Intelligent management method and device of terminal equipment based on edge computing equipment
By receiving the requests for adding artificial intelligence scenario application of terminal devices on the server side and updating resource pools using intelligent scheduling and artificial bee swarm algorithms, the problem of poor intelligent management reliability of video surveillance equipment is solved, and flexible allocation and failure of edge computing devices is realized Transfer has improved the intelligent management reliability and user experience of terminal devices.
Patent Information
- Application Number
- CN202411126063.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-16
AI Technical Summary
The existing video surveillance equipment has poor intelligent management reliability, and cannot provide intelligent analysis and calculations when edge computing equipment fails, which affects the user experience and lacks effective solutions.
By receiving the request for adding artificial intelligence scenario application of terminal devices on the server side, selecting available edge computing devices based on intelligent scheduling processing, and updating resource pools using artificial bee swarm algorithms, it realizes flexible allocation and failover of edge computing devices, ensuring terminal devices Intelligent management reliability.
It improves the reliability of intelligent management of terminal equipment, realizes flexible allocation and failover of edge computing devices, avoids the problem of unavailability of intelligent analysis scenarios caused by single-device failure, and improves user experience.
Smart Images

Figure CN118660046B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to an intelligent management method and apparatus for terminal devices based on edge computing devices. Background Art
[0002] With the continuous optimization of algorithms, artificial intelligence has quietly integrated into all aspects of our lives. From smart homes to smart cities, to industrial manufacturing, more and more application scenarios hope to combine with artificial intelligence to bring people a more intelligent experience. However, the number of hardware devices that have been deployed and do not have artificial intelligence capabilities is huge. If all of them are replaced, the cost will be high and it will be difficult. For example, with the rapid development of the video surveillance field, more and more monitoring equipment is widely used in various scenarios, such as public safety, traffic management, and enterprise monitoring. People's requirements for the functions of monitoring equipment are also getting higher and higher, hoping to achieve real-time video analysis and intelligent monitoring. In order to achieve a comprehensive upgrade of intelligent monitoring without intrusion, edge computing devices are introduced into related technologies.
[0003] Currently, edge computing devices are mainly deployed on the edge side of video surveillance devices, and intelligent analysis is performed on the monitoring devices in a fixed area based on the onvifi (OpenNetwork Video Interface Forum) protocol, and the analysis results are displayed on the large monitoring screen on the edge computing device side. Although edge computing devices can assist in the intelligent upgrade of existing old equipment, the reliability of intelligent management of monitoring equipment is poor. Edge computing devices and monitoring equipment are limited to fixed connection relationships. If an edge computing device fails, it will no longer be able to provide intelligent analysis and calculation, resulting in the unavailability of the corresponding monitoring equipment intelligent analysis scenario, affecting the user experience.
[0004] With regard to the problem of poor reliability of intelligent management in related technologies, no effective solution has been proposed so far. Summary of the invention
[0005] In this embodiment, a method and apparatus for intelligent management of terminal devices based on edge computing devices are provided to solve the problem of poor reliability of intelligent management in related technologies.
[0006] In a first aspect, in this embodiment, a terminal device intelligent management method based on an edge computing device is provided, which is applicable to an intelligent management system; the intelligent management system includes a terminal device, an edge computing device, and a server, and a target network where the terminal device is located has several edge computing devices deployed; the edge computing device is connected to the server; the method is applied to the server side; the method includes:
[0007] Receive a request for adding an artificial intelligence scenario application initiated by a target terminal device;
[0008] Based on the request to add an artificial intelligence scenario application, intelligently schedule the edge computing device in the target network to obtain a target edge computing device;
[0009] Control the target edge computing device to establish a connection with the target terminal device.
[0010] In some of the embodiments, before performing intelligent scheduling processing on the edge computing device in the target network based on the request to add the artificial intelligence scenario application to obtain the target edge computing device, the method includes:
[0011] Based on the request to add the artificial intelligence scenario application, determine whether there is an available edge computing device in the target network;
[0012] If the available edge computing device exists in the target network, preferentially selecting a target edge computing device from the available edge computing devices;
[0013] If the available edge computing device does not exist in the target network, an alarm notification is triggered.
[0014] In some embodiments, if the available edge computing device exists in the target network, preferentially selecting a target edge computing device from the available edge computing devices includes:
[0015] If the available edge computing device exists in the target network, determining a resource pool corresponding to the target terminal device based on the target terminal device information in the request to add the artificial intelligence scenario application;
[0016] If the resource pool is not empty, obtaining a target edge computing device from the resource pool;
[0017] If the resource pool is empty, the available edge computing devices are traversed, the resource pool is updated based on the artificial bee colony algorithm, and the target edge computing device is obtained from the updated resource pool.
[0018] In some embodiments, if the resource pool is empty, traversing the edge computing devices in the target network and updating the resource pool based on an artificial bee colony algorithm includes:
[0019] If the resource pool is empty, initialize the parameters of the artificial bee colony algorithm, and randomly select a preset number of edge computing devices from the target network as initial honey sources;
[0020] Based on the leader bee searching for new nectar sources near the initial nectar source;
[0021] Based on the follower bee continuing to search near the new nectar source with higher quality, a field search is performed until a local optimal nectar source is obtained;
[0022] The scout bees perform a global search on the local optimal nectar source until a maximum number of searches is reached, and then the optimal nectar source currently obtained by the scout bees is added to the resource pool.
[0023] In some embodiments, the method further comprises:
[0024] After the edge computing device is deployed in the target network, judging whether each edge computing device remains online based on device status perception detection;
[0025] When the target edge computing device is detected as being offline, the target edge computing device is redetermined from the resource pool, and the redetermined target edge computing device is controlled to establish a connection with the target terminal device.
[0026] In some of the embodiments, after the edge computing device is deployed in the target network, judging whether each edge computing device remains online based on device status perception detection includes:
[0027] After the edge computing device is deployed in the target network, heartbeat information regularly reported by the edge computing device is received, and an online status of the edge computing device is stored based on the received heartbeat information and an identity of the edge computing device;
[0028] After the target edge computing device establishes a connection with the target terminal device, judging whether the online state exists based on the identity of the target edge computing device;
[0029] If the online state does not exist, the target edge computing device is considered to be offline, and if the online state exists, the target edge computing device is considered to be online.
[0030] In some of the embodiments, after controlling the target edge computing device to establish a connection with the target terminal device, the process further includes:
[0031] Control the target edge computing device to obtain the data to be analyzed of the target terminal device, and perform intelligent calculation on the data to be analyzed based on a preset algorithm model;
[0032] Receive the intelligent computing result fed back by the target edge computing device.
[0033] In a second aspect, in this embodiment, a terminal device intelligent management device based on an edge computing device is provided, and the device is applicable to an intelligent management system; the intelligent management system includes a terminal device, an edge computing device, and a server, and a target network where the terminal device is located has several edge computing devices deployed; the edge computing device is connected to the server; the device is applied to the server side; the device includes:
[0034] A request receiving module, used to receive a request for adding an artificial intelligence scenario application initiated by a target terminal device;
[0035] A scheduling module, configured to perform intelligent scheduling processing on the edge computing device in the target network based on the request to add the artificial intelligence scenario application, so as to obtain a target edge computing device;
[0036] A connection module is used to control the target edge computing device to establish a connection with the target terminal device.
[0037] In a third aspect, an intelligent management system is provided in this embodiment, the system comprising: a terminal device, an edge computing device, and a server;
[0038] The terminal device and the edge computing device are deployed in the same target network in a target area, and the target network includes a plurality of terminal devices and a plurality of edge computing devices;
[0039] The server is connected to the edge computing device and is used to execute the steps of the intelligent management method of terminal devices based on edge computing devices as described in any one of the first aspects.
[0040] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the terminal device intelligent management method based on the edge computing device described in the first aspect is implemented.
[0041] Compared with the related art, the intelligent management method and apparatus for terminal devices based on edge computing devices provided in this embodiment are suitable for intelligent management systems; the intelligent management system includes terminal devices, edge computing devices and servers, and several edge computing devices are deployed in the target network where the terminal devices are located; the edge computing device is connected to the server; the method is applied on the server side; the method includes: receiving a request for adding an artificial intelligence scenario application initiated for a target terminal device; based on the request for adding an artificial intelligence scenario application, performing intelligent scheduling processing on the edge computing device in the target network to obtain a target edge computing device; controlling the target edge computing device to establish a connection with the target terminal device, thereby solving the problem of poor reliability of intelligent management, realizing flexible allocation of edge computing devices in a fixed area, and improving the reliability of intelligent management of terminal devices.
[0042] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0044] Figure 1 A hardware structure block diagram of a terminal of an intelligent management method for terminal devices based on an edge computing device in one embodiment;
[0045] Figure 2 is a schematic diagram of an intelligent management system in one embodiment;
[0046] Figure 3 A schematic diagram of a process of an intelligent management method of a terminal device based on an edge computing device in one embodiment;
[0047] Figure 4 A topological diagram of intelligent management system equipment in a preferred embodiment;
[0048] Figure 5 A schematic diagram of a process of an intelligent management method of terminal devices based on edge computing devices in a preferred embodiment;
[0049] Figure 6 A schematic diagram of the AI box resource estimation process in a preferred embodiment;
[0050] Figure 7 A schematic diagram of a process flow of real-time perception of device status in a preferred embodiment;
[0051] Figure 8A schematic diagram of a device failure transfer process in a preferred embodiment;
[0052] Fig. 9 A schematic diagram of a device state storage model in a preferred embodiment;
[0053] Fig.10 This is a structural block diagram of a terminal device intelligent management device based on an edge computing device in an embodiment. DETAILED DESCRIPTION
[0054] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.
[0055] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with general skills in the technical field to which this application belongs. The words "one", "a", "the", "these" and the like in this application do not indicate a quantitative limitation, and they may be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone. Generally, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.
[0056] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. For example, running on a terminal, Figure 1 : is a hardware structure block diagram of a terminal of the terminal device intelligent management method based on edge computing device of this embodiment. Figure 1 As shown, the terminal may include one or more ( Figure 1Only one is shown in the figure) a processor 102 and a memory 104 for storing data, wherein the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.
[0057] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the intelligent management method of terminal equipment based on edge computing equipment in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0058] The transmission device 106 is used to receive or send data via a network. The above network includes a wireless network provided by the communication provider of the terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.
[0059] In this embodiment, a terminal device intelligent management method based on edge computing device is provided, which is applicable to an intelligent management system; an intelligent management system such as Figure 2 As shown, it includes terminal devices, edge computing devices and servers. Several edge computing devices are deployed in the target network where the terminal devices are located; the edge computing devices are connected to the servers. Figure 3 is a flow chart of the terminal device intelligent management method based on edge computing device of this embodiment, which is applied to the server side, such as Figure 3 As shown, the process includes the following steps:
[0060] Step S210, receiving a request for adding an artificial intelligence scenario application initiated by a target terminal device.
[0061] Specifically, multiple terminal devices are deployed in the target network, including but not limited to IoT devices such as sensor devices and execution devices, preferably terminal-side monitoring devices or terminal network cameras. When the client uses one of the terminal devices as the target terminal device, the client initiates an add artificial intelligence scenario application request for the target terminal device to the server. The add artificial intelligence scenario application request will carry the device information (such as IP, Mac address and device code) of the corresponding target terminal device, user identity identification (such as mobile phone number and user ID), etc.
[0062] Step S220, based on the request to add an artificial intelligence scenario application, intelligent scheduling is performed on the edge computing device in the target network to obtain the target edge computing device.
[0063] Specifically, edge computing devices are computing nodes deployed at the edge of the network. They usually have the ability to process data, store information, and perform specific tasks. Edge computing devices can be dedicated hardware boxes, gateway devices, small servers in server clusters, or terminal devices. When in use, edge computing devices can be dynamically allocated based on the computing power and available channels of edge computing devices to achieve efficient use of resources.
[0064] Step S230, control the target edge computing device to establish a connection with the target terminal device.
[0065] Specifically, when the optimal computing resources, i.e., the target edge computing device, are sensed to exist in the target network, the target edge computing device is connected to the target terminal device to realize data transmission and data calculation on the edge side, completing the intelligent upgrade of the terminal device. A target terminal device can be connected to one or more target edge computing devices, and multiple target edge computing devices can perform different artificial intelligence computing processes respectively. A connection is established, and the binding relationship between the target edge computing device and the target terminal device is persistently stored.
[0066] In this embodiment, a request for adding an artificial intelligence scenario application is received for a target terminal device; based on the request for adding an artificial intelligence scenario application, the edge computing device in the target network is intelligently scheduled and processed to obtain the target edge computing device; the target edge computing device is controlled to establish a connection with the target terminal device, thereby solving the problem of edge computing unavailability caused by a single device failure in a fixed management mode, realizing flexible allocation of edge computing devices in a fixed area, and improving the reliability of intelligent management of terminal devices.
[0067] In some embodiments, the above step S220, based on the request to add an artificial intelligence scenario application, performs intelligent scheduling processing on the edge computing device in the target network to obtain the target edge computing device, includes:
[0068] Step S221, based on the request to add an artificial intelligence scenario application, determine whether there is an available edge computing device in the target network.
[0069] Specifically, the available edge computing devices may be edge computing devices that remain online. They may also be edge computing devices that match the task information in the request to add an artificial intelligence scenario application and remain online. After the edge computing devices are deployed in the target network, device status perception detection is performed based on the idea of a time wheel to determine whether each edge computing device remains online.
[0070] Step S222: If there are available edge computing devices in the target network, a target edge computing device is selected from the available edge computing devices.
[0071] Specifically, based on methods such as artificial bee colony algorithm, genetic algorithm or particle swarm optimization algorithm, an edge computing device with optimal resources can be directly screened out from the available edge computing devices, or a preset number of edge computing devices with higher available resources can be selected to form a resource pool, and any device can be selected from the resource pool as the target edge computing device, among which the edge computing devices with higher available resources have available resources that meet preset conditions and can handle additional tasks.
[0072] Step S223: If there is no available edge computing device in the target network, an alarm notification is triggered.
[0073] Specifically, the alarm notification can be sent to the operation and maintenance personnel to inform them to perform offline deployment.
[0074] In this embodiment, available edge computing devices are searched for, the efficiency of device allocation is improved through the resource pool, and the device deployment is improved based on alarm notifications.
[0075] In some embodiments, based on the above step S222, if there is an available edge computing device in the target network, the target edge computing device is preferentially selected from the available edge computing devices, specifically including the following steps:
[0076] Step S222-a: If the available edge computing device exists in the target network, determine the resource pool corresponding to the target terminal device based on the target terminal device information in the request to add the artificial intelligence scenario application.
[0077] Specifically, the terminal device information includes information such as device IP, Mac address and device code.
[0078] Step S222-b, if the resource pool is not empty, obtain the target edge computing device from the resource pool.
[0079] Specifically, the edge computing devices stored in the resource pool are devices with relatively low resource occupancy and have sufficient idle resources to be allocated. At this time, any edge computing device can be obtained from the resource pool as the target edge computing device based on the terminal device information. When the target edge computing device successfully receives the task, the current target edge computing device is removed from the resource pool to balance the use of the resources of each edge computing device. After removing the target edge computing device, it can also be marked as a new honey source for use as a comparison factor when updating the resource pool.
[0080] Step S222-c, if the resource pool is empty, traverse the edge computing devices in the target network, update the resource pool based on the artificial bee colony algorithm, and obtain the target edge computing device from the updated resource pool.
[0081] Specifically, based on the artificial bee colony algorithm, resource usage (including CPU usage, memory usage, network bandwidth and other information) is used as the selection basis to screen out the optimal resources that meet the available resource standards, and update the resource pool based on the optimal resources.
[0082] In this embodiment, the idea of combining artificial bee colony algorithm with resource pool is adopted to realize resource estimation of edge computing devices, which solves the problems of uneven device resource allocation, resource waste and failure to maximize edge computing efficiency caused by subjective evaluation of device resources or random selection of edge computing devices.
[0083] In some embodiments, based on the above step S223, if the resource pool is empty, traverse the edge computing devices in the target network and update the resource pool based on the artificial bee colony algorithm, specifically including the following steps:
[0084] Step S310: If the resource pool is empty, initialize the parameters of the artificial bee colony algorithm and randomly select a preset number of edge computing devices from the target network as initial honey sources.
[0085] Specifically, the parameters of the worker bee colony algorithm are initialized, including: initialization of the colony size NP, the nectar source counter count, the nectar source threshold limit, and the maximum number of searches Tmax. And through the random generation formula, n optimal resources (i.e., n edge computing devices) are randomly selected. The random generation formula is as follows:
[0086] ;
[0087] in, x i represents the i-th nectar source,x i (i=1,2,...,n) is a D-dimensional vector, D is the number of optimization parameters, j∈{1,2,...,D}, x ij Indicates honey source x i No. j The value of the dimension, x j min represents the lower bound of the search space, x j max Represents the upper limit of the search space.
[0088] Step S320, searching for new nectar sources near the initial nectar source based on the leader bee.
[0089] Step S330, based on the follower bee continuing to search near the new nectar source with higher quality, a field search is performed until the local optimal nectar source is obtained.
[0090] Step S340, a global search is performed on the local optimal nectar source based on the scout bees until the maximum number of searches is reached, and then the optimal nectar source currently obtained by the scout bees is added to the resource pool.
[0091] Specifically, first, in the leading bee stage, all nectar sources x i Assign a leader bee, which generates new nectar sources based on the search formula v i , record the optimal solution and update the nectar source counter count. The search formula is:
[0092] ;
[0093] in, v id It indicates a new nectar source generated by the leader bee’s search near nectar source i. x id represents the position of the leading bee in the dth dimension corresponding to nectar source i, x jd represents the position of the leading bee in the dth dimension corresponding to the nectar source j, φ is a random number between [-1,1], which determines the disturbance amplitude, Indicates that random perturbations are performed using random numbers generated in [-1,1], and whether to update is chosen according to the greedy strategy x i .
[0094] The follower bee then calculates the quality of the nectar source, that is, the fitness value, and calculates the probability of the nectar source being selected based on the fitness value. p i The formula is as follows:
[0095] ;
[0096] ;
[0097] Where NP represents the size of the bee colony, fit i represents the fitness value, f i It represents the fitness value calculation factor of the i-th nectar source, that is, the quality index of the nectar source. abs(f i ) It indicates the absolute value of the fitness calculation factor for the i-th nectar source.
[0098] The follower bee searches for the nectar source in the same way as the leader bee, records the optimal solution and updates the nectar source counter count. When the nectar source counter count reaches the nectar source threshold limit, the corresponding leader bee of the nectar source is changed to a scout bee. A new nectar source is generated in the scout bee stage to jump out of the local optimal solution. After the maximum number of searches Tmax is reached, the optimal solution output is the optimal resource.
[0099] In this embodiment, the honey source in the target network is initialized, and greedy selection is made after three stages: leading bee, following bee, and scout bee. After completing the maximum number of searches, the optimal solution is output as the optimal resource. By adding the optimal resource to the resource pool, resource information of the edge computing device can be quickly obtained.
[0100] In some embodiments, the method further comprises:
[0101] Step S410: After the edge computing device is deployed in the target network, based on device status perception detection, it is determined whether each edge computing device remains online.
[0102] Step S420: When the target edge computing device is detected as offline, the target edge computing device is redetermined from the resource pool, and the redetermined target edge computing device is controlled to establish a connection with the target terminal device.
[0103] Specifically, when the target edge computing device is detected as offline, the binding relationship of the currently offline target edge computing device is also obtained, and the binding relationship includes the terminal device bound to the current target edge computing device, the algorithm model for performing data analysis on the terminal device, etc. After the re-determined target edge computing device establishes a connection with the target terminal device, the obtained algorithm model is sent to the re-determined target edge computing device, and the target edge computing device is controlled to assist the target terminal device in performing intelligent computing based on the algorithm model.
[0104] In this embodiment, through real-time perception of the device status, the target edge computing device is timely re-determined and the binding relationship is transferred, thereby further improving the reliability of intelligent management of the device.
[0105] In some embodiments, based on the above step S410, after the edge computing device is deployed in the target network, based on the device status perception detection, it is determined whether each edge computing device remains online, specifically including the following steps:
[0106] Step S411, after the edge computing device is deployed in the target network, heartbeat information regularly reported by the edge computing device is received, and the online status of the edge computing device is stored based on the received heartbeat information and the identity of the edge computing device.
[0107] Specifically, after the installation and maintenance personnel complete the deployment of the edge computing device, the edge computing device periodically reports its heartbeat to the server. The server periodically (60s by default) senses the device heartbeat. After receiving the heartbeat information, it obtains the scale that the current detection task should be in based on the current time, device heartbeat interval, current slot, step frequency, and time wheel length, and adds the task to the scale to detect whether the device status is online in a fixed period. The server starts the time wheel. When the detection task successfully adds the specified scale, the online status is cached according to the identity ID of the edge computing device.
[0108] Step S412: After the target edge computing device establishes a connection with the target terminal device, it is determined whether an online state exists based on the identity of the target edge computing device.
[0109] Specifically, first, the time wheel worker thread checks the current time slot that should be executed. If the current slot is not empty, it obtains the current slot status task and checks whether the online status cache exists according to the device ID. If the device online status cache does not exist, it means that the device heartbeat has not been obtained due to timeout. At this time, the device status is considered offline. If the current slot is empty, it will rotate to the next time slot for judgment after a preset time.
[0110] Step S413: if there is no online status, the target edge computing device is considered to be offline; if there is an online status, the target edge computing device is considered to be online.
[0111] In this embodiment, through the state perception based on the time wheel, the online state of the equipment is efficiently and accurately monitored, which helps to timely discover faults, transfer binding relationships, and improve the reliability of intelligent management.
[0112] In some embodiments, based on the above step S230, after controlling the target edge computing device to establish a connection with the target terminal device, the following steps are further included:
[0113] Step S270 controls the target edge computing device to obtain the data to be analyzed from the target terminal device, and performs intelligent calculation on the data to be analyzed based on a preset algorithm model.
[0114] Step S280 receives the intelligent computing results fed back by the target edge computing device.
[0115] Specifically, the intelligent computing results on the edge side are gathered, and the analysis results are reported to the server in the dimension of the target terminal device. The server processes the data and applies it to the Internet user side, thus realizing the full link from the device side to the user side.
[0116] In this embodiment, an intelligent management system is also provided. Figure 2 The system includes: a terminal device, an edge computing device and a server.
[0117] The terminal device and the edge computing device are deployed in the same target network in the target area, and the target network includes several terminal devices and several edge computing devices.
[0118] The server is connected to the edge computing device and is used to execute any step of the above-mentioned terminal device intelligent management method based on the edge computing device.
[0119] Specifically, the terminal devices include but are not limited to devices in application fields such as video surveillance and industrial automation. Taking the video surveillance scenario as an example, the terminal devices can be monitoring devices such as IP Camera (IP Camera, IPC) and Network Video Recorder (Network Video Recorder, NVR); the edge computing device uses AI Box (AI BOX).
[0120] See also Figure 4 , deploy the AI box and IPC / NVR devices in the same LAN through the switch, perform ONVIF search in the AI box through the ONVIFI (Open Network Video Interface Forum) protocol, and add the video stream of the IPC / NVR device scanned by the AI box to the AI box side. After the addition is completed, the AI box can perform stream analysis on the added IPC / NVR on the edge side. It should be noted that before performing stream analysis, the algorithm binding and parameter configuration of the video stream of the corresponding device are required, such as confidence, trigger time, warning cycle, stream acquisition method, frame acquisition interval, detection range, etc. After the AI box is deployed, the AI box will calculate the IPC / NVR on the edge side. The specific calculation rules use the algorithm model configured for the video stream, and data aggregation is performed on the edge side.
[0121] After the intelligent upgrade of monitoring equipment in a fixed area is completed through the AI box, intelligent management and control are still required due to differences in AI box models, computing power, hardware and other factors. That is, the box resources are dynamically allocated according to the computing power and available channels of the AI box to achieve the purposes of reasonable resource allocation and fault transfer.
[0122] like Figure 5 As shown in the figure, based on the request to add artificial intelligence scenario application sent by the client, the server determines whether there is an available AI box in the target network; if there is an available AI box in the target network, the AI box is intelligently scheduled; if there is no available edge computing device in the target network, an alarm notification is triggered. When the monitoring device successfully adds the artificial intelligence scenario application, the AI box status and the binding relationship between the AI box and the monitoring device are persistently stored. Among them, intelligent scheduling includes resource estimation of AI box resources and device failure transfer by the server.
[0123] AI box resource estimation process is as follows Figure 6 As shown in the figure, when the client initiates a video surveillance device algorithm scenario application request to the server, it will bring the IPC device information (such as IP, Mac address and device code) and user identity identification (such as mobile phone number and user ID) to which the algorithm scenario needs to be applied. The server will obtain the optimal resource AI box from the resource pool in the area where the device is located based on the IPC device information. If the resource pool does not have the optimal resource, the AI box in the fixed area will be initialized as the honey source, and after the three stages of leading bee, following bee and scout bee, a greedy choice will be made. After completing the maximum number of searches, the optimal solution is output as the optimal resource. By putting the optimal resource AI box into the resource pool, the box resource information can be quickly obtained through the resource estimation process.
[0124] The process of real-time perception of equipment status is as follows Figure 7 As shown in the figure, based on the idea of the time wheel, after the installation and maintenance personnel complete the deployment of the AI box, the device will periodically report heartbeats to the server. When the server senses the heartbeat of the device, it obtains the scale that the current task should be in based on the current time, device heartbeat interval, current slot, step frequency, and time wheel length, and adds the task to the specified scale to detect whether the device status is online in a fixed period. When it is determined that the device does not report heartbeats periodically, the device fault transfer process is entered at this time, see Figure 8 , check whether the device status exists according to the AI box device ID cache. If not, obtain the IPC and algorithm model associated with the current offline AI box, and obtain the optimal resource AI box from the resource pool, and allocate the IPC and algorithm model associated with the offline AI box to the new online device in the same network environment.
[0125] The device state storage model is as follows Fig. 9As shown in the figure, the AI box device ID cache is placed in Redis, and the device status of all business logic is obtained from Redis. The data that needs to be persistently stored is placed in the Redis queue. Finally, the data is obtained from the Redis queue through a scheduled task and asynchronously written to Mysql for persistent storage.
[0126] Through real-time perception of AI box resources and status, resource estimation and instant fault transfer of AI boxes can be realized. The server can schedule and allocate edge computing devices in all fixed areas, complete the intelligent upgrade of massive IPCs without changing the original IPCs, and establish data aggregation points on the edge side of IPC devices. The server receives all AI analysis data aggregated from the edge side, processes the data for business purposes, and applies it to the user side.
[0127] In this embodiment, the server side can schedule and allocate edge computing devices in all fixed areas, complete the intelligent upgrade of massive monitoring equipment without changing the original monitoring equipment, solve the problem of unavailability of AI models applied to monitoring equipment due to single device failure, and improve the availability of the system.
[0128] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0129] In this embodiment, a terminal device intelligent management device based on an edge computing device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. The terms "module", "unit", "subunit", etc. used below can implement a combination of software and / or hardware for predetermined functions. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0130] Fig.10 is a structural block diagram of the terminal device intelligent management device based on edge computing device of this embodiment, such as Fig.10 As shown, the device is applicable to an intelligent management system; the intelligent management system includes a terminal device, an edge computing device and a server, and a plurality of edge computing devices are deployed in the target network where the terminal device is located; the edge computing device is connected to the server; the device is applied to the server side; the device includes:
[0131] The receiving request module 11 is used to receive a request to add an artificial intelligence scenario application initiated by a target terminal device.
[0132] The scheduling module 12 is used to perform intelligent scheduling processing on the edge computing device in the target network based on the request to add the artificial intelligence scenario application to obtain the target edge computing device.
[0133] The connection module 13 is used to control the target edge computing device to establish a connection with the target terminal device.
[0134] In some of the embodiments, the scheduling module 12 is also used to determine whether there is an available edge computing device in the target network based on the request to add an artificial intelligence scenario application; if the available edge computing device exists in the target network, the target edge computing device is selected from the available edge computing devices; if the available edge computing device does not exist in the target network, an alarm notification is triggered.
[0135] In some of the embodiments, the scheduling module 12 is also used to determine the resource pool corresponding to the target terminal device based on the target terminal device information in the request to add the artificial intelligence scenario application if the available edge computing device exists in the target network; if the resource pool is not empty, obtain the target edge computing device from the resource pool; if the resource pool is empty, traverse the available edge computing devices, update the resource pool based on the artificial bee colony algorithm, and obtain the target edge computing device from the updated resource pool.
[0136] In some of the embodiments, the scheduling module 12 is also used to initialize the parameters of the artificial bee colony algorithm if the resource pool is empty, and randomly select a preset number of the edge computing devices from the target network as the initial nectar source; search for a new nectar source near the initial nectar source based on the leading bee; continue to search near the new nectar source with higher quality based on the follower bee to perform a field search until a local optimal nectar source is obtained; and perform a global search for the local optimal nectar source based on the scout bee until the maximum number of searches is reached, and then add the optimal nectar source currently obtained by the scout bee to the resource pool.
[0137] In some of the embodiments, the scheduling module 12 is also used to determine whether each of the edge computing devices remains online based on device status perception detection after the edge computing device is deployed in the target network; when the target edge computing device is detected as offline, redetermine the target edge computing device from the resource pool, and control the redetermined target edge computing device to establish a connection with the target terminal device.
[0138] In some of the embodiments, the scheduling module 12 is also used to receive heartbeat information reported periodically by the edge computing device after the edge computing device is deployed in the target network, and store the online status of the edge computing device based on the received heartbeat information and the identity of the edge computing device; after the target edge computing device establishes a connection with the target terminal device, determine whether the online status exists based on the identity of the target edge computing device; if the online status does not exist, the target edge computing device is considered to be offline, and if the online status exists, the target edge computing device is considered to be online.
[0139] In some of the embodiments, the connection module 13 is also used to control the target edge computing device to obtain the data to be analyzed from the target terminal device, and perform intelligent calculations on the data to be analyzed based on a preset algorithm model; and receive the intelligent calculation results fed back by the target edge computing device.
[0140] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.
[0141] In this embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
[0142] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and will not be repeated in this embodiment.
[0143] In addition, in combination with the terminal device intelligent management method based on edge computing device provided in the above embodiment, a storage medium can also be provided in this embodiment to implement. The storage medium stores a computer program; when the computer program is executed by the processor, any terminal device intelligent management method based on edge computing device in the above embodiment is implemented.
[0144] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.
[0145] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.
[0146] The term "embodiment" in this application refers to a specific feature, structure or characteristic described in conjunction with the embodiment that can be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is clearly or implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict.
[0147] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.
Claims
1. A terminal device intelligent management method based on edge computing device, characterized in that: Applicable to intelligent management system; the intelligent management system includes terminal devices, edge computing devices and servers, and several edge computing devices are deployed in the target network where the terminal devices are located; The edge computing device is connected to the server; The method is applied to the server side; the method comprises: Receive a request for adding an artificial intelligence scenario application initiated by a target terminal device; Based on the request to add an artificial intelligence scenario application, intelligently schedule the edge computing device in the target network to obtain a target edge computing device; Controlling the target edge computing device to establish a connection with the target terminal device; Wherein, based on the request to add an artificial intelligence scenario application, intelligently scheduling the edge computing device in the target network to obtain the target edge computing device includes: Based on the request to add an artificial intelligence scenario application, determine whether there is an available edge computing device in the target network; if the available edge computing device exists in the target network, select the target edge computing device from the available edge computing devices; wherein the available edge computing device is the edge computing device that remains online; the edge computing device that remains online is obtained based on device state perception detection under the idea of a time wheel; Wherein, if the available edge computing device exists in the target network, preferentially selecting a target edge computing device from the available edge computing devices includes: If the available edge computing device exists in the target network, determining a resource pool corresponding to the target terminal device based on the target terminal device information in the request to add the artificial intelligence scenario application; If the resource pool is not empty, obtaining a target edge computing device from the resource pool; wherein the resource pool includes the edge computing devices that remain online and have sufficient idle resources to be allocated; if the resource pool is empty, traversing the available edge computing devices, updating the resource pool based on an artificial bee colony algorithm, and obtaining a target edge computing device from the updated resource pool; Wherein, if the resource pool is empty, traversing the edge computing devices in the target network and updating the resource pool based on an artificial bee colony algorithm includes: If the resource pool is empty, initialize the parameters of the artificial bee colony algorithm, and randomly select a preset number of edge computing devices from the target network as initial honey sources; Based on the leader bee searching for new nectar sources near the initial nectar source; Based on the follower bee continuing to search near the new nectar source with higher quality, a field search is performed until a local optimal nectar source is obtained; The scout bees perform a global search on the local optimal nectar source until a maximum number of searches is reached, and then the optimal nectar source currently obtained by the scout bees is added to the resource pool.
2. The intelligent management method of terminal equipment based on edge computing equipment according to claim 1 is characterized in that: The method further comprises: If the available edge computing device does not exist in the target network, an alarm notification is triggered.
3. The intelligent management method of terminal equipment based on edge computing equipment according to claim 1 is characterized in that: The method further comprises: After the edge computing device is deployed in the target network, judging whether each edge computing device remains online based on device status perception detection; When the target edge computing device is detected as being offline, the target edge computing device is redetermined from the resource pool, and the redetermined target edge computing device is controlled to establish a connection with the target terminal device.
4. The intelligent management method of terminal equipment based on edge computing equipment according to claim 3 is characterized in that: After the edge computing device is deployed in the target network, judging whether each edge computing device remains online based on device status perception detection includes: After the edge computing device is deployed in the target network, heartbeat information regularly reported by the edge computing device is received, and an online status of the edge computing device is stored based on the received heartbeat information and an identity of the edge computing device; After the target edge computing device establishes a connection with the target terminal device, judging whether the online state exists based on the identity of the target edge computing device; If the online state does not exist, the target edge computing device is considered to be offline, and if the online state exists, the target edge computing device is considered to be online.
5. The intelligent management method of terminal equipment based on edge computing equipment according to claim 1 is characterized in that: After the control of establishing a connection between the target edge computing device and the target terminal device, the method further includes: Control the target edge computing device to obtain the data to be analyzed of the target terminal device, and perform intelligent calculation on the data to be analyzed based on a preset algorithm model; Receive the intelligent computing result fed back by the target edge computing device.
6. An intelligent terminal device management device based on edge computing equipment, characterized in that: The device is suitable for an intelligent management system; the intelligent management system includes a terminal device, an edge computing device and a server, and a plurality of edge computing devices are deployed in the target network where the terminal device is located; The edge computing device is connected to the server; The device is applied to the server side; the device comprises: A request receiving module, used to receive a request for adding an artificial intelligence scenario application initiated by a target terminal device; A scheduling module, configured to perform intelligent scheduling processing on the edge computing device in the target network based on the request to add an artificial intelligence scenario application, to obtain a target edge computing device; wherein, based on the request to add an artificial intelligence scenario application, performing intelligent scheduling processing on the edge computing device in the target network to obtain a target edge computing device includes: Based on the request to add an artificial intelligence scenario application, determine whether there is an available edge computing device in the target network; if the available edge computing device exists in the target network, select the target edge computing device from the available edge computing devices; wherein the available edge computing device is the edge computing device that remains online; the edge computing device that remains online is obtained based on device state perception detection under the idea of a time wheel; Wherein, if the available edge computing device exists in the target network, preferentially selecting a target edge computing device from the available edge computing devices includes: If the available edge computing device exists in the target network, the resource pool corresponding to the target terminal device is determined based on the target terminal device information in the request to add the artificial intelligence scenario application; if the resource pool is not empty, the target edge computing device is obtained from the resource pool; wherein the resource pool includes the edge computing device that remains online and has sufficient idle resources to be allocated; If the resource pool is empty, traverse the available edge computing devices, update the resource pool based on the artificial bee colony algorithm, and obtain the target edge computing device from the updated resource pool; Wherein, if the resource pool is empty, traversing the edge computing devices in the target network and updating the resource pool based on an artificial bee colony algorithm includes: If the resource pool is empty, initialize the parameters of the artificial bee colony algorithm, and randomly select a preset number of edge computing devices from the target network as initial honey sources; Based on the leader bee searching for new nectar sources near the initial nectar source; Based on the follower bee continuing to search near the new nectar source with higher quality, a field search is performed until a local optimal nectar source is obtained; Performing a global search for the local optimal nectar source based on the scout bees until a maximum number of searches is reached, and then adding the optimal nectar source currently obtained by the scout bees to the resource pool; A connection module is used to control the target edge computing device to establish a connection with the target terminal device.
7. An intelligent management system, characterized in that: The system includes: a terminal device, an edge computing device and a server; The terminal device and the edge computing device are deployed in the same target network in a target area, and the target network includes a plurality of terminal devices and a plurality of edge computing devices; The server is connected to the edge computing device and is used to execute the steps of the intelligent management method of terminal devices based on edge computing devices according to any one of claims 1 to 5.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent management method of terminal devices based on edge computing devices as described in any one of claims 1 to 5 are implemented.
