level of feedback information, wherein the risk level includes at least: high, medium and low. The dispatch unit is used to send a takeover reminder for smart home devices to the resident when the risk level of the feedback information is high.

[0016] Compared with the prior art, the beneficial effects of the present invention are: By setting pre-action parameters, behavioral representations can be provided to users, identifying their impending target behavior before their actual needs occur. This transforms smart home device control from "reactive response" to "predictive response," improving the lead time and accuracy of device linkage. A pre-tuning layer allows for progressive pre-response to smart home devices before the target behavior occurs, ensuring timely linkage while reducing resource waste from false triggers. A core layer, combined with the pre-tuning layer, further verifies the true intent based on subsequent behavior. Only after confirming the target behavior is achieved are the complete control parameters in the core layer activated, thereby improving the accuracy of behavior recognition. To ensure the accuracy and reliability of device control and avoid the problem of all devices starting up simultaneously due to misjudgment, a behavior chain is constructed to associate discrete user behaviors in chronological order, forming a continuous behavior evolution process. This intuitively displays the user intent corresponding to real-time behavioral characteristics and records the behavioral differences of different residents in different time periods, significantly improving the accuracy of scene recognition. By selecting candidate paths and target paths, the linkage control of smart home devices can be adaptively adjusted to improve the accuracy of behavior prediction, thereby realizing the transformation from traditional rule-driven to proactive perception and comprehensively improving the intelligence level of smart home device control and the efficiency of device linkage control. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a device linkage control method for smart home scenarios provided in an embodiment of the present invention; Figure 2This is a first sub-flowchart of a device linkage control method for smart home scenarios provided in an embodiment of the present invention. Figure 3 This is a second sub-flowchart of the device linkage control method for smart home scenarios provided in an embodiment of the present invention; Figure 4 This is a third sub-flowchart of the device linkage control method for smart home scenarios provided in an embodiment of the present invention; Figure 5 This is a fourth sub-flowchart of the device linkage control method for smart home scenarios provided in an embodiment of the present invention; Figure 6 This is a block diagram of the device linkage control system for smart home scenarios provided in an embodiment of the present invention; Figure 7 This is a structural block diagram of the segmentation module in a device linkage control system for smart home scenarios provided in an embodiment of the present invention; Figure 8 This is a structural block diagram of the adjustment module in a device linkage control system for smart home scenarios provided in an embodiment of the present invention; Figure 9 This is a block diagram illustrating the composition of a judgment module in a device linkage control system for smart home scenarios, as provided in an embodiment of the present invention. Figure 10 This is a structural block diagram of the activation module in a device linkage control system for smart home scenarios provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In Example 1, Figure 1 The implementation flow of the device linkage control method for smart home scenarios provided by an embodiment of the present invention is illustrated below, and is described in detail below: S100: Construct usage scenarios for smart home devices, identify the residents, set the daily behaviors of each resident, select pre-actions, collect control parameters of smart home devices under each daily behavior, summarize and generate parameter groups, establish the correspondence between residents, daily behaviors and parameter groups, and divide the parameter groups into a pre-adjustment layer and a core layer.

[0020] The process involves defining the usage scenarios for smart home devices, which can include home or office environments. It identifies users who reside in the home environment long-term, determining the corresponding resident entities, including male and female homeowners, the elderly, and children. A behavioral profile is created for each resident, including historical device usage records, video surveillance data, door lock opening records, human movement trajectories, and time patterns. Based on each resident's use of smart home devices within their chosen scenarios, their daily behaviors are determined. These behaviors include waking up, using the toilet, washing, cooking, resting, leaving home, and falling asleep. From the consecutive actions preceding each daily behavior, action features that indicate the impending target behavior are selected to obtain pre-actions, such as picking up a phone and opening the toilet door, turning off the TV and walking to the kitchen.

[0021] After a pre-action occurs, control parameters of smart home devices under the corresponding life behavior are collected. These control parameters include: on / off status, brightness, color temperature, temperature, fan speed, and speed. The control parameters of all smart home devices in the usage scenario under the corresponding life behavior are integrated to obtain a parameter group. The parameter group can be understood as a data set representing the operating status and control parameters of various smart devices in the usage scenario during the execution of a specific life behavior by the resident. Each resident corresponds to a parameter group under a single life behavior. Each parameter group is divided into a pre-adjustment layer and a core layer according to the importance of device linkage, energy consumption level, and trigger sensitivity. The pre-adjustment layer refers to the basic control parameters used for advance response. For example, in the life behavior of using the toilet, the control parameters such as turning off the bedroom lights and turning off the TV are in the pre-adjustment layer, while the pre-starting of the smart toilet, turning on the bathroom lights and ventilation system are in the core layer.

[0022] S200: Using video surveillance equipment pre-installed in the usage scenario, the real-time behavioral characteristics of each resident are collected. Through posture recognition, the resident corresponding to the real-time behavioral characteristics is determined. When the real-time behavioral characteristics are the same as the previous actions, the pre-adjustment layer in the corresponding parameter group is extracted, and the smart home devices are adjusted.

[0023] By using video surveillance equipment installed in the usage scenario, video data is collected, and human behavior feature extraction algorithms are used to extract action features from the video data, namely real-time behavior features. Basic information such as human contours, key skeletal points, and movement trajectories are extracted. Then, through human posture recognition algorithms, the resident corresponding to each real-time behavior feature is determined. When a real-time behavior feature is the same as a certain preceding action, the pre-adjustment layer data in the corresponding parameter group associated with the preceding action is retrieved, and the control parameters such as the operating status, working mode, and functional parameters of the smart home device are adjusted to the same state as the pre-adjustment layer.

[0024] S300: Collect feedback information from the resident after adjustment, wherein the feedback information includes at least: voice feedback and motion feedback, determine whether to continue activating the core layer, and if not, revert the control parameters of the pre-adjusted layer.

[0025] After adjusting smart home devices using the pre-tuning layer, feedback information from the resident is collected. This feedback includes voice commands and physical actions. By semantically parsing the voice feedback and recognizing posture features and analyzing behavioral logic in the action feedback, it's determined whether the resident has reversed the smart home device's state. If so, it indicates a mismatch between the pre-tuning layer's parameter adjustments and the resident's actual usage intentions. The corresponding control parameters of the pre-tuning layer are then reverted, restoring the currently effective pre-tuning layer control parameters to their initial state before device adjustment—for example, closing a smart toilet seat that just automatically opened. If the resident hasn't adjusted the smart home device or has made adjustments consistent with the current control direction, it indicates that the pre-tuning layer's parameter adjustments match the resident's actual usage intentions. The control parameters corresponding to the core layer are then activated, and adjustments to the smart home device continue.

[0026] S400: Integrate real-time behavioral features and feedback information to generate behavioral chains, where each resident entity corresponds to at least one behavioral chain. With the resident entity as the parent node and the behavioral chain as the child node, construct a control binary tree, edit the application rules corresponding to each path, define the feedback information during the rollback process as interruption features, find paths containing interruption features in the control binary tree to obtain candidate paths, collect attribute data of the usage scenario at the current moment, the attribute data includes at least: time and historical behavioral data, traverse the candidate paths with the highest similarity between the application rules and attribute data to obtain the target path, and activate the parameter group corresponding to the target path.

[0027] Following a chronological order, the real-time behavioral characteristics and feedback information corresponding to each resident are correlated and fused to generate a behavior chain. A behavior chain is a continuous set of action characteristics of the same resident arranged chronologically. Within a time period (e.g., one day), the same resident can correspond to multiple behavior chains. Parent nodes are created, each corresponding to a resident, and child nodes are created, each corresponding to a behavior chain. The child nodes are then attached to the parent nodes, generating a control binary tree. Both parent and child nodes are logical nodes, used only to represent the resident and the behavior chain. The control binary tree is similar to a binary tree in existing technologies. In this embodiment, by constructing a control binary tree, the behavioral characteristics corresponding to different residents can be intuitively displayed. By retrieving preceding actions from the control binary tree, predictive control of smart home devices can be quickly achieved. For each path in the control binary tree, at least one corresponding application rule is set. For example, when the resident is the male homeowner and the behavior chain is opening the door and entering the kitchen, the application rule corresponding to this path is 11:00-12:00, with a stay time exceeding 5 minutes. In other words, if the male homeowner is detected entering the kitchen during this time period and staying there for more than 5 minutes, it indicates that the male homeowner may be cooking, and the system will trigger the control to turn on the indoor air purifier, etc. If the male homeowner leaves the kitchen and turns on the TV, the feedback information (interruption feature) is that the TV is turned on. In the control binary tree, other paths containing the action of turning on the TV are searched, and the searched paths are defined as candidate paths. Attribute data is collected based on usage scenarios. This data includes time information (such as time periods, holiday attributes, and day-night cycles) and historical behavior data (such as user activity trajectories, device operation records, behavior frequency, and duration within similar time periods). Multi-dimensional alignment is performed using cosine similarity, weighted matching, or machine learning models to identify the candidate path with the highest similarity between the current attribute data and the application rules. This candidate path is defined as the target path. The smart home devices are adjusted using the pre-tuning layer in the target path, and further adjustments are made using the core layer after the initial adjustment.

[0028] For example, an application rule states: from 23:00 to 6:00 the next day, when the bedroom light is turned on, the bathroom light should also be turned on. If the current time happens to be within the above time period and the bedroom light is on, the candidate path corresponding to the application rule is defined as the target path through cosine similarity calculation and semantic analysis, and the bathroom light is turned on accordingly.

[0029] In Example 2, Figure 2 The first sub-flowchart of the device linkage control method for smart home scenarios provided by an embodiment of the present invention is shown. The following details the steps of setting the daily life behavior of each resident, selecting the preceding action, collecting the control parameters of smart home devices under each daily life behavior, and summarizing and generating a parameter group: S101: Using the video surveillance equipment, collect video data in the usage scenario and select segments containing the resident subjects.

[0030] By using video surveillance equipment to collect video data in the usage scenario, and through human detection, target tracking and identity recognition algorithms, the system identifies and locates the human targets in the video data, and filters out segments containing the activity trajectory of the residents from all the video data.

[0031] S102: According to the preset time interval, extract several snapshots from the segment and determine the preceding action.

[0032] According to a preset time interval (e.g., 1 second), several video frames containing the resident subject are extracted from the clip to obtain snapshots, and the preceding action is selected from the snapshots.

[0033] In Example 3, Figure 2 This diagram illustrates the first sub-flowchart of a device linkage control method for smart home scenarios provided by an embodiment of the present invention. The following details the steps of establishing the correspondence between the resident subject, living behavior, and parameter groups, and dividing the parameter groups into a pre-adjustment layer and a core layer: S103: Collect multi-source data using sensing devices pre-deployed in the usage scenario, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data.

[0034] Multi-source data is collected by using sensing devices installed in different areas of the usage scenario.

[0035] S104: Integrate parameter groups and multi-source data to generate linkage control logs, summarize the log sets, and embed a rollback mechanism.

[0036] When the control parameters of smart home devices change, multi-source data at the corresponding time is collected, and the control parameters and multi-source data before and after the adjustment are integrated to generate linkage control logs. Each linkage control log records the trigger source, device linkage sequence, and parameter change process. The linkage control logs are sorted in chronological order to obtain a log set, and a rollback mechanism is embedded. The rollback mechanism means that the control parameters of smart home devices can be restored to the parameter state corresponding to a specified historical time according to user needs.

[0037] In Example 4, Figure 3 The second sub-flowchart of the device linkage control method for smart home scenarios provided by an embodiment of the present invention is shown. The following details the step of extracting the pre-tuning layer from the corresponding parameter group and adjusting the smart home device when the real-time behavior characteristics are the same as the previous action: S201: Record the adjustment time corresponding to the pre-adjustment layer, and simultaneously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident.

[0038] Record the time it takes to adjust smart home devices using the control parameters corresponding to the pre-tuning layer, i.e., the adjustment time. Use the sound pickup device installed in the usage scenario to collect the audio data of the residents in the usage scenario.

[0039] S202: Perform semantic analysis on the audio data, generate analysis conclusions, and correct the control parameters, wherein the correction includes: maintaining and reverting.

[0040] The audio data undergoes speech recognition, semantic parsing, and sentiment analysis to determine the resident's true intentions and generate analysis conclusions. These conclusions should contain keywords with clear sentimental connotations, such as confirmation of satisfaction and negative feedback. Based on the analysis results, control parameters are adjusted, including maintaining and reverting. For example, if the analysis conclusion indicates that the resident approves of the current device state (i.e., the "maintain current state" analysis conclusion is identified), the existing control parameters are maintained and the current target path continues to be executed. However, if the analysis conclusion indicates that the user rejects the current interactive behavior (e.g., the analysis conclusions such as "I don't need to use the toilet" or "I don't need to turn it so bright"), a reverting mechanism is triggered, restoring the control parameters to the previous state.

[0041] In Example 5, Figure 4 The diagram illustrates the third sub-flowchart of the device linkage control method for smart home scenarios provided by an embodiment of the present invention. The following details the step of collecting and adjusting the feedback information of the resident, wherein the feedback information includes at least voice feedback and motion feedback, as follows: S301: Configure the risk level of the feedback information, wherein the risk level includes at least: high, medium and low.

[0042] Extract sentiment words from the feedback information. For example, if the feedback information is "Turn off the living room lights quickly", the sentiment word is "quickly". Build a sentiment word dictionary, assign a corresponding sentiment score to each sentiment word, and sum the sentiment scores of all sentiment words in the feedback information. Divide the sum into multiple intervals, with each interval corresponding to a different risk level. This risk level is the risk level of the feedback information.

[0043] S302: When the risk level of the feedback information is high, a takeover reminder for smart home devices will be sent to the resident.

[0044] When the risk level of the feedback information is high, a takeover reminder for smart home devices will be sent to the resident, who can then directly adjust the smart home devices through voice dialogue or platform control.

[0045] In Example 6, Figure 5 The fourth sub-flowchart of the device linkage control method for smart home scenarios provided by an embodiment of the present invention is shown. The steps of constructing a control binary tree with the resident as the parent node and the behavior chain as the child nodes are described in detail below: S401: Establish a mapping between the resident entities and the control binary tree, integrate the control binary trees of all resident entities, and generate a control forest.

[0046] A corresponding control binary tree is constructed for each resident. These control binary trees are then integrated across all resident entities within the same smart home scenario to generate a control forest, which is essentially a collection of control binary trees. The advantage of this approach is that it enables centralized management and correlation analysis of the control logic for multiple resident entities within the same smart home environment.

[0047] S402: Locate the conflict items in the control forest, send them to the user terminals corresponding to the resident entities, and grant correction permissions.

[0048] The system traverses the control forest to identify conflicting commands. Conflicting commands refer to opposing control instructions from different residents for the same smart home device. For example, when the male homeowner is watching a movie in the living room, his path might send a control command to the living room air conditioner to "turn off the main living room light" to create an immersive viewing environment. Meanwhile, the female homeowner is reading in the living room, and her target path might include a control command to "turn on the main light to high brightness mode." These are conflicting commands. Once a conflict is detected, it is sent to the user terminal corresponding to the resident for manual confirmation.

[0049] Figure 6 This diagram illustrates the structural block diagram of a device linkage control system for smart home scenarios provided in an embodiment of the present invention. The device linkage control system 1 for smart home scenarios includes: The segmentation module 11 is used to construct the usage scenarios of smart home devices, determine the residents, set the living behaviors of each resident, select the pre-action, collect the control parameters of smart home devices under each living behavior, summarize and generate parameter groups, establish the correspondence between residents, living behaviors and parameter groups, and segment the parameter groups into a pre-adjustment layer and a core layer. The adjustment module 12 is used to collect the real-time behavioral characteristics of each resident using video surveillance equipment pre-installed in the usage scenario, determine the resident corresponding to the real-time behavioral characteristics through posture recognition, and extract the pre-adjustment layer from the corresponding parameter group when the real-time behavioral characteristics are the same as the previous actions, and adjust the smart home devices accordingly. The judgment module 13 is used to collect feedback information of the resident subject after adjustment, wherein the feedback information includes at least: voice feedback and action feedback, and to determine whether to continue to activate the core layer. If not, the control parameters of the pre-adjustment layer are rolled back. Activation module 14 is used to integrate real-time behavioral features and feedback information to generate behavioral chains, where each resident subject corresponds to at least one behavioral chain. With the resident subject as the parent node and the behavioral chain as the child node, a control binary tree is constructed. The application rules corresponding to each path are edited, and the feedback information during the rollback process is defined as interruption features. In the control binary tree, paths containing interruption features are found to obtain candidate paths. Attribute data of the usage scenario at the current moment is collected. The attribute data includes at least time and historical behavioral data. The candidate paths with the highest similarity between the application rules and the attribute data are traversed to obtain the target path, and the parameter group corresponding to the target path is activated.

[0050] Figure 7 This diagram illustrates the composition of a segmentation module 11 in a device linkage control system for smart home scenarios provided in an embodiment of the present invention. The segmentation module 11 includes: The selection unit 111 is used to collect video data in the usage scenario using the video surveillance equipment and select segments containing the resident subjects; The segmentation unit 112 is used to extract several snapshots from the segment according to a preset time interval to determine the preceding action; The acquisition unit 113 is used to acquire multi-source data using sensing devices pre-deployed in the usage scenario, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data. The aggregation unit 114 is used to integrate parameter groups and multi-source data, generate linkage control logs, aggregate the log sets, and embed a rollback mechanism.

[0051] Figure 8 This diagram illustrates the structural composition of the adjustment module 12 in a device linkage control system for smart home scenarios provided by an embodiment of the present invention. The adjustment module 12 includes: Recording unit 121 is used to record the adjustment time corresponding to the pre-adjustment layer and synchronously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident. The correction unit 122 is used to perform semantic analysis on the audio data, generate analysis conclusions, and correct the control parameters, wherein the correction includes: maintaining and reverting.

[0052] Figure 9 This diagram illustrates the structural composition of the judgment module 13 in a device linkage control system for smart home scenarios provided by an embodiment of the present invention. The judgment module 13 includes: Configuration unit 131 is used to configure the risk level of feedback information, wherein the risk level includes at least: high, medium and low; Unit 132 is used to send a takeover reminder for smart home devices to the resident when the risk level of the feedback information is high.

[0053] Figure 10 This diagram illustrates the structural composition of the activation module 14 in a device linkage control system for smart home scenarios provided in an embodiment of the present invention. The activation module 14 includes: Mapping unit 141 is used to establish a mapping between the resident entities and the control binary tree, integrate the control binary trees of all resident entities, and generate a control forest. Open unit 142 is used to find conflict items in the control forest, send them to the user terminal corresponding to the resident subject, and grant correction permissions.

[0054] The segmentation module 11 is mainly used to complete step S100, the adjustment module 12 is mainly used to complete step S200, the judgment module 13 is mainly used to complete step S300, and the activation module 14 is mainly used to complete step S400. The selection unit 111 is mainly used to complete step S101, the interception unit 112 is mainly used to complete step S102, the acquisition unit 113 is mainly used to complete step S103, and the summarization unit 114 is mainly used to complete step S104. Recording unit 121 is mainly used to complete step S201, and correction unit 122 is mainly used to complete step S202; Configuration unit 131 is mainly used to complete step S301, and distribution unit 132 is mainly used to complete step S302; The mapping unit 141 is mainly used to complete step S401, and the opening unit 142 is mainly used to complete step S402.

[0055] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0056] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A device linkage control method for smart home scenarios, characterized in that, The method includes: To construct usage scenarios for smart home devices, identify the residents, set the daily behaviors of each resident, select pre-actions, collect the control parameters of smart home devices under each daily behavior, summarize and generate parameter groups, establish the correspondence between residents, daily behaviors and parameter groups, and divide the parameter groups into a pre-adjustment layer and a core layer. By using video surveillance equipment pre-installed in the usage scenario, the real-time behavioral characteristics of each resident are collected. Through posture recognition, the resident corresponding to the real-time behavioral characteristics is determined. When the real-time behavioral characteristics are the same as the previous actions, the pre-adjustment layer in the corresponding parameter group is extracted, and the smart home devices are adjusted. Collect feedback information from the residents after adjustment, wherein the feedback information includes at least: voice feedback and motion feedback, determine whether to continue activating the core layer, and if not, revert the control parameters of the pre-adjusted layer; Integrating real-time behavioral features and feedback information, a behavioral chain is generated, with each resident subject corresponding to at least one behavioral chain. Using the resident subject as the parent node and the behavioral chain as the child node, a control binary tree is constructed. The application rules corresponding to each path are edited, and the feedback information during the rollback process is defined as an interruption feature. In the control binary tree, paths containing interruption features are searched to obtain candidate paths. Attribute data of the usage scenario at the current moment is collected, and the attribute data includes at least time and historical behavioral data. The candidate path with the highest similarity between the application rules and the attribute data is traversed to obtain the target path, and the parameter group corresponding to the target path is activated.

2. The device linkage control method for smart home scenarios according to claim 1, characterized in that, The steps of setting the daily behavior of each resident, selecting pre-actions, collecting control parameters of smart home devices under each daily behavior, and summarizing and generating parameter groups include: Using the video surveillance equipment, video data in the usage scenario is collected, and segments containing the resident subjects are selected; According to a preset time interval, several snapshots are extracted from the segment to determine the preceding action.

3. The device linkage control method for smart home scenarios according to claim 1, characterized in that, The steps of establishing the correspondence between the resident subject, living behavior, and parameter group, and dividing the parameter group into a pre-adjustment layer and a core layer, include: Using sensing devices pre-deployed in the usage scenario, multi-source data is collected, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data; Integrate parameter groups and multi-source data to generate linkage control logs, summarize the log sets, and embed a rollback mechanism.

4. The device linkage control method for smart home scenarios according to claim 3, characterized in that, The step of extracting the pre-tuning layer from the corresponding parameter group and adjusting the smart home device when the real-time behavioral characteristics are the same as the previous actions includes: Record the adjustment time corresponding to the pre-adjustment layer, and simultaneously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident; Semantic analysis is performed on the audio data to generate analysis conclusions, and the control parameters are corrected, wherein the correction includes maintaining and reverting.

5. The device linkage control method for smart home scenarios according to claim 1, characterized in that, The step of collecting and adjusting the feedback information of the resident, wherein the feedback information includes at least voice feedback and motion feedback, includes: Configure the risk level of the feedback information, wherein the risk level includes at least: high, medium and low; When the risk level of the feedback information is high, a takeover reminder for smart home devices will be sent to the resident.

6. The device linkage control method for smart home scenarios according to claim 5, characterized in that, The steps for constructing a control binary tree with the resident entity as the parent node and the behavior chain as the child nodes include: Establish a mapping between the inhabitants and the control binary trees, integrate the control binary trees of all inhabitants, and generate a control forest; Identify conflict items in the control forest, distribute them to the user terminals corresponding to the resident entities, and grant correction permissions.

7. A device linkage control system for smart home scenarios, characterized in that, The system includes: The segmentation module is used to construct the usage scenarios of smart home devices, determine the residents, set the living behaviors of each resident, select the pre-action, collect the control parameters of smart home devices under each living behavior, summarize and generate parameter groups, establish the correspondence between residents, living behaviors and parameter groups, and segment the parameter groups into a pre-adjustment layer and a core layer. The adjustment module is used to collect the real-time behavioral characteristics of each resident using video surveillance equipment pre-installed in the usage scenario. Through posture recognition, it determines the resident corresponding to the real-time behavioral characteristics. When the real-time behavioral characteristics are the same as the previous actions, it extracts the pre-adjustment layer from the corresponding parameter group and adjusts the smart home devices. The judgment module is used to collect feedback information from the residents after the adjustment, wherein the feedback information includes at least: voice feedback and action feedback, and to determine whether to continue activating the core layer. If not, the control parameters of the pre-adjustment layer are rolled back. The activation module integrates real-time behavioral features and feedback information to generate behavioral chains, where each resident entity corresponds to at least one behavioral chain. A control binary tree is constructed with the resident entity as the parent node and the behavioral chain as the child node. The application rules corresponding to each path are edited, and feedback information during the rollback process is defined as interruption features. Paths containing interruption features are found in the control binary tree to obtain candidate paths. Attribute data of the usage scenario at the current moment is collected, including at least time and historical behavioral data. The candidate paths with the highest similarity between the application rules and the attribute data are traversed to obtain the target path, and the parameter group corresponding to the target path is activated.

8. The device linkage control system for smart home scenarios according to claim 7, characterized in that, The segmentation module includes: The selection unit is used to collect video data in the usage scenario using the video surveillance equipment and select segments containing the resident subjects; The segmentation unit is used to extract several snapshots from the segment at preset time intervals to determine the preceding actions; The data acquisition unit is used to collect multi-source data using sensing devices pre-deployed in the usage scenario, wherein the multi-source data includes at least: human body sensing data and door and window opening and closing sensing data. The aggregation unit is used to integrate parameter groups and multi-source data, generate linkage control logs, aggregate the log sets, and embed a rollback mechanism.

9. The device linkage control system for smart home scenarios according to claim 8, characterized in that, The adjustment module includes: The recording unit is used to record the adjustment time corresponding to the pre-adjustment layer and synchronously start the sound pickup device pre-installed in the usage scenario to collect the audio data of the resident. The correction unit is used to perform semantic analysis on the audio data, generate analysis conclusions, and correct the control parameters, wherein the correction includes: maintaining and reverting.

10. The device linkage control system for smart home scenarios according to claim 7, characterized in that, The judgment module includes: A configuration unit is used to configure the risk level of feedback information, wherein the risk level includes at least: high, medium and low. The dispatch unit is used to send a takeover reminder for smart home devices to the resident when the risk level of the feedback information is high.

Images