An integrated sensing and control method and system for industrial internet of things

Abstract

The application provides an integrated sensing and control method and system for an industrial Internet of Things, comprising: a controller scheduling sensors in a work area to collect environmental data and production state data; receiving the data collected by the sensors and aggregating the data to generate a control task; evaluating the resource states of a local device and a cloud device, and calculating the total time delay of the control task in local and cloud processing; calculating the control cost based on optimal control theory; generating an offloading decision for the control task based on the total time delay and the control cost; distributing the control task to the local device or the cloud device for processing according to the offloading decision, and generating a processing result; generating a control instruction based on the processing result, and delivering the control instruction to corresponding production equipment and sensors for adjustment, and optimizing the sensor parameters and the offloading decision, thereby forming a control loop. The method provided by the application improves resource utilization, guarantees low time delay and high real-time performance, enhances system stability and reliability, and reduces operation and maintenance complexity and cost.

Description

Technical Field

[0001] This invention relates to the fields of industrial Internet of Things (IIoT) and data service technology, and in particular to an integrated method and system for sensing, computing and control for industrial IoT. Background Technology

[0002] The Industrial Internet of Things (IIoT), as a core technology for the digital transformation of industrial manufacturing, achieves intelligent and automated production processes by deeply integrating sensors, controllers, communication networks, and computing resources. However, with the increasing complexity and diversification of industrial scenarios, traditional technical architectures face severe challenges in resource management, communication quality, and equipment collaboration.

[0003] Specifically, existing technologies can be divided into two main categories: cloud-based industrial IoT systems and integrated sensing-transmission-control architectures.

[0004] For cloud-based industrial IoT systems, the architecture is "cloud-pipeline-edge-pipeline-terminal". The cloud is responsible for centralized data processing, providing the basic capabilities for business processing; the pipeline, through diverse remote communication technologies, ensures the breadth and reliability of device access; edge devices undertake real-time data acquisition, preprocessing, and uploading functions, effectively shortening the data processing chain; and terminal devices accurately collect various device information, enriching data sources. This technical solution improves the efficiency of data acquisition and analysis, providing better support for business applications. However, this technical solution relies on centralized processing in the cloud, which may lead to network latency issues, especially in scenarios requiring real-time response, making it difficult to meet the stringent low-latency requirements of the industrial sector. Furthermore, cloud resource orchestration is highly dependent on the network environment; in cases of poor network quality or interruption, the system may fail to operate normally, posing a potential threat to the continuity and stability of industrial production. Additionally, while utilizing cloud resources, the resource usage of the devices themselves in the industrial IoT is neglected, failing to fully utilize the local computing network resources of the devices.

[0005] For the integrated sensing-transmission-control architecture, based on the dynamically changing external disturbance sensing system state of the industrial system, state estimation is performed to support adaptive transmission for heterogeneous Quality of Service (QoS) requirements across multiple services, achieving optimized multi-domain resource allocation. This technical solution, through precise control and scheduling, can cope with complex network environments and changing operational demands, and can also optimize resource allocation, improving the overall system efficiency and response speed. However, this solution relies on sensing and transmission; network latency and bandwidth limitations may prevent real-time data from being transmitted in a timely manner, thus affecting the accuracy of control decisions. Simultaneously, the overall system lacks consideration for computing resources, which may affect the accuracy of sensing and control, especially in extreme environments or under resource constraints, potentially failing to meet the requirements for efficient and precise control. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide an integrated method and system for sensing, computing and control for the Industrial Internet of Things, in order to eliminate or improve one or more defects existing in the prior art.

[0007] On the one hand, the present invention provides an integrated sensing, computing, and control method for the Industrial Internet of Things, the method comprising the following steps:

[0008] Sensors within the scheduling work area collect environmental and production status data;

[0009] The system receives and aggregates data collected by various sensors to generate a control task; it assesses the resource status of both local and cloud environments and calculates the total processing latency of the control task locally and in the cloud, respectively; based on optimal control theory, it constructs a cost function to calculate the control cost; and based on the total latency and the control cost, it generates an offloading decision for the control task.

[0010] Based on the uninstallation decision, the control task is assigned to local or cloud processing to generate processing results;

[0011] Based on the processing results, control commands are generated and sent to the corresponding production equipment and sensors for adjustment, thereby optimizing the sensor parameters and the unloading decision.

[0012] In some embodiments of the present invention, sensors within the scheduling work area collect environmental data and production status data at a dynamic sampling rate, including:

[0013] The amount of environmental and production status data collected by the sensors satisfies the following calculation formula:

[0014]

[0015] in, α represents the amount of data collected by sensor k in work area n for control task m and control subtask i; α represents the sensor sampling rate; t represents the sampling time.

[0016] In some embodiments of the present invention, calculating the total latency of the control task being processed locally includes:

[0017] The data collected by each sensor is received and aggregated, resulting in a data aggregation delay, calculated as follows:

[0018]

[0019] Where max(·) represents taking the maximum value; This represents the sampling time of sensor k in work area n for control subtask i of control task m; Indicates the wireless transmission delay from the sensor to the controller;

[0020] The control task is processed to generate a local computation latency, calculated as follows:

[0021]

[0022] in, This represents the amount of data collected in control task m, specifically control subtask i, within work area n. Indicates the local computing speed;

[0023] The total local latency is obtained by adding the data aggregation latency and the local computation latency, calculated as follows:

[0024]

[0025] in, This indicates the data aggregation latency; This indicates the local computation latency.

[0026] In some embodiments of the present invention, the calculation step of the wireless transmission delay from the sensor to the controller includes:

[0027] Data collected by various sensors is received wirelessly, resulting in wireless transmission path loss, which is calculated as follows:

[0028]

[0029] Where, d n,m,k f represents the transmission distance between sensor k and controller n, which acquire data for control task m; n,m,k c represents the signal transmission frequency; c represents the speed of light.

[0030] The signal received power is calculated based on the wireless transmission path loss, using the following formula:

[0031]

[0032] in, Indicates the sensor's transmit power; ε tra Indicates the transmit gain; ε rec Indicates the receiver gain;

[0033] Based on Shannon's formula, the transmission rate from the sensor to the controller is calculated using the following formula:

[0034]

[0035] Among them, B n,m,k Indicates transmission bandwidth; N0 represents the received signal power; N0 represents the Gaussian white noise power spectral density.

[0036] The wireless transmission delay from the sensor to the controller is calculated based on the transmission rate from the sensor to the controller. The calculation formula is as follows:

[0037]

[0038] in, Indicates the amount of data transmitted wirelessly; This indicates the transmission rate from the sensor to the controller.

[0039] In some embodiments of the present invention, calculating the total latency of the control task being processed in the cloud includes:

[0040] The data collected by each sensor is received and aggregated, resulting in a data aggregation delay, calculated as follows:

[0041]

[0042] Where max(·) represents taking the maximum value; This represents the sampling time of sensor k in work area n for control subtask i of control task m; Indicates the wireless transmission delay from the sensor to the controller;

[0043] The control task is transmitted to the cloud and processed by the cloud, resulting in the transmission delay from the controller to the cloud and the cloud computing delay.

[0044] The formula for calculating the transmission latency from the controller to the cloud is:

[0045]

[0046] The formula for calculating the cloud computing latency is:

[0047]

[0048] in, This represents the amount of data collected in work area n for control task m and its subtask i; v n This indicates the transmission rate from the controller to the cloud; Indicates cloud computing speed;

[0049] The total cloud latency is obtained by adding the data aggregation latency, the transmission latency from the controller to the cloud, and the cloud computing latency. The calculation formula is as follows:

[0050]

[0051] in, This indicates the data aggregation latency; This indicates the transmission latency from the controller to the cloud; This indicates the cloud computing latency.

[0052] In some embodiments of the present invention, the method further includes:

[0053] Introducing a binary variable to represent whether the control task is processed locally or offloaded to the cloud; then the processing latency caused by the control task being processed locally or offloaded to the cloud is calculated as follows:

[0054]

[0055] in, Represents the binary variable, when This indicates that the control task is processed locally, when This indicates that the control task has been offloaded to the cloud for processing; This represents the amount of data collected in control task m, specifically control subtask i, within work area n. Indicates the local computing speed; This indicates the cloud computing speed.

[0056] In some embodiments of the present invention, a cost function is constructed based on optimal control theory to calculate the control cost, including:

[0057] The cost function satisfies the following formula:

[0058]

[0059] Where E[·] represents the mathematical expectation; T Indicates transpose; ξ i I represents the system state vector after the i-th control; n,m μ represents the total number of control operations performed by controller n on control task m. iLet represent the input vector for the i-th control; Δ represent the weight matrix of the system state vector ξ; and Θ represent the weight matrix of the input vector μ.

[0060] On the other hand, the present invention also provides an integrated sensing, computing, and control system for the industrial Internet of Things, the system comprising:

[0061] The perception layer consists of sensors from multiple work areas, used to collect environmental and production status data, and transmit the data to the controller layer via a wireless communication module.

[0062] The controller layer includes multiple controllers for implementing the steps of any of the methods described above;

[0063] The cloud layer, comprising multiple cloud servers, is used to receive and process control tasks offloaded by the controller layer; and to return the processing results to the controller layer.

[0064] The communication module connects the perception layer and the controller layer wirelessly; the controller layer and the cloud layer are connected via wired communication.

[0065] A closed-loop control module is used to optimize the sensor parameters and the unloading decision based on the adjustment effect.

[0066] In some embodiments of the present invention, the controller layer further includes a resource orchestration module for integrating the sensor's sensing resources, intra-system communication resources, the controller's and the cloud server's computing and intelligent resources through software-defined networking and virtualization technology to construct a virtualized resource pool.

[0067] On the other hand, the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of any of the methods mentioned above.

[0068] This invention provides an integrated sensing, computing, and control method and system for the Industrial Internet of Things (IIoT), comprising: a controller scheduling sensors within a work area to collect environmental and production status data; receiving and aggregating the data collected by each sensor to generate control tasks; evaluating the resource status of local and cloud environments, and calculating the total latency of processing the control tasks locally and in the cloud respectively; calculating the control cost based on optimal control theory; generating offloading decisions for the control tasks based on the total latency and control cost; allocating the control tasks to local or cloud environments for processing according to the offloading decisions, and generating processing results; generating control commands based on the processing results and sending them to the corresponding production equipment and sensors for adjustment, optimizing sensor parameters and offloading decisions, thereby forming a control closed loop. The method provided by this invention achieves reasonable scheduling and orchestration of diverse heterogeneous resources, fully utilizing the computing, storage, and communication resources of each device within the system. Simultaneously, the closed-loop control mechanism improves the consistency and adaptability of the system, enabling faster response to user needs and timely adaptive adjustments to changes in the work area environment, ensuring optimal control across local, regional, and global environments.

[0069] Furthermore, the integrated sensing, computing, and control system for the Industrial Internet of Things (IIoT) provided by this invention utilizes multi-mode communication to achieve interconnection and seamless connection between and within devices at different levels, realizing ultra-reliable low-latency communication to ensure real-time and stable communication, thereby enabling rapid and accurate response to control commands; and it uses software-defined networking and network virtualization technologies to construct a virtualized resource pool, supporting the integration and orchestration of heterogeneous resources, thereby improving the production performance of the IIoT.

[0070] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0071] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0072] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings:

[0073] Figure 1 This is a schematic diagram of the steps of an integrated sensing, computing, and control method for industrial Internet of Things in one embodiment of the present invention.

[0074] Figure 2This is a diagram of an integrated sensing, computing, and control system architecture for the Industrial Internet of Things (IIoT) according to one embodiment of the present invention.

[0075] Figure 3 This is a functional architecture diagram of an integrated sensing, computing, and control system for industrial IoT in one embodiment of the present invention.

[0076] Figure 4 This is a flowchart of an integrated sensing, computing, and control method for the industrial Internet of Things according to an embodiment of the present invention. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0078] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0079] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0080] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0081] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0082] It should be emphasized here that the step markers mentioned below are not a limitation on the order of the steps, but should be understood as meaning that the steps can be executed in the order mentioned in the embodiments, or in a different order than in the embodiments, or several steps can be executed simultaneously.

[0083] To address the problems of poor real-time performance, poor stability, low resource utilization, slow response speed, and insufficient computing resources in existing industrial IoT systems, this invention provides an integrated sensing, computing, and control method for industrial IoT. This integrated method refers to a processing approach that integrates communication, sensing, computing, and control, such as... Figure 1 As shown, the method includes the following steps S101 to S104:

[0084] Step S101: The sensors in the scheduling work area collect environmental data and production status data.

[0085] Step S102: Receive and aggregate the data collected by each sensor to generate a control task; evaluate the resource status of the local and cloud environments, and calculate the total latency of the control task in the local environment and the cloud environment respectively; construct a cost function based on optimal control theory to calculate the control cost; generate an offloading decision for the control task based on the total latency and control cost.

[0086] Step S103: Based on the uninstallation decision, the control task is assigned to local or cloud processing and the processing result is generated.

[0087] Step S104: Generate control commands based on the processing results and send them to the corresponding production equipment and sensors for adjustment, optimization of sensor parameters and unloading decisions.

[0088] Corresponding to this integrated sensing, computing, and control method for the Industrial Internet of Things (IIoT), the present invention also provides an integrated sensing, computing, and control system for the Industrial Internet of Things, such as... Figure 2 As shown, the system mainly includes a perception layer, a controller layer, and a cloud layer, and is equipped with a communication module and a closed-loop control module.

[0089] The sensing layer consists of multiple work zones. Because the production tasks and operations performed in different work zones vary, each work zone is equipped with a variety of differentiated sensors, such as temperature sensors, humidity sensors, and photosensors. These sensors are responsible for collecting environmental information and production status within their respective work zones and transmitting this data to the controller layer for processing.

[0090] The controller layer consists of multiple controllers, each with certain computing resources and a relatively high-precision Artificial Intelligence (AI) model deployed to process data transmitted from sensors. However, considering the limited computing and storage resources on the controllers, the AI ​​models deployed on the controllers are compressed, which correspondingly reduces the accuracy of the model's inference and control task processing. The controllers are responsible for receiving and aggregating data collected by various sensors, generating control tasks; evaluating the resource status of local and cloud resources, generating offloading decisions for control tasks; processing control tasks locally or offloading them to the cloud; and generating control commands and sending them to the perception layer.

[0091] In some embodiments, the controller layer further includes a resource orchestration module for integrating sensor sensing resources, intra-system communication resources, and the computing and intelligent resources of the controller and cloud server through software-defined networking and virtualization technologies to build a virtualized resource pool.

[0092] The cloud layer consists of multiple cloud servers. Compared to the processing power of the controller, the cloud servers have more powerful computing and storage resources. Therefore, more accurate artificial intelligence models can be deployed on the cloud servers to handle more complex intelligent control tasks. The cloud servers are responsible for receiving control tasks offloaded from the controller layer, processing them, and returning the processing results to the controller layer.

[0093] To enable communication between different layers, the system also includes a communication module. Specifically, the perception layer and the controller layer are connected wirelessly to minimize geographical limitations on device deployment locations and to meet the simultaneous access requirements of a large number of sensors, as well as the mobility needs of some sensors. The controller layer and the cloud layer are connected via wired communication, which ensures both the handling of massive amounts of control data and provides high-stability, high-speed communication.

[0094] To achieve a dynamic feedback optimization mechanism, the system also includes a closed-loop control module, which optimizes sensor parameters and unloading decisions based on the adjustment effect.

[0095] The preceding text, from a physical deployment perspective, proposed an integrated sensing, computing, and control system architecture for the Industrial Internet of Things (IIoT), clarifying the actual composition of the hardware / software components. In the following text, such as... Figure 3 As shown, from the perspective of logical functions, a functional architecture integrating sensing, computing and control for the Industrial Internet of Things is proposed, and the responsibilities and collaborative relationships of each level are described.

[0096] The integrated functional architecture for sensing, computing, and control in the Industrial Internet of Things (IIoT) includes an infrastructure layer, a computing and control layer, an application and service layer, and an orchestration and operation and maintenance layer. These layers are interdependent and coordinated to promote the realization of a closed-loop system control mechanism.

[0097] Infrastructure layer (corresponding to the perception layer and communication module):

[0098] The infrastructure layer utilizes Software Defined Network (SDN) and Network Functions Virtualization (NFV) technologies to integrate sensor perception resources, controller and cloud server computing and intelligent resources, and system communication resources to build a virtualized resource pool, i.e. a multi-dimensional heterogeneous resource fusion pool. This makes resource management more flexible, efficient, and scalable, providing a foundation for accurate and dynamic scheduling of resources within the system.

[0099] Computation and Control Layer (corresponding to Controller Layer):

[0100] The computational control layer integrates computation (based on local AI model data processing) and control (command generation) functions to achieve intelligent management of the Industrial Internet of Things (IIoT). By processing sensor data, it generates precise control commands, ensuring the continuity and reliability of the entire system, enabling real-time decision-making and adaptive response, and improving the efficiency and stability of industrial processes. Control functionality is the core of the IIoT, executed by the controller, including optimal control and end-to-end industrial optimization. It implements feedback-triggered control mechanisms, facilitating dynamic adjustments to system manufacturing and operation, and ensuring system adaptability and efficiency under changing conditions.

[0101] Service and Application Layer (corresponding to the cloud layer):

[0102] The service and application layer consists of two parts: service management and application support. Service management utilizes technologies such as Service Function Chain (SFC) and multi-precision artificial intelligence models to construct industrial service function chains, enabling full lifecycle management of services within the system. It integrates artificial intelligence technology to drive industrial services, improving service scalability, adaptability, and efficiency, and supporting diverse industrial services and dynamic demands. As mentioned above, compared to controllers, cloud servers are more powerful computing units, possessing abundant resources and more complete, uncompressed artificial intelligence models. Through parallel computing, they can more efficiently handle complex, data-intensive intelligent control tasks.

[0103] Orchestration and Operation Layer (corresponding to the closed-loop control module and resource orchestration module):

[0104] Based on the deep integration of sensing, communication, and computing, integrated intelligent multi-dimensional resource orchestration and intelligent operation are achieved, forming a closed-loop control. The resource orchestration function integrates and manages heterogeneous multi-dimensional resources through real-time status monitoring. When the controller load is too high, this layer enables collaborative computing between the controller and the cloud server, preventing control errors from affecting system safety through on-demand computation offloading. The orchestration layer plays a crucial role in ensuring efficient resource scheduling and rapid response to emergencies, and is key to ensuring system reliability. Intelligent operation and maintenance, based on resource orchestration, utilizes O&M technologies such as containers, AIOps, and serverless computing to achieve real-time system status perception and data-driven O&M. The Industrial Internet of Things (IIoT) combines human-machine interaction with Ultra-reliable and Low-latency Communication (URLLC) and high-performance intelligent computing, significantly improving production efficiency, optimizing resource allocation, reducing production costs, and promoting more intelligent and efficient industrial operations. The closed-loop control function enables global control and end-to-end management of the production line and system, while ensuring real-time correction of system deviations. It utilizes the feedback principle to automate system behavior and refine operations, thereby enhancing system robustness and ensuring stable and efficient industrial processes.

[0105] In the integrated sensing, computing, and control functional architecture for the Industrial Internet of Things (IIoT), the various layers are closely interconnected. The application and service layer provides highly available industrial services and relies on the optimized scheduling and task collaboration of the computing and control layer to achieve intelligent and automated functions. The computing and control layer, through functions such as feedback regulation and load balancing, relies on the sensing, communication, computing, and intelligent resources provided by the infrastructure layer to ensure the smooth execution of computing tasks. The infrastructure layer constructs a heterogeneous resource fusion pool to support the task requirements of the computing and control layer. The orchestration and operation layer plays a crucial role in this framework, coordinating and optimizing each layer through functions such as resource scheduling and status monitoring to ensure the overall stability and efficient operation of the system. A closed loop is formed between the layers, from resource support to service implementation to system optimization, with each layer depending on the others to drive the collaborative and efficient operation of the entire system.

[0106] Based on the above analysis and explanation of the integrated sensing, computing, and control system architecture and functional architecture for the Industrial Internet of Things (IIoT), the integrated sensing, computing, and control method for the Industrial Internet of Things (IIoT) will be further explained.

[0107] like Figure 4 The diagram shown is a flowchart of an integrated sensing, computing, and control method for the Industrial Internet of Things.

[0108] In step S101, the controller schedules the sensors in the work area to collect environmental data and production status data.

[0109] In some embodiments, the total amount of data collected by the sensor is calculated as shown in formula (1):

[0110]

[0111] in, α represents the amount of data collected by the controller n and sensor k in the control task m control subtask i within the work area; α represents the sensor sampling rate; t represents the sampling time.

[0112] In this invention, n represents the controller, k represents the sensor, m represents the control task, and i represents the subtask of the control task m, which will not be elaborated further below.

[0113] In step S102, the controller receives and aggregates the data collected by each sensor to generate a control task. It assesses the resource status of both local and cloud environments, calculates the total latency of the control task processing locally and in the cloud, constructs a comprehensive cost function based on the total latency, and generates an offloading decision for the control task by minimizing the comprehensive cost function.

[0114] In some embodiments, if the control task is processed locally, the total latency should include data aggregation latency and local computation latency, wherein the data aggregation latency includes data acquisition latency and wireless transmission latency.

[0115] Data acquisition latency is the sampling time for the sensor to complete data acquisition.

[0116] Wireless transmission latency is the time it takes for the sensor to transmit the collected data to the local controller via a wireless network. Specifically:

[0117] Assuming that the wireless transmission path loss between the sensor and the controller only considers free space path loss, the formula for calculating the wireless transmission path loss is as shown in formula (2):

[0118]

[0119] Where, d n,m,k Indicates the transmission distance between the sensor and the controller; f n,m,k represents the signal transmission frequency; c represents the speed of light.

[0120] The signal received power is calculated based on the wireless transmission path loss, as shown in formula (3):

[0121]

[0122] in, Indicates the sensor's transmit power; ε tra Indicates the transmit gain; ε rec Indicates the receive gain.

[0123] Based on Shannon's formula, the transmission rate from the sensor to the controller is calculated as shown in formula (4):

[0124]

[0125] Among them, B n,m,k Indicates transmission bandwidth; N represents the received signal power; N0 represents the Gaussian white noise power spectral density.

[0126] Therefore, the formula for calculating wireless transmission delay satisfies formula (5):

[0127]

[0128] in, Indicates the amount of data transmitted wirelessly; This indicates the transmission rate from the sensor to the controller.

[0129] The calculation formula for the data aggregation delay generated by the controller receiving and aggregating the data collected by each sensor is shown in formula (6):

[0130]

[0131] Where max(·) represents taking the maximum value; Indicates the sensor sampling time; This indicates the wireless transmission delay from the sensor to the controller.

[0132] Local computation latency is the time it takes for the controller to process received data (such as artificial intelligence model inference, control logic operations, etc.). The formula for calculating local computation latency satisfies formula (7):

[0133]

[0134] in, Indicates the size of the data; Indicates the local computing speed.

[0135] Therefore, if the control task is processed locally, the total latency satisfies formula (8):

[0136]

[0137] in, Indicates the latency of data aggregation; This indicates the local computation latency.

[0138] In some embodiments, if the control task is processed in the cloud, the total latency should include data aggregation latency, wired transmission latency, and cloud computing latency.

[0139] The calculation method for data aggregation latency can be found in formula (6).

[0140] Wired transmission latency is the time it takes for the controller to transmit the control task to the cloud via wired connection. It should be noted that since the amount of data after task processing is relatively small, the transmission latency from the cloud server back to the controller is negligible.

[0141] The calculation formula for the transmission latency from the controller to the cloud is shown in formula (9):

[0142]

[0143] in, Indicates the size of the data; v n This indicates the transmission rate from the controller to the cloud.

[0144] Cloud computing latency is the time it takes for a cloud server to process data (such as high-precision AI model analysis). The formula for calculating cloud computing latency satisfies formula (10):

[0145]

[0146] in, Indicates the size of the data; This indicates the cloud computing speed, and the cloud computing speed is greater than the local computing speed, i.e.

[0147] Therefore, the control task is processed in the cloud, and the total latency satisfies formula (11):

[0148]

[0149] in, Indicates the latency of data aggregation; This indicates the transmission latency from the controller to the cloud; This indicates the latency of cloud computing.

[0150] In some embodiments, a binary variable is introduced. This indicates whether the control task is processed locally or offloaded to the cloud. For example, when... This indicates that the control task is processed locally, when If the control task is offloaded to the cloud for processing, then the original local computation latency formula (7) and the original cloud computation latency formula (10) can be integrated and modified into formula (12):

[0151]

[0152] In formula (12), Partially representing local computation latency, Partially representing cloud computing latency. Through analysis of... By assigning the appropriate values, the computation latency of the control task can be obtained, whether it is local or in the cloud.

[0153] Similarly, the original transmission delay formula (9) from the controller to the cloud can be modified to formula (13):

[0154]

[0155] That is, only when When control tasks are offloaded to the cloud for processing, there is a transmission delay between the controller and the cloud.

[0156] Similarly, the original total latency formula (8) for local calculation and the original total latency formula (11) for cloud calculation can be integrated and modified into formula (14):

[0157]

[0158] In some embodiments, based on optimal control theory, a cost function is constructed to calculate the control cost, as shown in equation (15):

[0159]

[0160] Where E[·] represents the mathematical expectation; T Indicates transpose; ξ i I represents the system state vector after the i-th control; n,m μ represents the total number of control operations performed by controller n on control task m. i Let represent the input vector for the i-th control; Δ represent the weight matrix of the system state vector ξ; and Θ represent the weight matrix of the input vector μ.

[0161] By evaluating the total latency and control cost J, it can be determined whether the control task needs to be offloaded to the cloud to utilize more powerful computing and intelligent resources for processing, thereby minimizing the total latency and control cost of completing the control operation.

[0162] In step S103, based on the unloading decision generated in step S102, the control task is assigned to either local or cloud processing. If processed locally, the controller directly processes the control task and obtains the processing result. If processed in the cloud, the controller unloads the control task to the cloud, where it is processed and the processing result is returned.

[0163] In step S104, control commands are generated based on the processing results obtained from the controller or cloud processing. These control commands are then sent to the corresponding production equipment and sensors at the underlying level for adjustments, such as adjusting sensor parameters (sampling rate, transmission power, etc.) and unloading decisions, optimizing the equipment production process, and forming a control closed loop.

[0164] Corresponding to the above method, the present invention also provides an electronic device including a computer device, the computer device including a processor and a memory, the memory storing computer instructions, the processor executing the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the electronic device performs the steps of the method as described above.

[0165] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0166] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0167] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0168] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0169] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for integrated sensing, computing, and control for the Industrial Internet of Things, characterized in that, The method includes the following steps: Data is collected by sensors within the scheduling work area; the data includes environmental data and production status data. The system receives and aggregates data collected by various sensors to generate a control task; it assesses the resource status of both local and cloud environments and calculates the total processing latency of the control task locally and in the cloud, respectively; based on optimal control theory, it constructs a cost function to calculate the control cost; and based on the total latency and the control cost, it generates an offloading decision for the control task. Based on the uninstallation decision, the control task is assigned to local or cloud processing for processing, and the processing result is obtained; Based on the processing results, control commands are generated and sent to the corresponding production equipment and sensors for adjustment, thereby optimizing the sensor parameters and the unloading decision. The calculation of the total latency for processing the control task locally includes: The data collected by each sensor is received and aggregated, resulting in a data aggregation delay, calculated as follows: ； in, This indicates taking the maximum value; Indicates the controller within the work area Scheduling sensors Control task Control subtasks Sampling time; Indicates the wireless transmission delay from the sensor to the controller; The control task is processed to generate a local computation latency, calculated as follows: ； in, Indicates the work area Control tasks collected in Control subtasks The size of the data; Indicates the local computing speed; The total local latency is obtained by adding the data aggregation latency and the local computation latency, calculated as follows: ； in, This indicates the data aggregation latency; This indicates the local computation latency; Calculate the total latency of the control task being processed in the cloud, including: The data collected by each sensor is received and aggregated, resulting in a data aggregation delay, calculated as follows: ； The control task is transmitted to the cloud and processed by the cloud, resulting in the transmission delay from the controller to the cloud and the cloud computing delay. The formula for calculating the transmission latency from the controller to the cloud is: ； The formula for calculating the cloud computing latency is: ； in, This indicates the transmission rate from the controller to the cloud; Indicates cloud computing speed; The total cloud latency is obtained by adding the data aggregation latency, the transmission latency from the controller to the cloud, and the cloud computing latency. The calculation formula is as follows: ； in, This indicates the transmission latency from the controller to the cloud; This indicates the cloud computing latency; Based on optimal control theory, a cost function is constructed to calculate the control cost, including: The cost function satisfies the following formula: ； in, Represents the mathematical expectation; Indicates transpose; Indicates the first The system state vector after the second control; Indicates controller Control task Total number of control operations; Indicates the first The input vector for secondary control; Represents the system state vector The weight matrix; Represents the input vector The weight matrix.

2. The integrated sensing, computing, and control method for the Industrial Internet of Things according to claim 1, characterized in that, The amount of environmental and production status data collected by the sensors satisfies the following calculation formula: ； in, Indicates the controller within the work area Scheduling sensors Collected control tasks Control subtasks The size of the data; Indicates the sensor sampling rate; Indicates the sampling time.

3. The integrated sensing, computing, and control method for the Industrial Internet of Things according to claim 1, characterized in that, The steps for calculating the wireless transmission delay from the sensor to the controller include: Data collected by various sensors is received wirelessly, resulting in wireless transmission path loss, which is calculated as follows: ； in, Indicates data acquisition and control task Data sensors With controller Transmission distance between; Indicates the signal transmission frequency; Represents the speed of light; The signal received power is calculated based on the wireless transmission path loss, using the following formula: ； in, Indicates the sensor's transmit power; Indicates transmit gain; Indicates the receiver gain; Based on Shannon's formula, the transmission rate from the sensor to the controller is calculated using the following formula: ； in, Indicates transmission bandwidth; This indicates the received signal power; This represents the power spectral density of Gaussian white noise; The wireless transmission delay from the sensor to the controller is calculated based on the transmission rate from the sensor to the controller. The calculation formula is as follows: ； in, Indicates the amount of data transmitted wirelessly; This indicates the transmission rate from the sensor to the controller.

4. The integrated sensing, computing, and control method for the Industrial Internet of Things according to claim 1, characterized in that, The method further includes: Introducing a binary variable to represent whether the control task is processed locally or offloaded to the cloud; then the processing latency caused by the control task being processed locally or offloaded to the cloud is calculated as follows: ； in, Represents the binary variable, when This indicates that the control task is processed locally, when This indicates that the control task is offloaded to the cloud for processing; Indicates the work area Control tasks collected in Control subtasks The size of the data; Indicates the local computing speed; This indicates the cloud computing speed.

5. A sensing, computing, and control integrated system for the Industrial Internet of Things, characterized in that, The system includes: The perception layer consists of sensors from multiple work areas, used to collect environmental and production status data, and transmit the data to the controller layer via a wireless communication module. The controller layer includes multiple controllers for implementing the steps of the method as described in any one of claims 1 to 4; The cloud layer, comprising multiple cloud servers, is used to receive and process control tasks offloaded by the controller layer; and to return the processing results to the controller layer. The communication module connects the perception layer and the controller layer wirelessly; the controller layer and the cloud layer are connected via wired communication. A closed-loop control module is used to optimize the sensor parameters and the unloading decision based on the adjustment effect.

6. The integrated sensing, computing, and control system for the Industrial Internet of Things according to claim 5, characterized in that, The controller layer also includes a resource orchestration module, which integrates the sensor's sensing resources, the system's communication resources, the controller's and the cloud server's computing and intelligent resources through software-defined networking and virtualization technologies to build a virtualized resource pool.

7. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method as described in any one of claims 1 to 4.

Images