AI-driven 6G edge computing resource scheduling method and system

Abstract

The invention relates to the technical field of computing power scheduling, in particular to an AI-driven 6G edge computing resource scheduling method and system, and the method comprises the steps: determining an edge robot, a center computing end and a plurality of edge computing ends based on a resource scheduling instruction, and determining a task difficulty index based on the edge robot, the data transmission rate is confirmed based on the edge robots, the center calculation end, the center distance, the total bandwidth, the number of the robots and the robot transmitting power, the edge pressure index is confirmed based on the edge calculation end, and the delay constraint index is confirmed based on the maximum tolerable delay and the total delay; and inputting the task difficulty index, the edge pressure index, the total delay, the delay constraint index and the total energy consumption into a pre-constructed AI scheduling model to obtain a scheduling score, and completing resource scheduling based on the target edge computing end and the edge robot. The intelligent degree of resource scheduling can be improved, and the task processing efficiency of the edge robot is improved.

Description

Technical Field

[0001] The present invention relates to the field of computing power scheduling technology, and in particular to an AI-driven 6G edge computing resource scheduling method and system. Background Art

[0002] With the rapid development of edge computing and robotics technologies, edge robots have been widely used in industrial, logistics, security, and other scenarios. Scheduling computing resources for edge robots and then completing task distribution and processing has become a key step in edge robot applications.

[0003] In the existing technology, most resource scheduling solutions use static or heuristic algorithm-based methods to allocate tasks.

[0004] While this approach can schedule system resources, existing solutions fail to effectively balance task energy consumption and latency during task scheduling, resulting in low overall computing energy efficiency and failing to meet the actual demand for efficient computing. Therefore, improving the intelligence of scheduling computing resources for edge robots and improving their task processing efficiency has become a pressing technical challenge. Summary of the Invention

[0005] The present invention provides an AI-driven 6G edge computing resource scheduling method and system, the main purpose of which is to improve the intelligence level of resource scheduling and improve the task processing efficiency of edge robots.

[0006] To achieve the above objectives, the present invention provides an AI-driven 6G edge computing resource scheduling method, comprising: Receive a resource scheduling instruction, and identify an edge robot, a central computing end, and multiple edge computing ends based on the resource scheduling instruction, wherein the edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; Obtain edge robot processing task information, total bandwidth, number of robots, and robot transmission power. Processing task information includes: center distance, data size, and maximum tolerable delay; Determine the task difficulty index based on the edge robot; Determine the data transmission rate based on the edge robot, central computing terminal, center distance, total bandwidth, number of robots, and robot transmission power; Perform the following operations on each of the multiple edge computing terminals: Determine the edge pressure index based on the edge computing end; Obtain edge resource data and edge bandwidth, where edge resource data includes: task load, CPU frequency, load capacitance, operating voltage, and clock frequency; Determine the total latency based on data size, data rate, center distance, data rate, and edge bandwidth; Determine the total energy consumption based on data size, data transmission rate, robot transmission power and edge resource data; Determine the delay constraint index based on the maximum tolerable delay and the total delay; Input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain the scheduling score; Summarize the scheduling scores to obtain multiple scheduling scores, where the scheduling scores correspond one-to-one to the edge computing terminals; The target edge computing end is identified based on multiple scheduling scores, and resource scheduling is completed based on the target edge computing end and edge robot.

[0007] Optionally, determining a task difficulty index based on the edge robot includes: Obtain multiple sets of historical task information of the edge robot, where the historical task information includes: completion time, energy consumption, and failure rate; Extract multiple completion times, multiple energy consumptions, and multiple failure rates from multiple sets of historical task information; Normalizing the multiple completion times, the multiple energy consumptions, and the multiple failure rates respectively to obtain multiple standard times, multiple standard energy consumptions, and multiple standard failure rates; The weights of multiple standard times, multiple standard energy consumptions and multiple standard failure rates are calculated respectively to obtain time weights, energy consumption weights and failure rate weights; Determine an average time based on the multiple standard times, wherein the average time is an average value of the multiple standard times; Determine an average energy consumption based on multiple standard energy consumptions, wherein the average energy consumption is an average value of the multiple standard energy consumptions; Determining an average failure rate based on multiple standard failure rates, wherein the average failure rate is an average of the multiple standard failure rates; The task difficulty index is calculated based on the average time, average energy consumption, average failure rate, time weight, energy consumption weight, and failure rate weight. The calculation formula is as follows: in, represents the task difficulty index, represents the average time, represents the average energy consumption, represents the average failure rate, represents the time weight, represents the energy consumption weight, Represents the failure rate weight.

[0008] Optionally, determining the data transmission rate based on the edge robot, the central computing terminal, the center distance, the total bandwidth, the number of robots, and the robot transmission power includes: Noise is collected from the edge robot and the central computing end to obtain edge noise signals, wherein the collection frequency and collection time of the noise are preset; Performing a fast Fourier transform on the edge noise signal to obtain an amplitude spectrum, wherein the amplitude spectrum includes a plurality of frequency points and a plurality of amplitude values, wherein the frequency points correspond to the amplitude values ​​one to one; The average amplitude value is determined based on the amplitude spectrum. The calculation formula is as follows: in, represents the average amplitude value, Indicates the number of frequency points in multiple frequency points, Indicates the first of multiple frequency points The amplitude value of the frequency point, Indicates taking the absolute value; The noise power spectral density is calculated based on the acquisition frequency and average amplitude value. The calculation formula is as follows: in, represents the noise power spectral density, Indicates the acquisition frequency; The data transmission rate is calculated based on the center distance, total bandwidth, number of robots, robot transmission power, and noise power spectrum density. The calculation formula is as follows: in, Indicates the data transmission rate, Indicates the total bandwidth, Indicates the number of robots, Indicates the robot's transmission power, represents the center distance, is the preset noise index, is the natural logarithm.

[0009] Optionally, determining the edge pressure index based on the edge computing end includes: Obtain multiple historical queue information from the edge computing end, where the historical queue information includes: queue number and queue duration; Input multiple queue quantities from multiple historical queue information into the pre-built ARIMA model to obtain the future queue quantity; Determine an average queue time based on multiple queue times, where the average queue time is the average of the multiple queue times; The marginal pressure index is determined based on the number of future queues and the average queue time, wherein the marginal pressure index is the product of the number of future queues and the average queue time.

[0010] Optionally, determining the total delay based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth includes: The first delay is calculated based on the data size, data transmission rate and center distance. The calculation formula is as follows: in, represents the first delay, Indicates the data size, is the preset speed of light; The second delay is calculated based on the data size and edge bandwidth. The calculation formula is as follows: in, represents the second delay, represents the edge bandwidth; A total delay is determined based on the first delay and the second delay, wherein the total delay is the sum of the first delay and the second delay.

[0011] Optionally, determining the total energy consumption based on data size, data transmission rate, robot transmission power, and edge resource data includes: The first energy consumption is calculated based on the data size, data transmission rate and robot transmission power. The calculation formula is as follows: in, Indicates the first energy consumption; The second energy consumption is calculated based on the task load, CPU frequency, load capacitance, operating voltage, and clock frequency in the edge resource data. The calculation formula is as follows: in, represents the second energy consumption, represents the load capacitance, Indicates the operating voltage, Indicates the clock frequency, Indicates the CPU frequency, Indicates task load; A total energy consumption is determined based on the first energy consumption and the second energy consumption, wherein the total energy consumption is the sum of the first energy consumption and the second energy consumption.

[0012] Optionally, determining the delay constraint index based on the maximum tolerable delay and the total delay includes: The delay constraint index is calculated based on the maximum tolerable delay and the total delay. The calculation formula is as follows: in, represents the delay constraint index, represents the total delay, Indicates the maximum tolerable delay.

[0013] Optionally, the AI ​​scheduling model is as follows: in, represents the AI ​​scheduling model, represents the task difficulty index, is the edge pressure index, represents the delay constraint index, represents the total energy consumption, represents the total delay, is an exponential function, is the preset energy consumption delay ratio, is a natural constant.

[0014] Optionally, determining a target edge computing end based on multiple scheduling scores includes: Perform zero-removal screening on multiple scheduling scores to obtain a zero-removed scheduling score set; Determine the optimal scheduling score based on the zero-removal scheduling score set, wherein the optimal scheduling score is the minimum zero-removal scheduling score in the zero-removal scheduling score set; The edge computing end corresponding to the best scheduling score is used as the target edge computing end.

[0015] To achieve the above objectives, the present invention also provides an AI-driven 6G edge computing resource scheduling system, comprising: A basic equipment confirmation module is used to receive resource scheduling instructions and confirm the edge robot, the central computing end, and multiple edge computing ends based on the resource scheduling instructions. The edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; The task information processing module is used to obtain the processing task information, total bandwidth, number of robots, and robot transmission power of the edge robot. The processing task information includes: center distance, data size, and maximum tolerable delay. The task difficulty index is determined based on the edge robot, and the data transmission rate is determined based on the edge robot, center computing end, center distance, total bandwidth, number of robots, and robot transmission power. an edge information processing module, configured to perform the following operations on each of the plurality of edge computing terminals: determining an edge pressure index based on the edge computing terminal, obtaining edge resource data and edge bandwidth, wherein the edge resource data includes task load, CPU frequency, load capacitance, operating voltage, and clock frequency; determining a total delay based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth; determining a total energy consumption based on data size, data transmission rate, robot transmission power, and edge resource data; and determining a delay constraint index based on a maximum tolerable delay and total delay; The intelligent scoring scheduling module is used to input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain a scheduling score, summarize the scheduling scores, and obtain multiple scheduling scores. Among them, the scheduling score corresponds one-to-one to the edge computing end. Based on multiple scheduling scores, the target edge computing end is confirmed, and resource scheduling is completed based on the target edge computing end and the edge robot.

[0016] In order to solve the above problem, the present invention further provides an electronic device, comprising: a memory storing at least one instruction; and A processor executes instructions stored in the memory to implement the above-mentioned AI-driven 6G edge computing resource scheduling method.

[0017] In order to solve the above problems, the present invention also provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned AI-driven 6G edge computing resource scheduling method.

[0018] The present invention is to solve the problems described in the background technology. The present invention receives resource scheduling instructions and identifies edge robots, central computing terminals and multiple edge computing terminals based on the resource scheduling instructions. The edge robot and the central computing terminal have the function of data transmission, and the central computing terminal and each of the multiple edge computing terminals have the function of data transmission. It can be seen that the embodiment of the present invention confirms the edge robot, the central computing terminal and multiple edge computing terminals in advance, which facilitates the subsequent use of the central computing terminal to allocate the computing tasks generated by the edge robot to the edge computing terminal for computing and processing, thereby improving the intelligence of resource scheduling, and then obtaining the processing task information, total bandwidth, number of robots and robot transmission power of the edge robot. The processing task information includes: center distance, data size and maximum tolerable delay. It can be seen that the embodiment of the present invention obtains the processing task information, total bandwidth, number of robots and robot transmission power of the edge robot, which facilitates the subsequent analysis of the computing tasks generated by the edge robot, thereby calculating the data transmission rate, and confirming the task difficulty index based on the edge robot. It can be seen that the embodiment of the present invention analyzes the computing tasks generated by the edge robot and obtains the task difficulty index, which facilitates the subsequent automatic edge computing resource scheduling based on the task difficulty index, thereby improving the intelligence level of resource scheduling, and confirms the data transmission rate based on the edge robot, the center computing end, the center distance, the total bandwidth, the number of robots and the robot transmission power. Taking into account that data may be delayed during transmission, the data transmission rate is calculated to facilitate the subsequent edge computing resource scheduling based on the data transmission rate, improve the intelligence of resource scheduling, and perform the following operations on each edge computing end in multiple edge computing ends: Based on the edge computing end, the edge pressure index is confirmed. It can be seen that the embodiment of the present invention obtains the historical operation data of each edge computing end, thereby calculating the current edge pressure index of each edge computing end, obtaining edge resource data and edge bandwidth, wherein the edge resource data includes: task load, CPU frequency, load capacitance, operating voltage and clock frequency. It can be seen that the embodiment of the present invention obtains the real-time operation data of the edge computing end, which facilitates the subsequent calculation of total delay and total energy consumption. The total delay is determined based on the data size, data transmission rate, center distance, data transmission rate and edge bandwidth. The total energy consumption is determined based on the data size, data transmission rate, robot transmission power and edge resource data. It can be seen that the embodiment of the present invention comprehensively considers the computing tasks generated by the edge robot and the operating status of the edge computing terminal, and comprehensively calculates the total delay and total energy consumption required to complete the computing tasks, so as to facilitate the subsequent intelligent scheduling of edge computing resources according to the total delay and total energy consumption, so as to reduce delay and energy consumption and improve the task processing efficiency of the edge robot. The delay constraint index is determined based on the maximum tolerable delay and the total delay. It can be seen that the embodiment of the present invention determines the delay constraint index by comparing the maximum tolerable delay with the total delay, so as to facilitate the subsequent calculation of the scheduling score.Improve the task processing efficiency of the edge robot, input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain a scheduling score. It can be seen that the embodiment of the present invention uses the pre-built AI scheduling model to analyze the acquired data, thereby intelligently deriving a scheduling score, improving the intelligence level of resource scheduling, summarizing the scheduling scores, and obtaining multiple scheduling scores, wherein the scheduling scores correspond one-to-one to the edge computing end. It can be seen that the embodiment of the present invention calculates the scheduling score for each edge computing end and summarizes the scheduling scores to facilitate the subsequent confirmation of the best scheduling score, confirms the target edge computing end based on multiple scheduling scores, and completes resource scheduling based on the target edge computing end and the edge robot. It can be seen that the embodiment of the present invention confirms the best edge computing end through the best scheduling score, thereby distributing computing tasks to the best edge computing end, and then intelligently realizing edge computing resource scheduling. Therefore, the present invention can improve the intelligence level of resource scheduling and improve the task processing efficiency of the edge robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a flow chart of an AI-driven 6G edge computing resource scheduling method provided in one embodiment of the present invention; Figure 2 A functional module diagram of an AI-driven 6G edge computing resource scheduling system provided in one embodiment of the present invention; Figure 3 A schematic diagram of the structure of an electronic device for implementing the AI-driven 6G edge computing resource scheduling method provided in one embodiment of the present invention.

[0020] Description of reference numerals: 1. Electronic device; 10. Processor; 11. Storage; 12. Bus.

[0021] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0022] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0023] The embodiment of the present application provides an AI-driven 6G edge computing resource scheduling method. The execution subject of the AI-driven 6G edge computing resource scheduling method includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the AI-driven 6G edge computing resource scheduling method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0024] Reference Figure 1 FIG2 is a flow chart of an AI-driven 6G edge computing resource scheduling method according to an embodiment of the present invention. In this embodiment, the AI-driven 6G edge computing resource scheduling method includes: S1. Receive resource scheduling instructions, and identify edge robots, central computing terminals, and multiple edge computing terminals based on the resource scheduling instructions. The edge robot and the central computing terminal have the function of data transmission, and the central computing terminal and each of the multiple edge computing terminals have the function of data transmission.

[0025] It should be explained that the resource scheduling instruction is an instruction issued by a logistics dispatcher. For example, Xiao Zhang is a logistics dispatcher who needs to schedule computing resources for the edge robots in the logistics factory. Therefore, Xiao Zhang issues a resource scheduling instruction, and the resource scheduling instruction contains a specific number or code. Therefore, the corresponding edge robot, central computing end and multiple edge computing ends can be identified by the specific number or code. The edge robot refers to the sorting robot that performs sorting and handling tasks in the logistics factory. The central computing end is the computer responsible for distributing the computing tasks generated by the edge robot to the target edge computing end, wherein the computing task refers to the computing work that needs to be processed on the edge computing end during the process of sorting and handling goods in the logistics factory. For example, when the edge robot in the logistics factory is sorting and handling goods, it will generate task requirements including but not limited to image recognition, path planning, dynamic obstacle avoidance, real-time positioning and status monitoring, which are computing tasks. The edge computing end is the computer responsible for processing the computing tasks generated by the edge robot.

[0026] It's understandable that the central computing end serves as the task scheduling center for the logistics factory, responsible for unified management and distribution of computing tasks generated by multiple edge robots, ensuring that computing tasks are properly assigned to the target edge computing end. Distributing computing tasks to the target edge computing end can fully utilize the computing resources of the edge computing end, achieving optimization of computing efficiency and computing energy consumption. Through this distributed collaboration between the center and the edge, it not only ensures the local rapid response of the edge robots, but also achieves efficient scheduling of global tasks, thereby greatly improving the overall operational efficiency of the logistics factory.

[0027] S2. Obtain the processing task information, total bandwidth, number of robots, and robot transmission power of the edge robot. The processing task information includes: center distance, data size, and maximum tolerable delay.

[0028] It should be explained that processing task information refers to the data of computing tasks generated by edge robots, and total bandwidth refers to the bandwidth of the communication link between the edge robot and the central computing end, usually expressed in bits per second (bps). The number of robots refers to the number of edge robots in the logistics factory, and the robot transmission power refers to the power of the edge robot when transmitting data through wireless communication with the central computing end. The center distance refers to the distance between the edge robot and the central computing end, and the data size refers to the size of the data of the computing tasks generated by the edge robot. For example, the data size is 2mb. The maximum tolerable delay refers to the longest response time allowed by the edge robot from task initiation to completion in the process of completing the task, and the maximum tolerable delay is manually set by the logistics dispatcher based on the task performed by the robot.

[0029] S3. Determine the task difficulty index based on the edge robot, and determine the data transmission rate based on the edge robot, central computing terminal, center distance, total bandwidth, number of robots, and robot transmission power.

[0030] Specifically, determining the task difficulty index based on the edge robot includes: Obtain multiple sets of historical task information of the edge robot, where the historical task information includes: completion time, energy consumption, and failure rate; Extract multiple completion times, multiple energy consumptions, and multiple failure rates from multiple sets of historical task information; Normalizing the multiple completion times, the multiple energy consumptions, and the multiple failure rates respectively to obtain multiple standard times, multiple standard energy consumptions, and multiple standard failure rates; The weights of multiple standard times, multiple standard energy consumptions and multiple standard failure rates are calculated respectively to obtain time weight, energy consumption weight and failure rate weight; Determine an average time based on the multiple standard times, wherein the average time is an average value of the multiple standard times; Determine an average energy consumption based on multiple standard energy consumptions, wherein the average energy consumption is an average value of the multiple standard energy consumptions; Determining an average failure rate based on multiple standard failure rates, wherein the average failure rate is an average of the multiple standard failure rates; The task difficulty index is calculated based on the average time, average energy consumption, average failure rate, time weight, energy consumption weight, and failure rate weight. The calculation formula is as follows: in, represents the task difficulty index, represents the average time, represents the average energy consumption, represents the average failure rate, represents the time weight, represents the energy consumption weight, Represents the failure rate weight.

[0031] It should be explained that historical mission information refers to the data recorded when the edge robot performed tasks in the past, and historical mission information includes: completion time, energy consumption, and failure rate. Completion time refers to the time required for the edge robot to complete each historical mission, energy consumption refers to the energy consumed by the edge robot to complete each historical mission, and failure rate refers to the probability that the edge robot failed to complete the historical mission. The completion time, energy consumption, and failure rate are all obtained from the log of the edge robot's execution of tasks.

[0032] For example, if multiple groups of historical task information are: the first group (1s, 5j, 0.5), the second group (1s, 10j, 0.4), the third group (2s, 5j, 0.3), and the fourth group (1s, 6j, 0.2), then after extracting multiple completion times, multiple energy consumptions, and multiple failure rates from the multiple groups of historical task information, the multiple completion times are (1s, 1s, 2s, 1s), the multiple energy consumptions are (5j, 10j, 5j, 6j), and the multiple failure rates are (0.5, 0.4, 0.3, 0.2).

[0033] It should be explained that standard time refers to the normalized completion time. Standard energy consumption refers to the normalized energy consumption. Standard failure rate refers to the normalized failure rate. The time weight reflects the contribution of average completion time to changes in task difficulty, the energy consumption weight reflects the contribution of average energy consumption to changes in task difficulty, and the failure rate weight reflects the contribution of failure rate to changes in task difficulty. Task difficulty reflects the difficulty level of the edge robot in performing the task.

[0034] Specifically, the normalization operations are performed on the multiple completion times, the multiple energy consumptions, and the multiple failure rates to obtain the multiple standard times, the multiple standard energy consumptions, and the multiple standard failure rates, including: For each of the multiple completion times, perform the following operations: Determine a minimum completion time based on the multiple completion times, wherein the minimum completion time is a minimum value among the multiple completion times; determine a maximum completion time based on the multiple completion times, wherein the maximum completion time is a maximum value among the multiple completion times; The standard time is calculated based on the completion time, minimum completion time, and maximum completion time. The fairness calculation is as follows: in, Indicates standard time. Indicates the completion time, Indicates the minimum completion time, Indicates the maximum completion time; Aggregate standard times to obtain multiple standard times; Determine multiple standard energy consumptions based on multiple energy consumptions; A plurality of standard failure rates are determined based on the plurality of failure rates.

[0035] It should be understood that the method of determining multiple standard energy consumptions based on multiple energy consumptions and the method of determining multiple standard failure rates based on multiple failure rates are the same as the method of normalizing multiple completion times to obtain multiple standard times, and will not be repeated here.

[0036] Specifically, the weight calculations are performed on multiple standard times, multiple standard energy consumptions, and multiple standard failure rates to obtain time weights, energy consumption weights, and failure rate weights, including: The time weight is calculated based on multiple standard times. The calculation formula is as follows: in, represents the time weight, Indicates the number of standard times, Indicates the first of multiple standard times Standard time; Energy consumption weights are determined based on multiple standard energy consumptions, and failure rate weights are determined based on multiple standard failure rates.

[0037] It should be understood that the method of determining energy consumption weights based on multiple standard energy consumptions and the method of determining failure rate weights based on multiple standard failure rates are the same as the method of calculating weights for multiple standard times to obtain time weights, and will not be repeated here.

[0038] Specifically, determining the data transmission rate based on the edge robot, the central computing terminal, the center distance, the total bandwidth, the number of robots, and the robot transmission power includes: Noise is collected from the edge robot and the central computing end to obtain edge noise signals, wherein the collection frequency and collection time of the noise are preset; Performing a fast Fourier transform on the edge noise signal to obtain an amplitude spectrum, wherein the amplitude spectrum includes a plurality of frequency points and a plurality of amplitude values, wherein the frequency points correspond to the amplitude values ​​one to one; The average amplitude value is determined based on the amplitude spectrum. The calculation formula is as follows: in, represents the average amplitude value, Indicates the number of frequency points in multiple frequency points, Indicates the first of multiple frequency points The amplitude value of the frequency point, Indicates taking the absolute value; The noise power spectral density is calculated based on the acquisition frequency and average amplitude value. The calculation formula is as follows: in, represents the noise power spectral density, Indicates the acquisition frequency; The data transmission rate is calculated based on the center distance, total bandwidth, number of robots, robot transmission power, and noise power spectrum density. The calculation formula is as follows: in, Indicates the data transmission rate, Indicates the total bandwidth, Indicates the number of robots, Indicates the robot's transmission power, represents the center distance, is the preset noise index, is the natural logarithm.

[0039] It should be explained that the noise collection of the edge robot and the central computing end refers to: collecting the transmission signal in the wireless channel between the edge robot and the central computing end, processing the collected transmission signal through a filter, filtering out the useful signal components in the transmission signal, and then obtaining the remaining signal, which is the edge noise signal. Among them, the specific waveform characteristics of the useful signal are manually set by the logistics dispatcher of the logistics factory based on the characteristics of the signal waveform recorded during the data transmission between the edge robot and the central computing end in history. And the technology of processing the collected transmission signal through a filter and filtering out the useful signal components in the transmission signal is a prior art and will not be repeated here.

[0040] For example, since any periodic signal (or nearly periodic signal) can be viewed as a superposition of multiple sine waves or cosine waves of different frequencies, a periodic signal can be decomposed into multiple frequency components. By performing a fast Fourier transform on the edge noise signal, a complex number sequence is obtained, wherein the complex number sequence includes: multiple complex numbers, and each complex number in the complex number sequence corresponds to a frequency component of multiple different frequencies decomposed from the edge noise signal. The modulus of each of the multiple complex numbers is then calculated in sequence to obtain multiple modulus lengths. Since the complex numbers correspond to the frequency components one-to-one, the modulus length reflects the vibration amplitude of the edge noise signal at the frequency component corresponding to the modulus length. Therefore, a coordinate system is constructed with the modulus length as the vertical axis and the frequency as the horizontal axis. The multiple modulus lengths and their corresponding multiple frequency components are mapped to the coordinate system to obtain the amplitude spectrum. The multiple frequency components in the amplitude spectrum are the multiple frequency points, and the multiple modulus lengths in the amplitude spectrum are the multiple amplitude values. The above-mentioned fast Fourier transform and amplitude spectrum generation steps are both prior art and will not be repeated here.

[0041] It is understandable that the average amplitude value refers to the arithmetic mean of the amplitude values ​​of all frequency points in the amplitude spectrum, reflecting the strength of the overall edge noise signal. The sampling frequency refers to the time interval for collecting edge noise signals in the wireless channel between the edge robot and the central computing end. The noise power spectrum density refers to the power distribution of noise within a bandwidth of 1 Hz, reflecting the intensity of noise at different frequencies. The larger the noise power spectrum density, the stronger the noise interference within the bandwidth of 1 Hz, and the more serious the impact on signal transmission. The data transmission rate reflects the speed at which the data of the computing task generated by the edge robot is transmitted from the edge robot to the central computing end. The higher the data transmission rate, the faster the data of the computing task generated by the edge robot is transmitted from the edge robot to the central computing end.

[0042] Optionally, the acquisition frequency is 20 Hz and the acquisition time is 10 s.

[0043] It should be understood that the noise index is a value manually set by the logistics dispatcher of the logistics factory. Optionally, the noise index is 3.

[0044] S4. Perform the following operations on each of the multiple edge computing ends: determine the edge pressure index based on the edge computing end, obtain edge resource data and edge bandwidth, where the edge resource data includes: task load, CPU frequency, load capacitance, operating voltage and clock frequency.

[0045] It should be explained that edge resource data refers to data related to the operating status of the edge computing end, including: task load, CPU frequency, load capacitance, operating voltage and clock frequency. Edge bandwidth refers to the bandwidth of the communication link between the edge computing end and the central computing end. Task load refers to the computational amount of computing tasks that the edge computing end needs to process. CPU frequency refers to the number of oscillation cycles of the processor of the edge computing end per unit time. Load capacitance is used to measure the amount of charge required for the chip to switch within a unit clock cycle. Operating voltage refers to the power supply voltage of the CPU of the edge computing end during normal operation. Clock frequency refers to the frequency of the system clock signal of the CPU of the edge computing end.

[0046] Specifically, determining the edge pressure index based on the edge computing end includes: Obtain multiple historical queue information from the edge computing end, where the historical queue information includes: queue number and queue duration; Input multiple queue quantities from multiple historical queue information into the pre-built ARIMA model to obtain the future queue quantity; Determine an average queue time based on multiple queue times, where the average queue time is the average of the multiple queue times; The marginal pressure index is determined based on the number of future queues and the average queue time, wherein the marginal pressure index is the product of the number of future queues and the average queue time.

[0047] It should be noted that historical queue information refers to the task queue status data recorded by the edge computing end during operation, which can be obtained from the edge computing end's task processing log. The queue count refers to the number of tasks waiting to be processed in the edge computing end's task queue during the historical sampling period, and the queue duration refers to the waiting time for each task from entering the queue to the start of processing.

[0048] For example, a logistics factory contains multiple edge robots, each of which generates computing tasks. These tasks are then sent to the edge computing terminals via the central computing terminal for computation. Since the number of edge computing terminals is limited and it takes time to complete computational tasks, when the number of computational tasks exceeds the number of edge computing terminals, task queuing occurs. When queued tasks are queued at an edge computing terminal, the edge computing terminal must complete processing the current computational task before processing the queued tasks. The completion time of the queued tasks is also affected by the queueing time.

[0049] It is understandable that the main operating principle of the ARIMA model is: using the autoregressive (AR), difference (I) and moving average (MA) modules in the ARIMA model to perform time series modeling, stabilization and residual correction on multiple queue numbers to obtain the prediction results of the future queue numbers, and then use the difference module to eliminate the non-stationarity of the prediction results and obtain a stable result sequence, and finally use the sliding average module to perform white noise correction on the model residuals of the stable result sequence to obtain the future queue number, and the above processes are all disclosed technical solutions, and the embodiments of the present invention will not be repeated here. The future queue number refers to the number of queued tasks at the edge computing end within a certain time period in the future. The edge pressure index reflects the current pressure situation of the edge computing end. The larger the edge pressure index, the greater the current computing pressure of the edge computing end.

[0050] S5. Determine the total delay based on data size, data transmission rate, center distance, data transmission rate and edge bandwidth, and determine the total energy consumption based on data size, data transmission rate, robot transmission power and edge resource data.

[0051] Specifically, determining the total delay based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth includes: The first delay is calculated based on the data size, data transmission rate and center distance. The calculation formula is as follows: in, represents the first delay, Indicates the data size, is the preset speed of light; The second delay is calculated based on the data size and edge bandwidth. The calculation formula is as follows: in, represents the second delay, represents the edge bandwidth; A total delay is determined based on the first delay and the second delay, wherein the total delay is the sum of the first delay and the second delay.

[0052] It should be explained that the first delay reflects the time it takes for the data of the computing task generated by the edge robot to be transmitted from the edge robot to the central computing end. The greater the first delay, the longer it takes for the data of the computing task generated by the edge robot to be transmitted from the edge robot to the central computing end. The second delay reflects the time it takes for the data of the computing task generated by the edge robot to be transmitted from the central computing end to the edge computing end. The greater the second delay, the longer it takes for the data of the computing task generated by the edge robot to be transmitted from the central computing end to the edge computing end. The total delay reflects the total time required to complete the computing task generated by the robot. The greater the total delay, the longer the total time required to complete the computing task. The speed of light is .

[0053] Specifically, determining the total energy consumption based on data size, data transmission rate, robot transmission power, and edge resource data includes: The first energy consumption is calculated based on the data size, data transmission rate and robot transmission power. The calculation formula is as follows: in, Indicates the first energy consumption; The second energy consumption is calculated based on the task load, CPU frequency, load capacitance, operating voltage, and clock frequency in the edge resource data. The calculation formula is as follows: in, represents the second energy consumption, represents the load capacitance, Indicates the operating voltage, Indicates the clock frequency, Indicates the CPU frequency, Indicates task load; A total energy consumption is determined based on the first energy consumption and the second energy consumption, wherein the total energy consumption is the sum of the first energy consumption and the second energy consumption.

[0054] It should be explained that the first energy consumption reflects the amount of energy consumed when the data of the computing tasks generated by the edge robot is transmitted from the edge robot to the central computing end. The greater the first energy consumption, the more energy is consumed when the data of the computing tasks generated by the edge robot is transmitted from the edge robot to the central computing end. The second energy consumption reflects the amount of energy consumed when the data of the computing tasks generated by the edge robot is transmitted from the central computing end to the edge computing end. The greater the second energy consumption, the more energy is consumed when the data of the computing tasks generated by the edge robot is transmitted from the central computing end to the edge computing end. The total energy consumption reflects the total energy required to complete the computing tasks generated by the edge robot. The greater the total energy consumption, the greater the total energy required to complete the computing tasks.

[0055] S6. Determine the delay constraint index based on the maximum tolerable delay and total delay, and input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain the scheduling score.

[0056] Specifically, determining the delay constraint index based on the maximum tolerable delay and the total delay includes: The delay constraint index is calculated based on the maximum tolerable delay and the total delay. The calculation formula is as follows: in, represents the delay constraint index, represents the total delay, Indicates the maximum tolerable delay.

[0057] It should be explained that the delay constraint index is used to measure whether the current delay of the computing task meets the task requirements. If the total delay is less than or equal to the maximum tolerable delay, it means that the computing task is completed within the maximum tolerable delay range, and the value is assigned to 1, indicating that the task meets the delay requirements. If the total delay is greater than the maximum tolerable delay, the task times out and the value is assigned to 0, indicating that the task does not meet the delay requirements.

[0058] In detail, the AI ​​scheduling model is as follows: in, represents the AI ​​scheduling model, represents the task difficulty index, is the edge pressure index, represents the delay constraint index, represents the total energy consumption, represents the total delay, is an exponential function, is the preset energy consumption delay ratio, is a natural constant.

[0059] It should be explained that the scheduling score is the value calculated by inputting the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the AI ​​scheduling model. The scheduling score is used to quantify the suitability of the edge robot's computing tasks to be assigned to the edge computing end. The smaller the scheduling score, the more suitable the edge robot's computing tasks are to be assigned to the edge computing end. The energy consumption delay ratio is a value manually set by the logistics dispatcher of the logistics factory based on historical data. Optionally, the average energy consumption and average delay of the edge computing end in processing computing tasks in history are obtained, and the ratio of the average energy consumption to the sum of the average energy consumption and the average delay is used as the energy consumption delay ratio.

[0060] S7. Summarize the scheduling scores to obtain multiple scheduling scores, where the scheduling scores correspond one-to-one to the edge computing end. Based on the multiple scheduling scores, the target edge computing end is identified, and resource scheduling is completed based on the target edge computing end and the edge robot.

[0061] Specifically, the target edge computing end is determined based on multiple scheduling scores, including: Perform zero-removal screening on multiple scheduling scores to obtain a zero-removed scheduling score set; Determine the optimal scheduling score based on the zero-removal scheduling score set, wherein the optimal scheduling score is the minimum zero-removal scheduling score in the zero-removal scheduling score set; The edge computing end corresponding to the best scheduling score is used as the target edge computing end.

[0062] It should be explained that the zero-removing screening of multiple scheduling scores refers to removing the scheduling scores with a value of 0 from the multiple scheduling scores. For example, if the multiple scheduling scores have the values ​​of 1, 2, 3, 0, 2, and 1, then after zero-removing screening, the values ​​obtained are 1, 2, 3, 2, and 1, which is the zero-removed scheduling score set. The zero-removed scheduling score set refers to the set of scheduling scores after zero-removing screening. The target edge computing end refers to the edge computing end corresponding to the optimal scheduling score.

[0063] The present invention is to solve the problems described in the background technology. The present invention receives resource scheduling instructions and identifies edge robots, central computing terminals and multiple edge computing terminals based on the resource scheduling instructions. The edge robot and the central computing terminal have the function of data transmission, and the central computing terminal and each of the multiple edge computing terminals have the function of data transmission. It can be seen that the embodiment of the present invention confirms the edge robot, the central computing terminal and multiple edge computing terminals in advance, which facilitates the subsequent use of the central computing terminal to allocate the computing tasks generated by the edge robot to the edge computing terminal for computing and processing, thereby improving the intelligence of resource scheduling, and then obtaining the processing task information, total bandwidth, number of robots and robot transmission power of the edge robot. The processing task information includes: center distance, data size and maximum tolerable delay. It can be seen that the embodiment of the present invention obtains the processing task information, total bandwidth, number of robots and robot transmission power of the edge robot, which facilitates the subsequent analysis of the computing tasks generated by the edge robot, thereby calculating the data transmission rate, and confirming the task difficulty index based on the edge robot. It can be seen that the embodiment of the present invention analyzes the computing tasks generated by the edge robot and obtains the task difficulty index, which facilitates the subsequent automatic edge computing resource scheduling based on the task difficulty index, thereby improving the intelligence level of resource scheduling, and confirms the data transmission rate based on the edge robot, the center computing end, the center distance, the total bandwidth, the number of robots and the robot transmission power. Taking into account that data may be delayed during transmission, the data transmission rate is calculated to facilitate the subsequent edge computing resource scheduling based on the data transmission rate, improve the intelligence of resource scheduling, and perform the following operations on each edge computing end in multiple edge computing ends: Based on the edge computing end, the edge pressure index is confirmed. It can be seen that the embodiment of the present invention obtains the historical operation data of each edge computing end, thereby calculating the current edge pressure index of each edge computing end, obtaining edge resource data and edge bandwidth, wherein the edge resource data includes: task load, CPU frequency, load capacitance, operating voltage and clock frequency. It can be seen that the embodiment of the present invention obtains the real-time operation data of the edge computing end, which facilitates the subsequent calculation of total delay and total energy consumption. The total delay is determined based on the data size, data transmission rate, center distance, data transmission rate and edge bandwidth. The total energy consumption is determined based on the data size, data transmission rate, robot transmission power and edge resource data. It can be seen that the embodiment of the present invention comprehensively considers the computing tasks generated by the edge robot and the operating status of the edge computing terminal, and comprehensively calculates the total delay and total energy consumption required to complete the computing tasks, so as to facilitate the subsequent intelligent scheduling of edge computing resources according to the total delay and total energy consumption, so as to reduce delay and energy consumption and improve the task processing efficiency of the edge robot. The delay constraint index is determined based on the maximum tolerable delay and the total delay. It can be seen that the embodiment of the present invention determines the delay constraint index by comparing the maximum tolerable delay with the total delay, so as to facilitate the subsequent calculation of the scheduling score.Improve the task processing efficiency of the edge robot, input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain a scheduling score. It can be seen that the embodiment of the present invention uses the pre-built AI scheduling model to analyze the acquired data, thereby intelligently deriving a scheduling score, improving the intelligence level of resource scheduling, summarizing the scheduling scores, and obtaining multiple scheduling scores, wherein the scheduling scores correspond one-to-one to the edge computing end. It can be seen that the embodiment of the present invention calculates the scheduling score for each edge computing end and summarizes the scheduling scores to facilitate the subsequent confirmation of the best scheduling score, confirms the target edge computing end based on multiple scheduling scores, and completes resource scheduling based on the target edge computing end and the edge robot. It can be seen that the embodiment of the present invention confirms the best edge computing end through the best scheduling score, thereby distributing computing tasks to the best edge computing end, and then intelligently realizing edge computing resource scheduling. Therefore, the present invention can improve the intelligence level of resource scheduling and improve the task processing efficiency of the edge robot.

[0064] like Figure 2 , which is a functional module diagram of an AI-driven 6G edge computing resource scheduling system provided by one embodiment of the present invention.

[0065] The AI-driven 6G edge computing resource scheduling system 100 of the present invention can be installed in an electronic device. Depending on the functions implemented, the AI-driven 6G edge computing resource scheduling system 100 may include a basic device confirmation module 101, a task information processing module 102, an edge information processing module 103, and an intelligent scoring scheduling module 104. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can perform fixed functions, which are stored in the memory of the electronic device.

[0066] The basic equipment confirmation module 101 is used to receive resource scheduling instructions and confirm the edge robot, the central computing end and multiple edge computing ends based on the resource scheduling instructions, wherein the edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; The task information processing module 102 is used to obtain processing task information, total bandwidth, number of robots, and robot transmission power of the edge robot. The processing task information includes: center distance, data size, and maximum tolerable delay. The task difficulty index is determined based on the edge robot, and the data transmission rate is determined based on the edge robot, the center computing end, the center distance, the total bandwidth, the number of robots, and the robot transmission power. The edge information processing module 103 is configured to perform the following operations on each of the multiple edge computing terminals: determining an edge pressure index based on the edge computing terminal, obtaining edge resource data and edge bandwidth, wherein the edge resource data includes: task load, CPU frequency, load capacitance, operating voltage, and clock frequency; determining a total delay based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth; determining a total energy consumption based on data size, data transmission rate, robot transmission power, and edge resource data; and determining a delay constraint index based on a maximum tolerable delay and total delay; The intelligent scoring scheduling module 104 is used to input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into a pre-built AI scheduling model to obtain a scheduling score, summarize the scheduling scores, and obtain multiple scheduling scores, wherein the scheduling scores correspond one-to-one to the edge computing end, and the target edge computing end is identified based on the multiple scheduling scores, and resource scheduling is completed based on the target edge computing end and the edge robot.

[0067] In detail, the modules in the AI-driven 6G edge computing resource scheduling system 100 in the embodiment of the present invention adopt the same Figure 1 The AI-driven 6G edge computing resource scheduling method described in

[15] is the same technical means and can produce the same technical effects, so I will not go into details here.

[0068] like Figure 3 , which is a structural diagram of an electronic device for implementing an AI-driven 6G edge computing resource scheduling method provided by one embodiment of the present invention.

[0069] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an AI-driven 6G edge computing resource scheduling method program.

[0070] The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard drive, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard drive of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard drive, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc. equipped on the electronic device 1. Furthermore, the memory 11 includes both the internal storage unit of the electronic device 1 and an external storage device. The memory 11 can be used not only to store application software installed on the electronic device 1 and various types of data, such as the code of the AI-driven 6G edge computing resource scheduling method program, but also to temporarily store data that has been output or is about to be output.

[0071] In some embodiments, the processor 10 may be composed of an integrated circuit, such as a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor 10 is the control core (Control Unit) of the electronic device, connecting the various components of the entire electronic device using various interfaces and lines, and executing programs or modules stored in the memory 11 (such as an AI-driven 6G edge computing resource scheduling method program) and calling data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0072] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0073] Figure 3 Only the electronic device with components is shown, and it can be understood by those skilled in the art that Figure 3The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0074] For example, although not shown, the electronic device 1 may further include a power supply (e.g., a battery) to power various components. Preferably, the power supply may be logically connected to the at least one processor 10 via a power management device, thereby enabling functions such as charge management, discharge management, and power consumption management via the power management device. The power supply may further include any components such as one or more DC or AC power supplies, a recharging device, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which are not further described here.

[0075] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0076] Optionally, the electronic device 1 may further include a user interface, which may be a display or an input unit (such as a keyboard). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a display screen or a display unit, and is used to display information processed by the electronic device 1 and to display a visual user interface.

[0077] The AI-driven 6G edge computing resource scheduling method program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve: Receive a resource scheduling instruction, and identify an edge robot, a central computing end, and multiple edge computing ends based on the resource scheduling instruction, wherein the edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; Obtain edge robot processing task information, total bandwidth, number of robots, and robot transmission power. Processing task information includes: center distance, data size, and maximum tolerable delay; Determine the task difficulty index based on the edge robot; Determine the data transmission rate based on the edge robot, central computing terminal, center distance, total bandwidth, number of robots, and robot transmission power; Perform the following operations on each of the multiple edge computing terminals: Determine the edge pressure index based on the edge computing end; Obtain edge resource data and edge bandwidth, where edge resource data includes: task load, CPU frequency, load capacitance, operating voltage, and clock frequency; Determine the total latency based on data size, data rate, center distance, data rate, and edge bandwidth; Determine the total energy consumption based on data size, data transmission rate, robot transmission power and edge resource data; Determine the delay constraint index based on the maximum tolerable delay and the total delay; Input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain the scheduling score; Summarize the scheduling scores to obtain multiple scheduling scores, where the scheduling scores correspond one-to-one to the edge computing terminals; The target edge computing end is identified based on multiple scheduling scores, and resource scheduling is completed based on the target edge computing end and edge robot.

[0078] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0079] Furthermore, if the modules / units integrated into the electronic device 1 are implemented as software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. The computer-readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0080] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed by a processor of an electronic device, the computer program can implement: Receive a resource scheduling instruction, and identify an edge robot, a central computing end, and multiple edge computing ends based on the resource scheduling instruction, wherein the edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; Obtain edge robot processing task information, total bandwidth, number of robots, and robot transmission power. Processing task information includes: center distance, data size, and maximum tolerable delay; Determine the task difficulty index based on the edge robot; Determine the data transmission rate based on the edge robot, central computing terminal, center distance, total bandwidth, number of robots, and robot transmission power; Perform the following operations on each of the multiple edge computing terminals: Determine the edge pressure index based on the edge computing end; Obtain edge resource data and edge bandwidth, where edge resource data includes: task load, CPU frequency, load capacitance, operating voltage, and clock frequency; Determine the total latency based on data size, data rate, center distance, data rate, and edge bandwidth; Determine the total energy consumption based on data size, data transmission rate, robot transmission power and edge resource data; Determine the delay constraint index based on the maximum tolerable delay and the total delay; Input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain the scheduling score; Summarize the scheduling scores to obtain multiple scheduling scores, where the scheduling scores correspond one-to-one to the edge computing terminals; The target edge computing end is identified based on multiple scheduling scores, and resource scheduling is completed based on the target edge computing end and edge robot.

[0081] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only exemplary, and actual implementations may have other division methods.

[0082] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0083] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0084] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An AI-driven 6G edge computing resource scheduling method, characterized in that: The method comprises: Receive a resource scheduling instruction, and identify an edge robot, a central computing end, and multiple edge computing ends based on the resource scheduling instruction, wherein the edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; Obtain edge robot processing task information, total bandwidth, number of robots, and robot transmission power. Processing task information includes: center distance, data size, and maximum tolerable delay; Determine the task difficulty index based on the edge robot; Determine the data transmission rate based on the edge robot, central computing terminal, center distance, total bandwidth, number of robots, and robot transmission power; Perform the following operations on each of the multiple edge computing terminals: Determine the edge pressure index based on the edge computing end; Obtain edge resource data and edge bandwidth, where edge resource data includes: task load, CPU frequency, load capacitance, operating voltage, and clock frequency; Determine the total latency based on data size, data rate, center distance, data rate, and edge bandwidth; Determine the total energy consumption based on data size, data transmission rate, robot transmission power and edge resource data; Determine the delay constraint index based on the maximum tolerable delay and the total delay; Input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain the scheduling score; Summarize the scheduling scores to obtain multiple scheduling scores, where the scheduling scores correspond one-to-one to the edge computing terminals; The target edge computing end is identified based on multiple scheduling scores, and resource scheduling is completed based on the target edge computing end and edge robot.

2. The AI-driven 6G edge computing resource scheduling method according to claim 1, characterized in that: The determination of the task difficulty index based on the edge robot includes: Obtain multiple sets of historical task information of the edge robot, where the historical task information includes: completion time, energy consumption, and failure rate; Extract multiple completion times, multiple energy consumptions, and multiple failure rates from multiple sets of historical task information; Normalizing the multiple completion times, the multiple energy consumptions, and the multiple failure rates respectively to obtain multiple standard times, multiple standard energy consumptions, and multiple standard failure rates; The weights of multiple standard times, multiple standard energy consumptions and multiple standard failure rates are calculated respectively to obtain time weight, energy consumption weight and failure rate weight; Determine an average time based on the multiple standard times, wherein the average time is an average value of the multiple standard times; Determine an average energy consumption based on multiple standard energy consumptions, wherein the average energy consumption is an average value of the multiple standard energy consumptions; Determining an average failure rate based on multiple standard failure rates, wherein the average failure rate is an average of the multiple standard failure rates; The task difficulty index is calculated based on the average time, average energy consumption, average failure rate, time weight, energy consumption weight and failure rate weight.

3. The AI-driven 6G edge computing resource scheduling method according to claim 2, characterized in that: The data transmission rate is determined based on the edge robot, the central computing terminal, the center distance, the total bandwidth, the number of robots, and the robot transmission power, including: Noise is collected from the edge robot and the central computing end to obtain edge noise signals, wherein the collection frequency and collection time of the noise are preset; Performing a fast Fourier transform on the edge noise signal to obtain an amplitude spectrum, wherein the amplitude spectrum includes a plurality of frequency points and a plurality of amplitude values, wherein the frequency points correspond to the amplitude values ​​one to one; Determine the average amplitude value based on the amplitude spectrum; Calculate the noise power spectral density based on the acquisition frequency and average amplitude value; The data transmission rate is calculated based on the center distance, total bandwidth, number of robots, robot transmission power, and noise power spectrum density.

4. The AI-driven 6G edge computing resource scheduling method according to claim 3, characterized in that: The determining of the edge pressure index based on the edge computing end includes: Obtain multiple historical queue information from the edge computing end, where the historical queue information includes: queue number and queue duration; Input multiple queue quantities from multiple historical queue information into the pre-built ARIMA model to obtain the future queue quantity; Determine an average queue time based on multiple queue times, where the average queue time is the average of the multiple queue times; The marginal pressure index is determined based on the number of future queues and the average queue time, wherein the marginal pressure index is the product of the number of future queues and the average queue time.

5. The AI-driven 6G edge computing resource scheduling method according to claim 4, characterized in that: The total latency is determined based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth, including: Calculating a first delay based on data size, data transmission rate, and center distance; Calculate the second delay according to the data size and edge bandwidth; A total delay is determined based on the first delay and the second delay, wherein the total delay is the sum of the first delay and the second delay.

6. The AI-driven 6G edge computing resource scheduling method according to claim 5, characterized in that: The total energy consumption is determined based on data size, data transmission rate, robot transmission power and edge resource data, including: Calculate the first energy consumption based on the data size, data transmission rate and robot transmission power; Calculating a second energy consumption based on the task load, CPU frequency, load capacitance, operating voltage, and clock frequency in the edge resource data; A total energy consumption is determined based on the first energy consumption and the second energy consumption, wherein the total energy consumption is the sum of the first energy consumption and the second energy consumption.

7. The AI-driven 6G edge computing resource scheduling method according to claim 6, characterized in that: Determining the delay constraint index based on the maximum tolerable delay and the total delay includes: The delay constraint index is calculated based on the maximum tolerable delay and the total delay. The calculation formula is as follows: in, represents the delay constraint index, represents the total delay, Indicates the maximum tolerable delay.

8. The AI-driven 6G edge computing resource scheduling method according to claim 7, characterized in that: The AI ​​scheduling model is as follows: in, represents the AI ​​scheduling model, represents the task difficulty index, is the edge pressure index, represents the delay constraint index, represents the total energy consumption, represents the total delay, is an exponential function, is the preset energy consumption delay ratio, is a natural constant.

9. The AI-driven 6G edge computing resource scheduling method according to claim 8, characterized in that: The determining of the target edge computing end based on multiple scheduling scores includes: Perform zero-removal screening on multiple scheduling scores to obtain a zero-removed scheduling score set; Determine the optimal scheduling score based on the zero-removal scheduling score set, wherein the optimal scheduling score is the minimum zero-removal scheduling score in the zero-removal scheduling score set; The edge computing end corresponding to the best scheduling score is used as the target edge computing end.

10. An AI-driven 6G edge computing resource scheduling system, characterized in that: The system comprises: A basic equipment confirmation module is used to receive resource scheduling instructions and confirm the edge robot, the central computing end, and multiple edge computing ends based on the resource scheduling instructions. The edge robot and the central computing end have a data transmission function, and the central computing end and each of the multiple edge computing ends have a data transmission function; The task information processing module is used to obtain the processing task information, total bandwidth, number of robots, and robot transmission power of the edge robot. The processing task information includes: center distance, data size, and maximum tolerable delay. The task difficulty index is determined based on the edge robot, and the data transmission rate is determined based on the edge robot, center computing end, center distance, total bandwidth, number of robots, and robot transmission power. an edge information processing module, configured to perform the following operations on each of the plurality of edge computing terminals: determining an edge pressure index based on the edge computing terminal, obtaining edge resource data and edge bandwidth, wherein the edge resource data includes task load, CPU frequency, load capacitance, operating voltage, and clock frequency; determining a total delay based on data size, data transmission rate, center distance, data transmission rate, and edge bandwidth; determining a total energy consumption based on data size, data transmission rate, robot transmission power, and edge resource data; and determining a delay constraint index based on a maximum tolerable delay and total delay; The intelligent scoring scheduling module is used to input the task difficulty index, edge pressure index, total delay, delay constraint index and total energy consumption into the pre-built AI scheduling model to obtain a scheduling score, summarize the scheduling scores, and obtain multiple scheduling scores. Among them, the scheduling score corresponds one-to-one to the edge computing end. Based on multiple scheduling scores, the target edge computing end is confirmed, and resource scheduling is completed based on the target edge computing end and the edge robot.

Images