An edge-computing-based intelligent home screen local control system and method

Abstract

The application discloses an intelligent home screen local control system and method based on edge computing, and belongs to the technical field of home equipment control. The method forms a control network through building a home scene edge computing node and a local control link of the intelligent home screen, records link transmission states and marks instruction behaviors, generates a bimodal control feedback information set, calculates a characteristic value based on the feedback set and evaluates instruction priority, and outputs a control sequence in order. The system corresponds to control link building, state recording, feedback set generation, priority evaluation and a main control module. The application reduces cloud dependence through local link building, realizes efficient scheduling of instructions through bimodal feedback and priority evaluation, solves the problems of existing intelligent home control response lag and unreasonable resource allocation, and improves the real-time performance and reliability of intelligent equipment control in a home scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of home equipment control, in particular to a smart home screen local control system and method based on edge computing. BACKGROUND

[0002] With the rapid development of smart home technology, the number of smart devices in the home scene is increasing year by year. The control response efficiency and stability of the smart home screen, as the core control terminal, directly affect the user experience. Existing smart home control solutions rely on cloud servers to process control instructions. When the network fluctuates or is interrupted, the transmission delay of control instructions increases or even fails, resulting in device response lag. At the same time, the traditional control scheme does not dynamically optimize the resource allocation of the edge computing node. When multiple devices simultaneously initiate control requests, node resource preemption conflicts may occur, further reducing control reliability.

[0003] In addition, the existing technology lacks precise monitoring and feedback mechanism for control instruction response state. When the smart home screen does not respond to the control instruction, it is difficult to quickly locate the root cause of the problem (such as link failure or node resource shortage), and manual troubleshooting is required, increasing maintenance costs. Although some local control solutions reduce cloud dependence, they do not establish a scientific instruction scheduling mechanism, and the instruction execution order is chaotic, making it difficult to prioritize critical control requirements (such as security device alarm instructions) and meet the diverse and high real-time control requirements in the home scene. Therefore, there is an urgent need for a local control scheme based on edge computing to achieve efficient instruction transmission, dynamic resource allocation and precise scheduling, and improve the control performance of the smart home system. SUMMARY

[0004] The present application aims to provide a smart home screen local control system and method based on edge computing to solve the problems raised in the background art. Based on edge computing technology, edge computing nodes are deployed in the home scene and local instruction protocol control links are built to construct a decentralized smart home edge control network. Control instructions can be transmitted locally without going through the cloud, shortening the transmission path (corresponding to step S1). By recording link state parameters (including instruction characteristics and node resource information), the whole link monitoring of the instruction transmission process is realized (corresponding to step S2). Based on the dual-mode control feedback information set, the instruction behavior characteristics are captured from the "link-node" and "node-device" two dimensions respectively. By calculating the calling degree and resource cache degree characteristic values, the priority of instruction execution is accurately evaluated to ensure the priority execution of critical instructions (corresponding to steps S3 and S4).

[0005] To solve the above technical problems, the present application provides the following technical solutions:

[0006] The application discloses an edge-computing-based intelligent home screen local control system, which comprises a control link construction and instruction processing module, a link state recording and marking module, a dual-mode feedback set generation module, an instruction priority evaluation and sequence output module and a main control unit for coordinating data interaction and process control of the modules, so as to realize the whole-process local control from the control link construction to the instruction sequence output.

[0007] The control link construction and instruction processing module is used for constructing a local instruction protocol control link of an edge computing node and an intelligent home screen and forming a control network, processing a control request instruction and extracting feature information when the control request instruction is not responded.

[0008] The link state recording and marking module is used for recording state parameters when the control link transmits an instruction, locking a response object and an associated node and completing instruction behavior marking.

[0009] The dual-mode feedback set generation module is used for generating a dual-mode control feedback information set based on an instruction behavior, and determining two types of pointing features corresponding to the instruction behavior.

[0010] The instruction priority evaluation and sequence output module is used for calculating two types of pointing feature values, evaluating instruction behavior priority and generating a control sequence output in sequence.

[0011] Preferably, the control link construction and instruction processing module comprises a control network construction unit and an instruction feature extraction unit.

[0012] The control network construction unit is used for constructing a local instruction protocol control link between edge computing nodes with control response capability in a home scene, realizing the connection of the nodes and the intelligent home screen and forming a control network.

[0013] The instruction feature extraction unit is used for monitoring the response state of a control request instruction in the control network, and extracting a response object and a response waiting time length for a non-responded instruction.

[0014] Preferably, the link state recording and marking module comprises a state parameter recording unit and an instruction behavior marking unit.

[0015] The state parameter recording unit is used for recording state parameters of an edge computing node calling link when the control link carries a control request instruction, including instruction feature information and node resource cache information.

[0016] The instruction behavior marking unit is used for locking a response object and an associated edge computing node based on the state parameters, and marking the instruction behavior of each control link with the node as an instruction task center.

[0017] Preferably, the dual-mode feedback set generation module comprises a first mode set generation unit and a second mode set generation unit.

[0018] The first modal set generation unit is configured to collect information of edge computing nodes connected to the same control link, and generate a first modal control feedback information set to determine link-node pointing features of the instruction behavior.

[0019] The second modal set generation unit is configured to collect information of smart home screens involved in the instruction behavior, and generate a second modal control feedback information set to determine node-device pointing features of the instruction behavior.

[0020] Preferably, the instruction priority evaluation and sequence output module comprises a feature value calculation unit and a control sequence generation unit.

[0021] The feature value calculation unit is configured to calculate calling degree feature values of the edge computing nodes and the control link based on the first modal feedback set, and calculate resource caching degree feature values of the nodes to the smart home screens based on the second modal feedback set.

[0022] The control sequence generation unit is configured to calculate the instruction behavior priority by multiplying the two types of feature values, and generate and output the instruction behavior control sequence in descending order of priority.

[0023] A smart home screen local control method based on edge computing, comprising the following steps:

[0024] Step S1: Building a local instruction protocol control link between edge computing nodes and smart home screens in a home scene to form a smart home edge control network, and extracting corresponding feature information in the case of unresponsive control request instructions.

[0025] Step S2: Recording link state parameters when transmitting control request instructions through the local instruction protocol control link, locking the response object and associated edge computing nodes based on the link state parameters, and marking the instruction behavior of each control link.

[0026] Step S3: Generating a dual-modal control feedback information set based on the marked instruction behavior, and determining the link-node pointing features, i.e., the first pointing features, and the node-device pointing features, i.e., the second pointing features, corresponding to the instruction behavior.

[0027] Step S4: Calculating the corresponding pointing feature values based on the dual-modal control feedback information set, evaluating the instruction behavior priority through the pointing feature values, and generating and outputting the instruction behavior control sequence in descending order of priority.

[0028] Preferably, the specific implementation process of step S1 comprises:

[0029] In the same home scenario, local command protocol control links are established between edge computing nodes with control and response capabilities, and each edge computing node is connected to each smart home screen through the local command protocol control links to form a smart home edge control network.

[0030] The smart home edge control network generates control request commands for each smart home screen. If each smart home screen does not respond to the control request command within a preset time, feature information is extracted from the control request command. The feature information includes the response object and the response waiting time of the response object. The response object refers to each smart home screen that the control request command is intended to control.

[0031] Preferably, the specific implementation process of step S2 includes:

[0032] When the local command protocol control link carries a control request command, it records the link status parameters used by the edge computing node to transmit the control request command by calling the local command protocol control link. The link status parameters include the characteristic information of the control request command and the resource cache information of the edge computing node.

[0033] Based on link state parameters, the system locks the response object and the connected edge computing nodes in the local command protocol control link by controlling request commands. Using the edge computing node as the command task center, it marks the command behavior of each local command protocol control link, denoting the i-th edge computing node as... Let the r-th local command protocol control link be denoted as Let the e-th smart home screen be denoted as Edge computing nodes Connecting local command protocol control link Time triggers smart home screen Control request instructions, marked as instruction behavior. , where x is the unique attribute code number of the control request instruction.

[0034] Preferably, the specific implementation process of step S3 includes:

[0035] Based on command behavior, a dual-modal control feedback information set is generated. The first modal control feedback information set is used to collect data from each edge computing node connected to the same local command protocol control link to determine command behavior. First pointing feature The second modal control feedback information set is used to collect smart home data on command behavior in order to determine the command behavior. Second pointing feature ;

[0036] The link will be controlled by the local command protocol. The generated first modal control feedback information set is denoted as The second mode control feedback information set generated by the edge computing nodes is denoted as... Where I represents the encoding sequence number of the edge computing node and E represents the encoding sequence number of the smart home screen.

[0037] Preferably, the specific implementation process of step S4 includes:

[0038] Based on the first mode control feedback information set, evaluate the edge computing node. Connecting local command protocol control link The first pointing characteristic value at time is used to reflect the connection of the local command protocol control link. Edge computing nodes call degree In the formula, Let the counting representation function be used, if the edge computing nodes are... Existing in the first mode control feedback information set In the middle, then order If edge computing nodes Not present in the first mode control feedback information set In the middle, then order R represents the total number of local command protocol control links;

[0039] Based on the second-modal control feedback information set, evaluate the command behavior. The second pointing characteristic value is used to reflect the edge computing node's response when the smart home screen does not respond to control request commands within a preset time. Resource caching level for various smart home screens , Represents edge computing nodes Smart Home Screen Service The response time of smart home screens;

[0040] Evaluate instruction behavior based on the first and second pointer feature values. priority Generate and output the control sequence of instruction behaviors in descending order of priority.

[0041] Compared with the prior art, the present application has the beneficial effects that: by building a local control link of a home scene edge computing node and a smart home screen to form a control network, recording link transmission state and marking instruction behavior, generating a dual-mode control feedback information set, calculating a characteristic value based on the feedback set and evaluating the instruction priority, and outputting a control sequence in order, the present application reduces the dependence on the cloud through the local link, realizes efficient scheduling of instructions through dual-mode feedback and priority evaluation, solves the problems of response lag and unreasonable resource allocation of existing smart home control, improves the real-time performance and reliability of smart device control in a home scene, and at the same time, through link state recording and dual-mode feedback, the root cause (link failure / node resource shortage) of the unresponsive instruction can be quickly located. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application together with the embodiments thereof, and explain the principles of the present application, and are used to explain the present application but do not limit the present application in any way.

[0043] Figure 1 is a step schematic diagram of a smart home screen local control method based on edge computing. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0045] In the first embodiment: a smart home screen local control system based on edge computing is provided, which comprises: a control link building and instruction processing module, a link state recording and marking module, a dual-mode feedback set generation module, an instruction priority evaluation and sequence output module, and a main control unit for coordinating data interaction and process control of each module, to realize the whole process of local control from control link building to instruction sequence output;

[0046] The control link building and instruction processing module is used to build a local instruction protocol control link of an edge computing node and a smart home screen and form a control network, and to process control request instructions and feature information extraction when unresponsive;

[0047] The control link building and instruction processing module comprises a control network construction unit and an instruction feature extraction unit.

[0048] The control network construction unit is configured to build a local instruction protocol control link among edge computing nodes with control response capability in a home scene, realize the connection of each node with a smart home screen, and form a control network.

[0049] The instruction feature extraction unit is configured to monitor the response state of a control request instruction in the control network, and extract a response object and a response waiting time for an unresponsive instruction.

[0050] The link state recording and marking module is configured to record state parameters when the control link transmits an instruction, lock a response object and associated nodes, and complete instruction behavior marking.

[0051] The link state recording and marking module includes a state parameter recording unit and an instruction behavior marking unit.

[0052] The state parameter recording unit is configured to record state parameters of an edge computing node calling a link when the control link carries a control request instruction, including instruction feature information and node resource cache information.

[0053] The instruction behavior marking unit is configured to lock a response object and associated edge computing nodes based on state parameters, and mark the instruction behavior of each control link with the node as the instruction task center.

[0054] The dual-modal feedback set generation module is configured to generate a dual-modal control feedback information set based on the instruction behavior, and determine two types of pointing features corresponding to the instruction behavior.

[0055] The dual-modal feedback set generation module includes a first modal set generation unit and a second modal set generation unit.

[0056] The first modal set generation unit is configured to collect information of each edge computing node connected to the same control link, generate a first modal control feedback information set, and determine a link-node pointing feature of the instruction behavior.

[0057] The second modal set generation unit is configured to collect smart home screen information involved in the instruction behavior, generate a second modal control feedback information set, and determine a node-device pointing feature of the instruction behavior.

[0058] The instruction priority evaluation and sequence output module is configured to calculate two types of pointing feature values, evaluate the priority of the instruction behavior, and generate a control sequence output in sequence.

[0059] The instruction priority evaluation and sequence output module includes a feature value calculation unit and a control sequence generation unit.

[0060] The feature value calculation unit is configured to calculate a calling degree feature value of an edge computing node and a control link based on the first modal feedback set, and calculate a resource cache degree feature value of a node to a smart home screen based on the second modal feedback set.

[0061] The control sequence generation unit is configured to calculate the priority of the instruction behavior by the two types of eigenvalue product, generate the instruction behavior control sequence according to the priority from high to low, and output the instruction behavior control sequence.

[0062] Referring to Figure 1 In the second embodiment, a local control method of smart home screens based on edge computing is provided, which is applicable to the first embodiment. The second embodiment takes a three-room and two-hall family residence as an application scenario. Five edge computing nodes (located in the living room, master bedroom, secondary bedroom, kitchen, and bathroom, respectively) and eight smart home screens (corresponding to smart device control terminals in each room) are deployed in the scenario. The scenario covers 15 smart devices such as smart lighting, air conditioning, curtains, and security cameras, and needs to realize local and efficient scheduling and response of device control instructions. The preset control instruction response timeout time is 2 seconds, and the instruction priority threshold is 0.6 (higher than the threshold is considered as a high-priority instruction).

[0063] The method comprises the following steps:

[0064] Step S1: Build a local instruction protocol control link between edge computing nodes and smart home screens in a home scenario to form a smart home edge control network, and extract corresponding feature information for unresponsive control request instructions.

[0065] Specifically, a local instruction protocol control link is built between edge computing nodes with control response capabilities in the same home scenario, and each edge computing node is connected to each smart home screen through the local instruction protocol control link to form a smart home edge control network.

[0066] The smart home edge control network generates control request instructions for each smart home screen. If each smart home screen does not respond to the control request instructions within a preset time, the feature information of the control request instructions is extracted, including the response object and the response waiting time of the response object. The response object refers to each smart home screen controlled by the control request instruction.

[0067] For example, a local instruction protocol control link is built between five edge computing nodes in an Ethernet and Wi-Fi 6 combination manner. Each node is connected to the corresponding regional smart home screen to form a smart home edge control network. Control request instructions such as light adjustment and temperature control are sent to eight smart home screens to simulate three instruction unresponsive scenarios (response waiting time is 2.5 seconds, 3 seconds, and 2.2 seconds, respectively). The response object (corresponding to three smart home screens) and the response waiting time feature information are extracted.

[0068] Step S2: Record the link state parameters when transmitting the control request instructions through the local instruction protocol control link, lock the response object and the associated edge computing node based on the link state parameters, and mark the instruction behavior of each control link.

[0069] Specifically, when the local command protocol control link carries a control request command, it records the link status parameters used by the edge computing node to transmit the control request command. The link status parameters include the characteristic information of the control request command and the resource cache information of the edge computing node.

[0070] Based on link state parameters, the system locks the response object and the connected edge computing nodes in the local command protocol control link by controlling request commands. Using the edge computing node as the command task center, it marks the command behavior of each local command protocol control link, denoting the i-th edge computing node as... Let the r-th local command protocol control link be denoted as Let the e-th smart home screen be denoted as Edge computing nodes Connecting local command protocol control link Time triggers smart home screen Control request instructions, marked as instruction behavior. Where x is the unique attribute code number of the control request instruction;

[0071] For example, when the control link transmits commands, the status parameters of each link are recorded, including command characteristics (such as light adjustment commands, response object: living room smart screen) and node resource cache information (memory usage of each node: living room node 35%, master bedroom node 28%, secondary bedroom node 42%, kitchen node 22%, bathroom node 18%). Based on the parameters, the response object and associated edge computing node are locked, and the command behavior of the 5 control links is marked to ensure that there are no mis-marking or omissions.

[0072] Step S3: Generate a dual-modal control feedback information set based on the marked command behavior, and determine the link-node pointing feature corresponding to the command behavior, i.e. the first pointing feature, and the node-device pointing feature, i.e. the second pointing feature;

[0073] Specifically, based on command behavior, a dual-modal control feedback information set is generated. The first modal control feedback information set is used to collect data from each edge computing node connected to the same local command protocol control link to determine command behavior. First pointing feature The second modal control feedback information set is used to collect smart home screen data on command behavior in order to determine the command behavior. Second pointing feature ;

[0074] The link will be controlled by the local command protocol. The generated first modal control feedback information set is denoted as A second modality control feedback information set generated by the edge computing node is denoted as wherein I represents the encoding serial number of the edge computing node, and E represents the encoding serial number of the smart home screen.

[0075] For example, based on the instruction behavior of the label, a first modality control feedback information set (recording the edge computing node corresponding to each link, such as link 1 corresponding to the living room and the master bedroom node) and a second modality control feedback information set (recording the smart home screen served by each node, such as the living room node corresponding to the living room and the dining room smart screen) are generated, which supports verification through device logs.

[0076] Step S4: calculating the corresponding pointing feature value based on the dual modality control feedback information set; evaluating the priority of the instruction behavior through the pointing feature value, generating and outputting the instruction behavior control sequence in the order of priority;

[0077] Specifically, based on the first modality control feedback information set, the first pointing feature value of the edge computing node when connecting the local instruction protocol control link is evaluated, which reflects the calling degree of the edge computing node when connecting the local instruction protocol control link , wherein is a counting representation function, if the edge computing node exists in the first modality control feedback information set , then , if the edge computing node does not exist in the first modality control feedback information set , then , and R represents the total number of local instruction protocol control links.

[0078] Based on the second modality control feedback information set, the second pointing feature value of the instruction behavior is evaluated, which reflects the resource cache degree of the edge computing node serving each smart home screen when the smart home screen does not respond to the control request instruction within the preset time. , represents the response waiting time of the smart home screen when the edge computing node serves the smart home screen .

[0079] Based on the first pointing feature value and the second pointing feature value, the priority of the instruction behavior is evaluated, and the control sequence of the instruction behavior is generated in the order from large to small according to the priority, and is outputted.

[0080] ​For example, the first modal characteristic value (call degree) is calculated: the call degree of the living room node and link 1 is 0.8 (the node participates in the instruction transmission of 4 links in 5 links), and the call degree of the master bedroom node and link 2 is 0.6; the second modal characteristic value (resource cache degree) is calculated: the resource cache degree of the living room node to the living room smart screen is 0.75 (the response waiting time is accumulated for 3 seconds, and the total waiting time is 4 seconds), and the resource cache degree of the secondary bedroom node to the secondary bedroom smart screen is 0.4; the instruction priority is calculated by multiplication: the instruction priority of the living room node-link 1-living room smart screen is 0.8*0.75=0.6 (high priority), and the instruction priority of the master bedroom node-link 2-master bedroom smart screen is 0.6*0.5=0.3 (low priority); the control sequence is generated by code debugging according to the priority, the average response time of the high-priority instruction execution is 0.8 seconds, which is lower than the preset timeout, and the average response time of the low-priority instruction execution is 1.5 seconds, which meets the control requirements.

[0081] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0082] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, although the foregoing embodiments of the present application are described in detail, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A localized control method for smart home screens based on edge computing, characterized in that, The method comprises the following steps: Step S1: Build a local instruction protocol control link between the edge computing node and the smart home screen in the home scene, form a smart home edge control network, and extract corresponding feature information for the case where the control request instruction is not responded; Step S2: Record the link state parameters when the local instruction protocol control link transmits the control request instruction, lock the response object and the associated edge computing node based on the link state parameters, and mark the instruction behavior of each control link; Step S3: Generate a dual-modal control feedback information set based on the marked instruction behavior, and determine the link-node pointing feature corresponding to the instruction behavior, i.e., the first pointing feature, and the node-device pointing feature, i.e., the second pointing feature; Step S4: Calculate the corresponding pointing feature values based on the dual-modal control feedback information set; evaluate the instruction behavior priority through the pointing feature values, and generate and output the instruction behavior control sequence in order of priority; The specific implementation process of step S2 comprises: When the local instruction protocol control link carries the control request instruction, record the link state parameters of the edge computing node calling the local instruction protocol control link to transmit the control request instruction, the link state parameters including the feature information of the control request instruction and the resource cache information of the edge computing node; Based on the link state parameters, the response object and the connected edge computing node in the local instruction protocol control link are locked through the control request instruction, and the edge computing node is taken as the instruction task center. The instruction behavior of each local instruction protocol control link is marked, and the i-th edge computing node is recorded as The r-th local instruction protocol control link is recorded as The e-th smart home screen is recorded as The edge computing node The connected local instruction protocol control link The control request instruction of the smart home screen is triggered when the time is triggered, and the instruction behavior is marked , wherein x is the unique attribute code number of the control request instruction. The specific implementation process of step S3 comprises: Based on the instruction behavior, a dual-modal control feedback information set is generated, wherein a first modal control feedback information set is used to collect edge computing nodes connected to the same local instruction protocol control link to determine a first pointing feature of the instruction behavior A second modal control feedback information set is used to collect smart home screens in the instruction behavior to determine a second pointing feature of the instruction behavior ;​​ The link controlled by the local instruction protocol is The first modal control feedback information set generated is denoted as The second modal control feedback information set generated by the edge computing node is denoted as Wherein, I represents the encoding serial number of the edge computing node, and E represents the encoding serial number of the smart home screen. The specific implementation process of step S4 comprises: based on the first modal control feedback information set, evaluate the edge computing node a connected local instruction protocol control link a first pointing feature value at a time, to reflect the connected local instruction protocol control link the edge computing node the calling degree , wherein, is a counting representation function, if the edge computing node exists in the first modal control feedback information set , then , if the edge computing node does not exist in the first modal control feedback information set , then , and R represents the total number of local instruction protocol control links; based on the second modality control feedback information set, evaluate the instruction behavior second pointing feature value of the smart home screen in the preset time, the edge computing node serves the resource cache degree of each smart home screen , indicates the response waiting time of the smart home screen when the edge computing node serves the smart home screen ​ Based on the first and second pointing feature values, the instruction behavior is evaluated in terms of priority , a control sequence of the instruction behavior is generated in descending order of priority, and output. 2.The method according to claim 1, wherein, The specific implementation process of step S1 comprises: Build a local instruction protocol control link between the edge computing nodes with control response capability in the same home scene, and connect each smart home screen through the local instruction protocol control link to form a smart home edge control network; The smart home edge control network generates a control request instruction for each smart home screen, and if each smart home screen does not respond to the control request instruction within a preset time, the control request instruction is extracted for feature information, including the response object and the response waiting time of the response object, and the response object refers to each smart home screen controlled by the control request instruction.

3. An edge computing based smart home screen local control system, which executes an edge computing based smart home screen local control method according to any one of claims 1-2, characterized in that, The system comprises: a control link building and instruction processing module, a link state recording and marking module, a dual-modal feedback set generation module, an instruction priority evaluation and sequence output module, and a main control unit for coordinating data interaction and process control of each module to realize the whole process of local control from control link building to instruction sequence output; The control link building and instruction processing module is used to build a local instruction protocol control link between the edge computing node and the smart home screen and form a control network, and to process the control request instruction and the feature information extraction when not responding; The link state recording and marking module is used to record the state parameters when the control link transmits the instruction, lock the response object and the associated node, and complete the instruction behavior marking; The dual-modal feedback set generation module is used to generate a dual-modal control feedback information set based on the instruction behavior, and determine two types of pointing features corresponding to the instruction behavior; The instruction priority evaluation and sequence output module is used to calculate the values of the two types of pointing features, evaluate the instruction behavior priority, and generate and output the control sequence in order. 4.The edge computing based smart home screen local control system according to claim 3, characterized in that, The control link building and instruction processing module comprises a control network building unit and an instruction feature extraction unit; The control network building unit is configured to build a local instruction protocol control link among edge computing nodes with control response capability in a home scene, realize the connectivity of each node and a smart home screen, and form a control network; The instruction feature extraction unit is configured to monitor the response state of a control request instruction in the control network, and extract a response object and a response waiting time for an unresponsive instruction. 5.The edge computing based smart home screen local control system according to claim 3, wherein, The link state recording and marking module comprises a state parameter recording unit and an instruction behavior marking unit; The state parameter recording unit is configured to record state parameters of an edge computing node calling link when the control link carries a control request instruction, including instruction feature information and node resource cache information; The instruction behavior marking unit is configured to lock a response object and an associated edge computing node based on the state parameters, and mark the instruction behavior of each control link with the node as the instruction task center. 6.The edge computing based smart home screen local control system according to claim 3, characterized in that, The dual-modal feedback set generation module comprises a first modal set generation unit and a second modal set generation unit; The first modal set generation unit is configured to collect information of each edge computing node connected to the same control link, generate a first modal control feedback information set, and determine a link-node pointing feature of the instruction behavior; The second modal set generation unit is configured to collect smart home screen information involved in the instruction behavior, generate a second modal control feedback information set, and determine a node-device pointing feature of the instruction behavior. 7.The edge computing based smart home screen local control system according to claim 3, characterized in that, The instruction priority evaluation and sequence output module comprises a feature value calculation unit and a control sequence generation unit; The feature value calculation unit is configured to calculate a calling degree feature value of the edge computing node and the control link based on the first modal feedback set, and calculate a resource cache degree feature value of the node to the smart home screen based on the second modal feedback set; The control sequence generation unit is configured to calculate the priority of the instruction behavior by multiplying the two types of feature values, generate an instruction behavior control sequence in descending order of priority, and output the instruction behavior control sequence.

Images