Intelligent factory multi-device cooperative control method and device based on 6G network and medium

Abstract

The invention discloses an intelligent factory multi-device cooperative control method and device based on a 6G network and a medium, and can be applied to the technical field of the 6G network. According to the invention, after real-time data obtained by data acquisition of a plurality of devices in an intelligent factory is obtained through a 6G base station, traffic prediction of the plurality of devices is carried out according to the real-time data to obtain future traffic data, and then the task priority corresponding to each device in the intelligent factory is obtained; carrying out AI dynamic network resource scheduling according to the future traffic data and the task priority; after the device channel state after AI dynamic network resource scheduling and the 6G signal fingerprint and sensor data corresponding to each device are obtained, device energy consumption optimization is carried out on the intelligent factory according to the device channel state, device positioning is carried out on the intelligent factory according to the 6G signal fingerprint, and a 3D model of the intelligent factory is updated according to the sensor data; therefore, the positioning precision and the data fusion efficiency can be effectively improved, and the dynamic adjustment of resources is realized.

Description

Technical Field

[0001] This application relates to the field of 6G network technology, and in particular, to a multi-device collaborative control method, device, and medium for an intelligent factory based on a 6G network. Background Art

[0002] In related technologies, with the rapid development of industry and intelligent manufacturing, factories have an increasing demand for real-time communication, high-precision collaboration, and intelligent decision-making. Although 5G technology has made breakthroughs in bandwidth (1 - 10 Gbps) and latency (1 - 10 ms), its performance still has problems such as rigid network resources, insufficient spatio-temporal synchronization accuracy, and low data fusion efficiency in high-concurrency industrial scenarios (such as multi-AR device collaboration, real-time 3D model loading). Although 6G networks have been applied in industry currently, there are still problems such as rigid network resources, insufficient spatio-temporal synchronization accuracy, and low data fusion efficiency in the current application of 6G networks.

[0003] In summary, the technical problems existing in related technologies need to be improved. Summary of the Invention

[0004] The main purpose of the embodiments of this application is to propose a multi-device collaborative control method, device, and medium for an intelligent factory based on a 6G network, which can effectively improve the positioning accuracy, data fusion efficiency, and achieve dynamic resource adjustment.

[0005] To achieve the above object, on the one hand, an embodiment of this application proposes a multi-device collaborative control method for an intelligent factory based on a 6G network, and the method includes the following steps:

[0006] Obtain real-time data collected by multiple data acquisition modules for multiple devices in the intelligent factory, and the data acquisition modules transmit data through 6G base stations;

[0007] Perform traffic prediction on multiple devices in the intelligent factory according to the real-time data to obtain future traffic data;

[0008] Obtain the task priority corresponding to each device in the intelligent factory;

[0009] Perform AI dynamic network resource scheduling according to the future traffic data and the task priority;

[0010] Obtain the device channel state, 6G signal fingerprint corresponding to each device, and sensor data in the intelligent factory after performing AI dynamic network resource scheduling;

[0011] Optimize the device energy consumption of the intelligent factory according to the device channel state;

[0012] Locate the devices in the smart factory according to the 6G signal fingerprint;

[0013] Update the 3D model of the smart factory according to the sensor data.

[0014] In some embodiments, the traffic prediction of multiple devices in the smart factory according to the real-time data includes:

[0015] Input the real-time traffic data of the real-time data at multiple time points into a long short-term memory network to perform traffic prediction on multiple devices in the smart factory; wherein, the prediction formula is as follows:

[0016] yt = LSTM(x t-1 , x t-2 ,..., x t-n ) + ∈ t ;

[0017] In the formula, x t-n represents the real-time data collected at time point t - n; yt represents the predicted future traffic data; ∈ t represents the reinforcement learning dynamic correction term, and LSTM represents the long short-term memory network.

[0018] In some embodiments, the AI dynamic network resource scheduling according to the future traffic data and the task priority includes:

[0019] Obtain the network load rate and the number of device connections. The number of device connections is used to characterize the load number of connections to the network, and the network load rate is used to characterize the ratio of the load of data transmission connected to the network to the number of device connections;

[0020] When the number of device connections is greater than a preset threshold, perform AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections.

[0021] In some embodiments, the AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections includes:

[0022] Calculate the average delay and packet loss rate corresponding to all tasks in the current network state;

[0023] Determine the reward value according to the average delay and the packet loss rate;

[0024] Determine the state space according to the future traffic data, the task priority, the network load rate, and the number of device connections;

[0025] Adjust the broadband occupancy ratio of the network slice in the action space according to the reward value and the state space.

[0026] In some embodiments, the AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections further includes:

[0027] Determine the task urgency weight according to the task priority;

[0028] Determine the required rate of the current task for the corresponding device according to the future traffic data;

[0029] Calculate the priority weight according to the task urgency weight and the required rate of the current task;

[0030] Adjust the broadband occupancy ratio of the network slice according to the priority weight.

[0031] In some embodiments, the device energy consumption optimization of the intelligent factory according to the device channel state includes:

[0032] Perform Nash equilibrium calculation according to the device channel state to obtain the first candidate transmission power of each device;

[0033] Construct the objective function of minimizing the total device energy consumption corresponding to the intelligent factory;

[0034] Calculate the objective function of minimizing the total device energy consumption according to the task completion constraint condition and the power limit condition corresponding to the intelligent factory to obtain the second candidate transmission power;

[0035] Adjust the first candidate transmission power according to the second candidate transmission power.

[0036] In some embodiments, the device positioning of the intelligent factory according to the 6G signal fingerprint includes:

[0037] Calculate the first positioning error data according to the 6G signal fingerprint, and the first positioning error data is used to characterize the error formed by positioning based on electromagnetic waves;

[0038] Calculate the second positioning error data corresponding to each device, and the second positioning error data is used to characterize the error formed by positioning based on wireless communication;

[0039] Calculate the third positioning error data corresponding to each device, and the third positioning error data is used to characterize the error formed by positioning based on the accelerometer and gyroscope;

[0040] Calculate the total positioning error of each device according to the first positioning error data, the second positioning error data, and the third positioning error data;

[0041] Determine the target position of each device in the smart factory according to the total positioning error.

[0042] To achieve the above object, another aspect of the embodiments of the present application provides a multi-device collaborative control device for a smart factory based on a 6G network, the device includes:

[0043] A first module, configured to obtain real-time data collected by a variety of data collection modules for multiple devices in the smart factory, and the data collection module transmits data through a 6G base station;

[0044] A second module, configured to perform traffic prediction on multiple devices in the smart factory according to the real-time data to obtain future traffic data;

[0045] A third module, configured to obtain the task priority corresponding to each device in the smart factory;

[0046] A fourth module, configured to perform AI dynamic network resource scheduling according to the future traffic data and the task priority;

[0047] A fifth module, configured to obtain the device channel state, the 6G signal fingerprint corresponding to each device, and sensor data in the smart factory after performing AI dynamic network resource scheduling;

[0048] A sixth module, configured to optimize the device energy consumption of the smart factory according to the device channel state;

[0049] A seventh module, configured to perform device positioning on the smart factory according to the 6G signal fingerprint;

[0050] An eighth module, configured to update the 3D model of the smart factory according to the sensor data.

[0051] To achieve the above object, another aspect of the embodiments of the present application provides a computer device, including:

[0052] At least one processor;

[0053] At least one memory, configured to store at least one program;

[0054] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0055] To achieve the above object, another aspect of the embodiments of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0056] The embodiments of the present application at least include the following beneficial effects: The present application provides a multi-device collaborative control method, device, and medium for an intelligent factory based on a 6G network. After obtaining the real-time data collected by the data collection module for multiple devices in the intelligent factory through a 6G base station, future traffic data is obtained by predicting the traffic of multiple devices in the intelligent factory based on the real-time data. Then, after obtaining the task priority corresponding to each device in the intelligent factory, AI dynamic network resource scheduling is performed based on the future traffic data and task priority. Next, after obtaining the device channel state, 6G signal fingerprint, and sensor data of each device in the intelligent factory after AI dynamic network resource scheduling, device energy consumption optimization is performed on the intelligent factory according to the device channel state, device positioning is performed on the intelligent factory according to the 6G signal fingerprint, and the 3D model of the intelligent factory is updated according to the sensor data, thereby effectively improving the positioning accuracy, data fusion efficiency, and realizing dynamic resource adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flowchart of the multi-device collaborative control method for an intelligent factory based on a 6G network provided by the embodiments of the present application;

[0058] Figure 2 is a schematic diagram of data processing of the AI-driven multi-device collaborative optimization mechanism provided by the embodiments of the present application;

[0059] Figure 3 is a schematic diagram of data processing of the ultra-reliable and low-latency dynamic resource allocation mechanism provided by the embodiments of the present application;

[0060] Figure 4 is a schematic diagram of resource scheduling based on task priority provided by the embodiments of the present application;

[0061] Figure 5 is a schematic diagram of hybrid positioning provided by the embodiments of the present application;

[0062] Figure 6 is a schematic diagram of the fusion of the real-time 3D model and sensor data provided by the embodiments of the present application;

[0063] Figure 7 is a schematic diagram of energy consumption optimization provided by the embodiments of the present application;

[0064] Figure 8 is a schematic diagram of real-time data processing and feedback based on edge computing provided by the embodiments of the present application;

[0065] Figure 9 is a general schematic diagram of the multi-device collaborative control method for an intelligent factory based on a 6G network provided by the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0066] In order to make the objectives, technical solutions, and advantages of this application more clearly understood, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the embodiments of this application. They are merely examples of devices and methods that are consistent with some aspects of the embodiments of this application.

[0067] It can be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, the first information may also be referred to as the second information. Similarly, the second information may also be referred to as the first information. Depending on the context, as used herein, the words "if", "when" can be interpreted as "when...", "when...", or "in response to determining".

[0068] The terms "at least one", "multiple", "each", "any one", etc. used in this application, at least one includes one, two, or more than two, multiple includes two or more than two, each refers to each of the corresponding multiple, and any one refers to any one of the multiple.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0070] In related technologies, with the rapid development of industry and intelligent manufacturing, factories have an increasing demand for real-time communication, high-precision collaboration, and intelligent decision-making. Although 5G technology has made breakthroughs in bandwidth (1 - 10 Gbps) and latency (1 - 10 ms), its performance still has problems such as rigid network resources, insufficient spatio-temporal synchronization accuracy, and low data fusion efficiency in high-concurrency industrial scenarios (such as multi-AR device collaboration, real-time 3D model loading). Although 6G networks have been applied in industry currently, however, the current application of 6G networks still has problems such as rigid network resources, insufficient spatio-temporal synchronization accuracy, and low data fusion efficiency.

[0071] In view of this, embodiments of this application provide an intelligent factory multi-device collaborative control method, device, and medium based on 6G networks, which can effectively improve positioning accuracy, data fusion efficiency, and achieve dynamic resource adjustment.

[0072] The embodiments of the present application will be specifically described below with reference to the accompanying drawings:

[0073] Figure 1 is an alternative flowchart of the multi-device collaborative control method for an intelligent factory based on a 6G network provided by the embodiments of the present application. Figure 1 The method in may include but is not limited to steps S110 to S180:

[0074] Step S110, obtain the real-time data collected by multiple data acquisition modules for multiple devices in the intelligent factory, where the data acquisition modules transmit data through a 6G base station;

[0075] Step S120, perform traffic prediction on multiple devices in the intelligent factory based on the real-time data to obtain future traffic data;

[0076] Step S130, obtain the task priority corresponding to each device in the intelligent factory;

[0077] Step S140, perform AI dynamic network resource scheduling based on the future traffic data and task priorities;

[0078] Step S150, obtain the device channel status, the 6G signal fingerprint corresponding to each device, and sensor data in the intelligent factory after performing AI dynamic network resource scheduling;

[0079] Step S160, optimize the device energy consumption of the intelligent factory according to the device channel status;

[0080] Step S170, locate the devices in the intelligent factory according to the 6G signal fingerprints;

[0081] Step S180, update the 3D model of the intelligent factory according to the sensor data.

[0082] It can be understood that the data acquisition modules can be AR glasses, industrial sensors, or cameras. Among them, the AR glasses support 4K resolution display and SLAM (Simultaneous Localization and Mapping) function, and receive 3D operation guidance in real time. The industrial sensors include a temperature sensor (accuracy of ±0.1°C), a vibration sensor (sampling rate of 10 kHz), and a pressure sensor (range of 0 - 100 MPa). The cameras can be deployed with wide-angle 4K cameras (frame rate of 120 fps) for dynamic monitoring of the production line. The 6G (Sixth Generation Mobile Communication) base station can support the terahertz band (0.1 - 1 THz) and large-scale MIMO (256 antenna array), providing a peak rate of more than 10 Gbps. The edge server can be equipped with a GPU acceleration card (such as NVIDIA A100) and run lightweight AI models (model parameter quantity ≤ 1 MB).

[0083] Specifically, this embodiment can be applied to the edge intelligence layer. Among them, the edge intelligence layer is equipped with an AI dynamic resource scheduling engine, a high-precision positioning module, a real-time data fusion engine, and an energy consumption optimization controller. The AI dynamic resource scheduling engine includes three modules: traffic prediction, bandwidth allocation, and computing offloading. The high-precision positioning module is used to fuse 6G signal fingerprints, UWB (Ultra-Wideband), and IMU (Inertial Measurement Unit) data. The real-time data fusion engine can dynamically update the 3D model in the AR glasses based on Kalman filtering and LOD (Level of Detail) technology. The energy consumption optimization controller can dynamically allocate the device transmission power using a game theory model.

[0084] It can be understood that the multi-device collaborative control method for intelligent factories based on 6G networks proposed in this embodiment can adopt an AI (Artificial Intelligence)-driven multi-device collaborative optimization mechanism. Specifically, as Figure 2 shown, during the execution of the multi-device collaborative optimization mechanism, it includes but is not limited to the following steps:

[0085] Multi-device data collection: Devices such as AR glasses, sensors, and cameras collect real-time data.

[0086] Edge node traffic prediction: Based on the LSTM (Long Short-Term Memory) model, predict the data traffic distribution within the next 5 ms.

[0087] Dynamic bandwidth allocation: Dynamically adjust the bandwidth ratio of each device according to the prediction results and task priorities.

[0088] Local computing offloading: Migrate non-essential cloud processing tasks to edge nodes to reduce latency.

[0089] Cross-device data fusion: Perform spatio-temporal alignment and format conversion on multi-source data at the edge node.

[0090] Real-time feedback: Real-time feedback the processed data to AR devices and sensors to support industrial operations.

[0091] Specifically, the processing details of the execution process of the multi-device collaborative optimization mechanism include but are not limited to:

[0092] 1. Standardized data encapsulation:

[0093] 1.1 Data format: Adopt a unified JSON format, including device ID, data type (video / sensor / instruction), timestamp, and data payload.

[0094] Field example:

[0095] {

[0096] "device_id":"AR_001",

[0097] "type":"3D_model",

[0098] "timestamp":1625097600.123,

[0099] "data":{"model_id":"M100","vertices":[x1,y1,z1,...]}

[0100] }

[0101] 2. Flow prediction module:

[0102] Structure of the flow prediction model: A two-layer LSTM (Long Short-Term Memory network) is adopted. The input is the flow sequences of each device within the past 10 ms, and the output is the flow distribution in the next 5 ms.

[0103] Training data: Peak flow records in the factory's historical task logs (such as equipment maintenance, production line changeovers).

[0104] After the training of the flow prediction model in this embodiment, the real-time flow data of multiple time points is input into the Long Short-Term Memory network to perform flow prediction for multiple devices in the intelligent factory, and the future flow data is obtained. Among them, the prediction formula:

[0105] yt=LSTM(x t-1 ,x t-2 ,...,x t-n )+∈ t ;

[0106] Among them, x t-n represents the real-time data collected at time point t-n; yt represents the future flow data obtained by prediction; ∈ t represents the reinforcement learning dynamic correction term, and LSTM represents the Long Short-Term Memory network.

[0107] 3. Dynamic bandwidth allocation:

[0108] 3.1 Network slicing strategy:

[0109] High-priority slice: Allocate 70% of the bandwidth, dedicated to AR instructions and emergency alerts (delay ≤ 1 ms, jitter ≤ 0.1 ms).

[0110] Elastic slice: Allocate 30% of the bandwidth for regular monitoring videos (allowing a delay ≤ 10 ms).

[0111] 3.2 Preemption mechanism: When a high-priority task is triggered, the bandwidth of the elastic slice can be temporarily reduced to 10% to release resources to ensure critical tasks.

[0112] 4. Local computing offloading:

[0113] Unloading decision model: Based on DNN (Deep Neural Network) to judge the task processing path, with the input being the data volume, latency requirement, and edge node load.

[0114] Example scenario:

[0115] AR model rendering: Complete 3D model lightweighting (the number of vertices is reduced from 1 million to 200,000) at the edge node, and only transmit the differential data to the AR glasses.

[0116] Sensor filtering: The edge node performs FFT (Fast Fourier Transform) noise reduction on the vibration sensor data, reducing the uploaded data volume by 80%.

[0117] 5. Cross-device data fusion:

[0118] Spatio-temporal alignment algorithm:

[0119] Spatial alignment: Match the positioning coordinates (x, y, z) of the 6G base station with the SLAM data of the AR glasses, and the error compensation formula:

[0120]

[0121] Temporal alignment: Adopt PTP (Precision Time Protocol) to synchronize the clocks of each device, with the error ≤ 0.1ms.

[0122] In this embodiment, a multi-device collaborative optimization mechanism is set as the core decision engine, which is responsible for scheduling tasks and coordinating device behaviors.

[0123] It can be understood that after obtaining the future traffic data corresponding to the devices in the intelligent factory in this embodiment, dynamic network resource allocation is performed through an ultra-reliable and low-latency dynamic resource allocation mechanism. Specifically, in this embodiment, after continuously monitoring the network load rate and the number of device connections, when the number of device connections is greater than the preset threshold, AI dynamic network resource scheduling is performed according to the future traffic data, task priority, network load rate, and the number of device connections. Exemplarily, as Figure 3 shown, when the number of device connections (load number) is greater than the threshold, maintain the current network configuration; when the number of device connections is greater than the threshold, trigger the Q-learning algorithm to reallocate network resources, and calculate the optimal bandwidth ratio of each network slice according to the task priority and load situation. At the same time, this embodiment will also verify whether the latency and reliability meet the standards. If they meet the standards, the network configuration will be updated; otherwise, it will be recalculated.

[0124] The process of dynamic network resource allocation in this embodiment can adopt the Q-learning resource scheduling model. When performing resource scheduling based on the Q-learning resource scheduling model, in this embodiment, after calculating the average delay and packet loss rate corresponding to all tasks in the current network state, the reward value is determined according to the average delay and packet loss rate, and then the state space is determined according to the future traffic data, task priority, network load rate, and number of device connections. Then, according to the reward value and the state space, the broadband occupancy ratio of the network slice in the action space is adjusted. Specifically, in this embodiment, the network load rate (0%-100%), the number of device connections, and the task priority distribution are used as the state space, and adjusting the bandwidth occupancy ratio of each network slice (step size 5%) is used as the action space. Then, the reward function is calculated based on the following formula:

[0125]

[0126] In the formula, D avg (average delay):

[0127] Definition: The average end-to-end delay of all tasks in the current network state.

[0128] Unit: millisecond (ms).

[0129] Function: One of the core indicators to measure network performance. The lower the delay, the better the user experience, and the stronger the real-time performance of industrial tasks.

[0130] P drop (packet loss rate):

[0131] Definition: The ratio of the number of lost data packets to the total number of sent data packets during network transmission.

[0132] Unit: percentage (%).

[0133] Function: A key indicator to measure network reliability. The lower the packet loss rate, the higher the stability of network transmission.

[0134] R (reward value):

[0135] Definition: A scalar value used to evaluate the quality of the current action in the Q-learning algorithm.

[0136] Function: Positive rewards encourage the algorithm to select actions that can reduce delay and packet loss rate. Negative rewards punish actions that lead to performance degradation and drive the algorithm to explore better strategies.

[0137] It can be understood that when performing dynamic resource allocation in this embodiment, the priority of tasks can also be considered for allocation. Specifically, in this embodiment, after calculating the priority weight, the resources are reallocated based on the priority weight:

[0138] wi = α·UrgencyLevel + β·SafetyImpact;

[0139] In the formula, α and β are the assigned weight values, and tasks with higher weights can preempt priority resources with lower weights.

[0140] After determining the resource reallocation strategy in this embodiment, when α = 0.7 and β = 0.3, the resource allocation model is simulated, and the simulation delay is optimized from the static allocation delay of 15 ms to the dynamic allocation delay of 2 ms; at the same time, the reliability is improved, and the packet loss rate of critical tasks is reduced from 0.1% to 0.001%.

[0141] In some other embodiments, when performing resource allocation in this embodiment, the task urgency weight is also determined according to the task priority, then the required rate of the current task of the corresponding device is determined according to the future traffic data, and then the priority weight is calculated according to the task urgency weight and the required rate of the current task, and the broadband occupancy ratio of the network slice is adjusted according to the priority weight. Specifically, as Figure 4 shown, when a new task arrives in the devices of the smart factory, the task priority is judged and it is determined whether the task is an urgent task (such as device fault alarm) according to the task priority. If it is an urgent task, a dedicated network slice is allocated to ensure the effective execution of high-priority tasks. If it is an ordinary task, the ordinary task is added to the elastic queue for weighted round-robin scheduling. It can be understood that when executing an urgent task in this embodiment, the bandwidth resources of ordinary tasks can be temporarily preempted and the resources are released after the task is completed for subsequent tasks to use.

[0142] In this embodiment, the priority weight of task i can be calculated by the following formula:

[0143]

[0144] In the formula, w i is the task urgency weight, R i is the required rate of the task, and B total is the total bandwidth.

[0145] Based on the priority weight, the bandwidth B is allocated to task i through the following formula i :

[0146] B i = W i ·B total ;

[0147] If a high-priority task k arrives, the bandwidth of the low-priority task i is temporarily reduced:

[0148] B i ' = B i - ΔBk ;

[0149] where, ΔB k is the bandwidth required for task k.

[0150] It can be understood that after resource reallocation in this embodiment, device positioning in the smart factory can be performed according to 6G signal fingerprints. In this embodiment, first positioning error data can be calculated based on 6G signal fingerprints, and the first positioning error data is used to characterize the error formed by positioning based on electromagnetic waves; at the same time, second positioning error data corresponding to each device and used to characterize the positioning error formed by wireless communication is calculated, and third positioning error data corresponding to each device and used to characterize the positioning error formed by the accelerometer and gyroscope is calculated. Then, the total positioning error of each device is calculated based on the first positioning error data, the second positioning error data, and the third positioning error data; and then the target position of each device in the smart factory is determined according to the total positioning error. Among them, the first positioning error data can be RSSI error data, the second positioning error data can be UWB error data, and the third positioning error data can be IMU error data.

[0151] Exemplarily, as Figure 5 shown, the 6G base station sends a reference signal to the AR device. After receiving the 6G signal, UWB signal, and IMU data, the AR device calculates the RSSI fingerprint (device position) based on the 6G signal strength. At the same time, the positioning data is supplemented by UWB ranging and IMU inertial navigation. Then, the multi-source positioning data is fused through Kalman filtering to output centimeter-level coordinates. Finally, the high-precision coordinates are fed back to the AR device to support virtual information superposition. In this example, the process of 6G signal fingerprint positioning can first construct a fingerprint database. Among them, the construction of the fingerprint database can deploy reference points (with a spacing of 1 m) in a grid in the factory, and collect the RSSI (Received Signal Strength Indicator) and CSI (Channel State Information) at each point. The case number uses the KNN (K-Nearest Neighbor) algorithm to match the real-time signal with the fingerprint database, and the fingerprint database corresponding to a positioning error ≤ 3 cm is used as the target distance corresponding to the target first positioning error data.

[0152] The UWB-assisted positioning process can be to calculate the distance between the device and the base station based on ToF (Time of Flight), and the formula:

[0153]

[0154] where, c is the speed of light, and the ranging accuracy ≤ 5 cm.

[0155] IMU error compensation can first calculate the inertial navigation equation:

[0156]

[0157] In this embodiment, the displacement of the device is calculated from the data of the gyroscope and the accelerometer, and it is calibrated with the UWB positioning result every 10 ms.

[0158] At the same time, time synchronization calibration is also performed in this embodiment. Among them, master-slave clock synchronization: the edge server acts as the PTP master clock and broadcasts synchronization messages to all devices.

[0159] The time delay compensation calculation formula is as follows:

[0160]

[0161] In the formula, t1 is the time when the master clock sends, t2 is the time when the slave clock receives, t3 is the response time of the slave clock, and t4 is the time when the master clock receives the response.

[0162] Finally, the hybrid positioning error model is executed for positioning error calibration: specifically, by fusing 6G signal fingerprints, UWB, and IMU positioning data:

[0163]

[0164] In the formula, σ RSSI2 , σ UWB and σ IMU are the errors of each positioning technology, that is, the first positioning error data, the second positioning error data, and the third positioning error data respectively.

[0165] It can be understood that this embodiment will also update the 3D model of the intelligent factory according to the sensor data. Specifically, as Figure 6 shown, after real-time data collection through temperature, pressure, or vibration sensors in this embodiment, the sensor data is filtered by Kalman filtering to remove noise. Then, the 3D basic model of the factory equipment is loaded, and the 3D model parameters are updated according to the sensor data. At the same time, the model is dynamically LOD lightweight rendered according to the user's distance to adjust the model detail level and reduce the rendering load. Finally, the updated 3D model is sent to the AR device to support real-time operation guidance. Among them, the state equation of Kalman filtering denoising is as follows:

[0166] x k = Fx k-1 + Bu k + w k ;

[0167] In the formula, F is the state transition matrix, and w k is the process noise.

[0168] The observation equation of Kalman filtering denoising is as follows:

[0169] z k = Hx k + vk ;

[0170] Modify the predicted state through the sensor observation value z k Modify the predicted state.

[0171] The level-of-detail formula for LOD rendering optimization is as follows:

[0172]

[0173] Dynamically reduce the number of polygons according to the distance between the user and the model.

[0174] It can be understood that this embodiment can also optimize the device energy consumption of the intelligent factory according to the device channel state. Specifically, as Figure 7 shown, this embodiment can perform Nash equilibrium calculation according to the device channel state to obtain the first candidate transmission power of each device; at the same time, construct the objective function of minimizing the total device energy consumption corresponding to the intelligent factory; calculate the second candidate transmission power according to the task completion constraint conditions and power limit conditions corresponding to the intelligent factory for the objective function of minimizing the total device energy consumption; adjust the first candidate transmission power according to the second candidate transmission power. Among them, the objective function of minimizing the total device energy consumption is used to minimize the total device energy consumption while meeting the task requirements and network performance constraints:

[0175]

[0176] where P i is the transmission power of the i-th device.

[0177] The task completion constraint conditions are as follows:

[0178]

[0179] where h i is the channel gain, σ 2 is the noise power, and R min is the minimum rate required for the task.

[0180] The power limit conditions are as follows:

[0181] 0 ≤ Pi ≤ Pmax;

[0182] where Pmax is the maximum transmission power of the device.

[0183] Nash equilibrium power allocation:

[0184] Solve the optimal power allocation through the game theory model:

[0185]

[0186] where λ is the Lagrange multiplier, and (x) + represents max(x, 0).

[0187] In the embodiments of the present application, the embodiments will also process and feedback real-time data based on edge computing. Specifically, as Figure 8 shown, when real-time data is generated by AR devices, sensors, etc., it is determined whether the real-time data is a real-time instruction (such as an operation guide). If it is a real-time instruction, local processing is performed at the edge node. If it is historical data, it is uploaded to the cloud for long-term storage and analysis. Among them, the edge node is used to generate real-time decisions and feedback the real-time decisions to the execution unit (such as AR glasses) for execution. It can be understood that the offloading revenue function is defined in the computing offloading decision model as follows:

[0188] Profit = α·T save -β·E edge ;

[0189] where T save is the saved time delay, E edge is the energy consumption of edge computing, and α and β are weight coefficients.

[0190] In the offloading decision rule, if the offloading revenue is greater than the threshold θ, edge computing is selected:

[0191]

[0192] where T edge is the edge computing time, and T cloud is the cloud computing time.

[0193] In edge computing load balancing, the computing load L of edge node j j is:

[0194]

[0195] where C i is the computing amount of task i, and F j is the computing power of edge node j.

[0196] From the above, it can be seen that in the application process of the method of the embodiments of the present application, as Figure 9 shown, after real-time data is collected through AR glasses, sensors, and cameras, it is sent to the edge intelligent node through the 6G base station, and AI dynamic resource scheduling is performed in the edge intelligent node to achieve high-precision positioning, real-time update of the 3D model of the intelligent factory, and improvement of device battery life.

[0197] Therefore, the method of this embodiment has the following beneficial effects:

[0198] First, through the AI-driven dynamic resource allocation mechanism and edge computing optimization, the end-to-end latency is reduced, significantly improving the real-time performance and user experience of AR-assisted operations.

[0199] Second, by integrating 6G signal fingerprint, UWB, and IMU positioning technologies, the positioning accuracy is improved to 1 cm, ensuring the precise matching of virtual information with actual devices.

[0200] Third, through the ultra-reliable low-latency dynamic resource allocation mechanism and network slicing technology, the transmission success rate of critical tasks is increased to ≥99.9999%, ensuring the uninterrupted operation of industrial tasks.

[0201] Fourth, through the Q-learning-driven dynamic resource scheduling and priority task preemption mechanism, the packet loss rate is reduced to ≤0.001%, ensuring the integrity and reliability of data transmission.

[0202] Fifth, through the game theory-based dynamic power control mechanism, the device battery life is extended, reducing the efficiency loss caused by frequent charging.

[0203] Sixth, through the AI-driven multi-device collaboration optimization mechanism, it supports up to 1000 devices to access simultaneously, and the resource allocation is dynamically adjusted to ensure stable performance in high-concurrency scenarios.

[0204] Seventh, through the task priority scheduling mechanism, bandwidth and computing resources can be dynamically allocated according to the task urgency (such as device fault alarms) to ensure the priority execution of critical tasks.

[0205] Eighth, through the high-precision positioning and time synchronization mechanism, centimeter-level positioning and sub-millisecond-level synchronization can be achieved in complex environments, ensuring the accuracy and reliability of operations, and is applicable to complex industrial scenarios.

[0206] This application embodiment provides a multi-device collaborative control device for an intelligent factory based on a 6G network corresponding to Figure 1 The device includes:

[0207] The first module is used to obtain real-time data collected by multiple data collection modules for multiple devices in the intelligent factory, where the data collection modules transmit data through 6G base stations;

[0208] The second module is used to perform traffic prediction on multiple devices in the intelligent factory based on the real-time data to obtain future traffic data;

[0209] The third module is used to obtain the task priority corresponding to each device in the intelligent factory;

[0210] The fourth module is used to perform AI dynamic network resource scheduling based on the future traffic data and the task priority;

[0211] The fifth module is used to obtain the device channel status, the 6G signal fingerprint corresponding to each device, and the sensor data in the intelligent factory after AI dynamic network resource scheduling;

[0212] The sixth module is used to optimize the device energy consumption of the intelligent factory according to the device channel status;

[0213] The seventh module is used to locate the devices in the intelligent factory according to the 6G signal fingerprint;

[0214] The eighth module is used to update the 3D model of the intelligent factory according to the sensor data.

[0215] It can be understood that the content in the above method embodiments is applicable to the device embodiments of the present application. The functions specifically implemented in the device embodiments of the present application are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0216] The embodiment of the present application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above method is implemented. The computer device can be any intelligent terminal including a tablet computer, an in-vehicle computer, etc.

[0217] It can be understood that the content in the above method embodiments is applicable to the device embodiments of the present application. The functions specifically implemented in the device embodiments of the present application are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0218] The embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0219] It can be understood that the content in the above method embodiments is applicable to the storage medium embodiments of the present application. The functions specifically implemented in the storage medium embodiments of the present application are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those in the above method embodiments.

[0220] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation to the technical solutions provided by the embodiments of the present application. Those skilled in the art know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0221] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation to the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0222] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0223] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations.

[0224] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of this application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0225] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0226] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the above units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0227] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0228] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0229] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store programs.

[0230] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the rights of the embodiments of the present application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the rights of the embodiments of the present application.

Claims

1. A multi-device collaborative control method for intelligent factories based on 6G networks, characterized in that, The method includes the following steps: Obtain the real-time data collected by multiple data acquisition modules for multiple devices in the intelligent factory, and the data acquisition modules transmit data through 6G base stations; Perform traffic prediction on multiple devices in the intelligent factory based on the real-time data to obtain future traffic data; Obtain the task priority corresponding to each device in the intelligent factory; Perform AI dynamic network resource scheduling according to the future traffic data and the task priority; Obtain the device channel state, the 6G signal fingerprint corresponding to each device, and the sensor data in the intelligent factory after AI dynamic network resource scheduling; Optimize the device energy consumption of the intelligent factory according to the device channel state; Locate the devices in the intelligent factory according to the 6G signal fingerprint; Update the 3D model of the intelligent factory according to the sensor data.

2. The method according to claim 1, characterized in that, The performing traffic prediction on multiple devices in the intelligent factory according to the real-time data includes: Input the real-time traffic data of the real-time data at multiple time points into a long short-term memory network to perform traffic prediction on multiple devices in the intelligent factory; wherein, the prediction formula is as follows: yt = LSTM(x t-1 , x t-2 ,..., x t-n ) + ∈ t ; In the formula, x t-n represents the real-time data collected at time point t - n; yt represents the predicted future flow data; ∈ t represents the reinforcement learning dynamic correction term, and LSTM represents the long short-term memory network.

3. The method according to claim 1, characterized in that, The performing AI dynamic network resource scheduling according to the future traffic data and the task priority includes: Obtain the network load rate and the number of device connections, where the number of device connections is used to represent the load number connected to the network, and the network load rate is used to represent the ratio of the load connected to the network and performing data transmission to the number of device connections; When the number of device connections is greater than a preset threshold, perform AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections.

4. The method according to claim 3, wherein The performing AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections includes: Calculate the average delay and packet loss rate corresponding to all tasks in the current network state; Determine the reward value according to the average delay and the packet loss rate; Determine the state space according to the future traffic data, the task priority, the network load rate, and the number of device connections; Adjust the broadband ratio of the network slice in the action space according to the reward value and the state space.

5. The method according to claim 4, characterized in that, The performing AI dynamic network resource scheduling according to the future traffic data, the task priority, the network load rate, and the number of device connections further includes: Determine the task urgency weight according to the task priority; Determine the required rate of the current task of the corresponding device according to the future traffic data; Calculate the priority weight according to the task urgency weight and the required rate of the current task; Adjust the broadband ratio of the network slice according to the priority weight.

6. The method according to claim 1, wherein The optimizing the device energy consumption of the intelligent factory according to the device channel state includes: Perform Nash equilibrium calculation according to the device channel state to obtain the first candidate transmission power of each device; Construct the objective function for minimizing the total device energy consumption corresponding to the intelligent factory; Calculate the objective function of minimizing the total device energy consumption according to the task completion constraint conditions and power limit conditions corresponding to the intelligent factory, and obtain the second candidate transmission power; Adjust the first candidate transmission power according to the second candidate transmission power.

7. The method according to claim 1, wherein The device positioning of the intelligent factory according to the 6G signal fingerprint includes: Calculate the first positioning error data according to the 6G signal fingerprint, and the first positioning error data is used to characterize the error formed by positioning based on electromagnetic waves; Calculate the second positioning error data corresponding to each device, and the second positioning error data is used to characterize the error formed by positioning based on wireless communication; Calculate the third positioning error data corresponding to each device, and the third positioning error data is used to characterize the error formed by positioning based on the accelerometer and gyroscope; Calculate the total positioning error of each device according to the first positioning error data, the second positioning error data, and the third positioning error data; Determine the target position of each device in the intelligent factory according to the total positioning error.

8. An intelligent factory multi-device collaborative control device based on a 6G network, characterized in that, The device includes: The first module is used to obtain the real-time data collected by multiple data collection modules for multiple devices in the intelligent factory, and the data collection module transmits data through a 6G base station; The second module is used to perform traffic prediction on multiple devices in the intelligent factory according to the real-time data to obtain future traffic data; The third module is used to obtain the task priority corresponding to each device in the intelligent factory; The fourth module is used to perform AI dynamic network resource scheduling according to the future traffic data and the task priority; The fifth module is used to obtain the device channel state, the 6G signal fingerprint corresponding to each device, and sensor data in the intelligent factory after performing AI dynamic network resource scheduling; The sixth module is used to optimize the device energy consumption of the intelligent factory according to the device channel state; The seventh module is used to perform device positioning on the intelligent factory according to the 6G signal fingerprint; The eighth module is used to update the 3D model of the intelligent factory according to the sensor data.

9. A computer device, characterized in that, Include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program implements the method according to any one of claims 1 to 7 when executed by a processor.

Images