CN111556469A - Computing system and computing method for artificial intelligence edge - Google Patents

Abstract

The present invention relates to an artificial intelligence edge computing system, which includes at least one Internet of Things device, an edge collection computing device and a mobile edge server; wherein, the Internet of Things device is used to collect environmental information and convert the environmental information into the data to be processed; the edge collection computing device is used to obtain monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, so as to predict whether an abnormality occurs in the designated area; the mobile edge The server is used for data collection and data sorting, wherein the mobile edge server includes a display unit, a storage unit, a wireless network interface unit, a judgment module and an alarm instruction generation unit. The invention also discloses a corresponding method. By implementing the present invention, the data processing efficiency can be improved, the calculation pressure can be reduced, the data delay can be reduced, and the edge computing security can be improved.

Description

An artificial intelligence edge computing system and computing method

technical field

The invention relates to the data processing field of edge computing, in particular to an artificial intelligence edge computing system and computing method.

Background technique

The front-end data collection point (ie edge device) in the Internet of Things initially only has the function of data collection and transmission. Due to the current high integration of sensors, the amount of data acquired by edge devices increases, and the customer's border side needs to calculate results. The speed is also increasing, which has prompted the birth of edge computing theory.

Edge computing refers to the use of an open platform that integrates network, computing, storage, and application core capabilities on the side close to the source of objects or data to provide the most recent services nearby. Its applications are initiated on the edge side to generate faster network service responses, which can meet the basic needs of the industry in real-time business, application intelligence, security and privacy protection. Edge computing is relative to cloud computing, which is essentially a distributed computing.

The theory of edge computing involves a large amount of data processing, and the requirements for data processing are getting higher and higher. At present, the general data processing adopts cloud computing. The specific cloud computing is to pass all the data through the network through each node server. Then upload it to the central server for processing.

Traditional edge computing is a method of physically approaching the location where data is generated and processing data, such as various sensing nodes in the Internet of Things. Some sensing nodes are not only responsible for collecting and transmitting data, but also realize some functions of data processing and result output. However, due to the lack of corresponding edge computing industry standards, specifications and protocols, the provided edge computing solutions lack technical and product commonality, and are difficult to be universally used in industrial communications, that is, to a certain extent, it cannot reduce the pressure of cloud computing and solve data delays. question.

SUMMARY OF THE INVENTION

In order to solve the above-mentioned problems in the prior art, the present invention provides an artificial intelligence edge computing system and computing method, which can improve data processing efficiency, reduce computing pressure and data delay, and improve the security of edge computing sex.

In order to solve the above technical problems, one aspect of the present invention provides an artificial intelligence edge computing system, which includes at least one Internet of Things device, an edge collection computing device and a mobile edge server; wherein,

The IoT device is used to collect environmental information and convert the environmental information into the data to be processed;

The edge collection computing device is used to obtain monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, so as to predict whether an abnormality occurs in the designated area; wherein the edge collection computing device includes AIOS Operating system unit, data acquisition unit, data conversion unit, data cleaning unit and data aggregation calculation unit;

The mobile edge server is used for data collection and data sorting, wherein the mobile edge server includes a display unit, a storage unit, a wireless network interface unit, a judgment module, and an alarm instruction generation unit.

Preferably, the data acquisition unit is used for the acquisition of contrast, temperature, humidity, and PM2.5 environmental data, and the data collected by the data acquisition unit is sent to the data conversion unit after classification processing.

The data conversion unit is used for format conversion of the data received from the data acquisition unit, and sent to the data cleaning unit;

The data cleaning unit is used to clean the data received from the data conversion unit, and send it to the data aggregation computing unit;

The data aggregation calculation unit is used to perform aggregation calculation on the data received from the data cleaning unit to form sorted data; the sorted data is input into the cloud deep learning model in the AIOS operating system, and the result is output.

Preferably, the AIOS operating system unit includes a bus and a device driver for connecting the data acquisition unit, the data conversion unit, the data cleaning unit and the data aggregation computing unit, as well as for providing a deep learning algorithm framework, a neural network model, various algorithms The platform layer of the SDK and services; the AIOS operating system unit can support a single linux operating system and a linux+RTOS heterogeneous operating system.

Preferably, in the edge collection computing device, the AIOS operating system unit further includes an AI core processor, and the AI core processor includes a multi-core ARM processor, a multi-core GPU, a multi-core neural network processor NNIE and a multi-core DSP.

Preferably, in the mobile edge server, the display unit is one of LED, LCD or OLED, and the display unit realizes the display of device operation status or configuration information;

The storage unit includes onboard storage and external storage media, wherein the storage unit includes EMMC, E2PROM, FLASHSD card, SSD, hard disk, and the storage unit stores data written or read by the intelligent blockchain processing unit or routing processing unit. ;

The wireless network interface unit includes serial port, IO port, RS485, USB, audio and video interfaces, the wireless network interface unit realizes the input of external information, and the wireless network interface unit is responsible for inputting and outputting other forms of data except network data.

Preferably, the edge collection computing device further includes a security protection unit, and the security protection unit has a public key authentication mechanism and a private key authentication mechanism to provide a single or batch edge device authentication and verification process.

Preferably, the mobile edge server further includes a judgment module and an alarm instruction generation module, wherein the wireless network interface unit collects data related to the execution end; the judgment module determines whether the data related to the execution end is abnormal; the alarm instruction generation The module is used for generating an alarm instruction when the judgment result is yes; the wireless network interface unit sends the alarm instruction to the display unit, and the display unit executes the alarm instruction.

As another aspect of the present invention, a computing method for artificial intelligence edge is also provided, which is implemented in the aforementioned system and includes the following steps:

Step S10: the mobile edge server and the edge collection computing device use the collected large amount of historical data to perform deep learning model training, and finally generate a cloud deep learning model; deliver the deep learning model to the AIOS operating system unit;

Step S11: The IoT device collects environmental information, and converts the environmental information into data to be processed;

Step S12: The edge collection computing device obtains the monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, and after processing, sends it to the cloud deep learning model in the AIOS operating system, and outputs the results to predict the specified data. Whether an exception occurs in the area;

Step S13, the wireless network interface unit of the mobile edge server collects the data related to the execution end; and judges whether the data related to the execution end and the cloud deep learning model is abnormal through the judgment module; when it is judged that the data is abnormal, the alarm instruction generation module generates alarm command;

Step S14, the wireless network interface unit sends the alarm instruction to a display unit, and the display unit executes the alarm instruction.

Preferably, the step S12 further includes:

The data acquisition unit collects the environmental data of luminance, temperature, humidity and PM2.5, the collected data is analog data or digital data, and the data collection unit sends the collected data to the data conversion unit after classifying and processing;

The data conversion unit performs format conversion on the data and sends it to the data cleaning unit;

The data cleaning unit cleans the data, and then the data aggregation computing unit performs the aggregation calculation to form the sorted data;

Input the sorted data into the cloud deep learning model in the AIOS operating system, and output the result.

Preferably, it further includes: the security protection unit implements the authentication and verification process of providing single or batch edge devices according to the public key authentication mechanism and the private key authentication mechanism it has.

Implementing the embodiment of the present invention has the following beneficial effects:

First, in the embodiment of the present invention, the data conversion unit is connected to the data collector, the data conversion unit is connected to the data cleaning unit, the data cleaning unit is connected to the data aggregation calculation unit, and the data conversion unit, the data cleaning unit and the data aggregation calculation unit cooperate with each other. It is used to organize the data collected by the data acquisition unit to improve the data processing efficiency, thereby reducing the calculation pressure and reducing the data delay.

Secondly, in the embodiment of the present invention, the edge collection computing device further includes a security protection unit, and the security protection unit has a public key authentication mechanism and a private key authentication mechanism, which realizes the authentication and verification process of providing single or batch edge devices, and improves the edge Computing security.

In addition, in the embodiment of the present invention, the wireless network interface unit includes serial ports, IO ports, RS485, USB, audio and video interfaces, the wireless network interface unit realizes the input of external information, and the wireless network interface unit is responsible for input and output except for network data. It has a variety of data access protocols, and can support a variety of device access functions. In addition to the traditional routing function, it can further process and calculate the acquired data.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention.

1 is a schematic structural diagram of an embodiment of an artificial intelligence edge computing system provided by the present invention;

FIG. 2 is a schematic structural diagram of the edge collection computing device in FIG. 1;

FIG. 3 is a schematic structural diagram of the mobile edge server in FIG. 1;

FIG. 4 is a schematic main flow diagram of an embodiment of an artificial intelligence edge computing method provided by the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments.

As shown in FIG. 1, it shows a schematic structural diagram of an embodiment of an artificial intelligence edge computing system provided by the present invention; as shown in FIG. 2 and FIG. 3 together, in the embodiment of the present invention, the artificial intelligence The computing system of the intelligent edge includes at least an IoT device 1 , an edge collection computing device 2 and a mobile edge server 3 .

The IoT device 1 is used to collect environmental information and convert the environmental information into data to be processed. Among them, multiple IoT devices belong to the same designated area.

The edge collection computing device 2 is used to obtain monitoring feature data collected by multiple IoT devices 1 connected thereto, so as to predict whether an abnormality occurs in a designated area. More specifically, the artificial intelligence edge collection computing device 2 includes a data collection unit 20 , a data conversion unit 21 , a data cleaning unit 22 and a data aggregation computing unit 23 .

In the edge collection computing device 2, the data collection unit 20 is used for the collection of contrast, temperature, humidity, PM2.5 environmental data, the data collected by the data collection unit 20 is analog data or digital data, and the data collected by the data collection unit 20 The data are classified and processed by the wireless network interface unit and then sent to the data conversion unit 21 .

The data conversion unit 21 is configured to perform format conversion on the data from the data acquisition unit 20 and send it to the data cleaning unit 22;

The data cleaning unit 22 is used to perform data cleaning on the data from the data cleaning unit 22 (remove unnecessary data);

The data aggregation calculation unit 23 is used to aggregate and calculate the data from the data cleaning unit 22 to form sorted data; the sorted data is input into the cloud deep learning model of the AIOS operating system 25, and the result is output .

Further, the edge collection computing device 2 further includes a security protection unit 24, and the security protection unit 24 has a public key authentication mechanism and a private key authentication mechanism, and provides a single or batch edge device authentication and verification process.

The edge collection computing device 2 also includes an AIOS operating system unit 25, and the AIOS operating system unit 25 includes a bus and equipment for connecting the data collection unit 20, the data conversion unit 21, the data cleaning unit 22 and the data aggregation computing unit 23. Drivers, and a platform layer for providing deep learning algorithm frameworks, neural network models, various algorithm SDKs and services; the AIOS operating system unit 25 can support a single linux operating system and a linux+RTOS heterogeneous operating system.

In the edge collection computing device, the AIOS operating system unit 25 also includes an AI core processor, and the AI core processor includes a multi-core ARM processor, a multi-core GPU, a multi-core neural network processor NNIE and a multi-core DSP.

The mobile edge server 3 is used for data collection and data sorting, wherein the mobile edge server 3 at least includes a display unit 30 , a storage unit 31 , a wireless network interface unit 32 , a judgment module 33 and an alarm instruction generation module 34 .

Wherein, the display unit 30 is one of LED, LCD or OLED, and the display unit 30 realizes the display of the operating state or configuration information of the device.

The storage unit 31 includes onboard storage and external storage media, wherein the storage unit 31 includes EMMC, E2PROM, FLASHSD card, SSD, and hard disk, and the storage unit 31 stores the data written or read by the intelligent blockchain processing unit or routing processing unit. data.

The wireless network interface unit 32 includes serial ports, IO ports, RS485, USB, audio and video interfaces, the wireless network interface unit 32 realizes the input of external information, collects data related to the execution end, and the wireless network interface unit 32 is responsible for input and output in addition to network data. other forms of data.

The judgment module 33 is used for judging whether the output result of the cloud deep learning model and the data related to the execution terminal are both abnormal. If the judgment result is both abnormal, it is determined that the data is abnormal; otherwise, it is determined that the data is not abnormal;

The alarm instruction generation module 34 is configured to generate an alarm instruction when the judgment result of the judgment module 33 is that the data is abnormal. The wireless network interface unit 32 sends the alarm instruction to the display unit 30, and the display unit 30 executes the alarm instruction.

As shown in FIG. 4 , it shows a schematic diagram of the main flow of an embodiment of an artificial intelligence edge computing method provided by the present invention. In this embodiment, it is implemented in the system as shown in FIG. 1 to FIG. 3 . Specifically, the computing method of the artificial intelligence edge includes the following steps:

Step S10: The mobile edge server and the edge collection computing device use the collected large amount of historical data to perform deep learning model training, and finally generate a cloud deep learning model; and deliver the deep learning model to the AIOS operating system unit.

Step S11: The IoT device collects environmental information, and converts the environmental information into data to be processed.

Step S12: The edge collection computing device obtains the monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, and after processing, sends it to the cloud deep learning model in the AIOS operating system, and outputs the results to predict the specified data. Whether an exception has occurred in the area. Specifically, the step S12 includes the following steps:

The data acquisition unit collects the environmental data of luminance, temperature, humidity and PM2.5, the collected data is analog data or digital data, and the data collection unit sends the collected data to the data conversion unit after classifying and processing;

The data conversion unit performs format conversion on the data and sends it to the data cleaning unit;

The data cleaning unit cleans the data, and then the data aggregation computing unit performs the aggregation calculation to form the sorted data;

Input the sorted data into the cloud deep learning model in the AIOS operating system, and output the result;

Step S13, the wireless network interface unit of the mobile edge server collects the data related to the execution end; and judges whether the data related to the execution end and the cloud deep learning model is abnormal through the judgment module; when it is judged that the data is abnormal, the alarm instruction generation module generates alarm command;

Step S14, the wireless network interface unit sends the alarm instruction to a display unit, and the display unit executes the alarm instruction.

Further, the method further includes the following steps: Step S14: The security protection unit implements a single or batch edge device authentication verification process according to the public key authentication mechanism and the private key authentication mechanism it has.

More details can be combined with the descriptions of the foregoing FIG. 1 to FIG. 3 , and will not be described in detail here.

Implementing the embodiment of the present invention has the following beneficial effects:

First, in the embodiment of the present invention, the data conversion unit is connected to the data collector, the data conversion unit is connected to the data cleaning unit, the data cleaning unit is connected to the data aggregation calculation unit, and the data conversion unit, the data cleaning unit and the data aggregation calculation unit cooperate with each other. It is used to organize the data collected by the data acquisition unit to improve the data processing efficiency, thereby reducing the calculation pressure and reducing the data delay.

Secondly, in the embodiment of the present invention, the edge collection computing device further includes a security protection unit, and the security protection unit has a public key authentication mechanism and a private key authentication mechanism, which realizes the authentication and verification process of providing single or batch edge devices, and improves the edge Computing security.

In addition, in the embodiment of the present invention, the wireless network interface unit includes serial ports, IO ports, RS485, USB, audio and video interfaces, the wireless network interface unit realizes the input of external information, and the wireless network interface unit is responsible for input and output except for network data. It has a variety of data access protocols, and can support a variety of device access functions. In addition to the traditional routing function, it can further process and calculate the acquired data.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be Modifications or equivalent replacements are made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention shall be included within the protection scope of the claims of the present invention.

Claims (10)

1. a computing system of artificial intelligence edge, is characterized in that, comprises at least one Internet of Things device, edge collection computing device and mobile edge server; Wherein, The IoT device is used to collect environmental information and convert the environmental information into the data to be processed; The edge collection computing device is used to obtain monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, so as to predict whether an abnormality occurs in the designated area; wherein the edge collection computing device includes AIOS Operating system unit, data acquisition unit, data conversion unit, data cleaning unit and data aggregation calculation unit; The mobile edge server is used for data collection and data sorting, wherein the mobile edge server includes a display unit, a storage unit, a wireless network interface unit, a judgment module, and an alarm instruction generation unit. 2. the computing system of artificial intelligence edge according to claim 1, is characterized in that: The data collection unit is used for collection of contrast, temperature, humidity, and PM2.5 environmental data, and the data collected by the data collection unit is sent to the data conversion unit after being classified and processed. The data conversion unit is used for format conversion of the data received from the data acquisition unit, and sent to the data cleaning unit; The data cleaning unit is used to clean the data received from the data conversion unit, and send it to the data aggregation computing unit; The data aggregation calculation unit is used to perform aggregation calculation on the data received from the data cleaning unit to form sorted data; the sorted data is input into the cloud deep learning model in the AIOS operating system, and the result is output. 3. the computing system of artificial intelligence edge according to claim 2 is characterized in that: described AIOS operating system unit comprises the bus and equipment that is used to connect data acquisition unit, data conversion unit, data cleaning unit and data aggregation computing unit A driver, and a platform layer for providing a deep learning algorithm framework, a neural network model, various algorithm SDKs and services; the AIOS operating system unit can support a single linux operating system and a linux+RTOS heterogeneous operating system. 4. The artificial intelligence edge computing system according to claim 3, characterized in that: in the edge collection computing device, the AIOS operating system unit further comprises an AI core processor, and the AI core processor comprises a multi-core processor ARM processor, multi-core GPU, multi-core neural network processor NNIE and multi-core DSP. 5. The computing system of artificial intelligence edge according to any one of claims 1 to 4, is characterized in that: In the mobile edge server, the display unit is one of LED, LCD or OLED, and the display unit realizes the display of device operation status or configuration information; The storage unit includes onboard storage and external storage media, wherein the storage unit includes EMMC, E2PROM, FLASHSD card, SSD, hard disk, and the storage unit stores data written or read by the intelligent blockchain processing unit or routing processing unit. ; The wireless network interface unit includes serial port, IO port, RS485, USB, audio and video interfaces, the wireless network interface unit realizes the input of external information, and the wireless network interface unit is responsible for inputting and outputting other forms of data except network data. 6. The artificial intelligence edge computing system according to claim 1, wherein the edge collection computing device further comprises a security protection unit, and the security protection unit has a public key authentication mechanism and a private key authentication mechanism, and provides Single or batch edge device authentication verification process. 7. The computing system of artificial intelligence edge according to claim 1, is characterized in that: described mobile edge server also comprises judging module and alarm instruction generation module, wherein, described wireless network interface unit collects data related to execution end Judging module, to judge whether the data related to the execution end is abnormal; Alarm instruction generation module, for when the judgment result is yes, generate an alarm instruction; The wireless network interface unit sends the alarm instruction to a display unit, and the display unit executes the the alarm command. 8. a computing method of artificial intelligence edge, it is realized in the system as described in any one of claim 1 to 7, it is characterized in that, comprise the following steps: Step S10, the mobile edge server and the edge collection computing device use the collected large amount of historical data to perform deep learning model training, generate a cloud deep learning model and deliver it to the AIOS operating system unit; Step S11, the IoT device collects environmental information, and converts the environmental information into data to be processed; In step S12, the edge collection computing device obtains the monitoring feature data collected by multiple IoT devices connected to the edge collection computing device, and after processing, it is sent to the cloud deep learning model in the AIOS operating system, and the results are output to predict the specified data. Whether an exception occurs in the area; Step S13, the wireless network interface unit of the mobile edge server collects the data related to the execution end; and judges whether the data related to the execution end and the cloud deep learning model is abnormal through the judgment module; when it is judged that the data is abnormal, the alarm instruction generation module generates alarm command; Step S14, the wireless network interface unit sends the alarm instruction to a display unit, and the display unit executes the alarm instruction. 9. The method of claim 8, wherein the step S12 further comprises: The data acquisition unit collects the environmental data of luminance, temperature, humidity and PM2.5, the collected data is analog data or digital data, and the data collection unit sends the collected data to the data conversion unit after classifying and processing; The data conversion unit performs format conversion on the data and sends it to the data cleaning unit; The data cleaning unit cleans the data, and then the data aggregation computing unit performs the aggregation calculation to form the sorted data; Input the sorted data into the cloud deep learning model in the AIOS operating system, and output the result. 10 . The method of claim 9 , further comprising: the security protection unit implements a single or batch edge device authentication verification process according to the public key authentication mechanism and the private key authentication mechanism it has. 11 .

Images