CN116634466A - Task offloading and resource allocation method based on unmanned aerial vehicle cooperative multi-access edge computing - Google Patents

Abstract

The method comprises the steps of constructing an optimization model with the aim of minimizing service delay and guaranteeing service fairness among user equipment according to the requirements of each user equipment in terms of service experience, simplifying a problem model based on a Dinkelbach method and a convex optimization theory, providing a four-stage alternate iterative optimization algorithm, decomposing the optimization target into four sub-optimization targets of an unmanned plane track decision, a task unloading decision, a service cache decision and a resource allocation decision, and utilizing an alternate solving iterative algorithm to carry out solving calculation until the target converges to obtain the task unloading and resource allocation decision in the unmanned plane cooperative multi-access edge calculation network. The present disclosure enables lower service delays while guaranteeing better fairness among all user devices.

Description

Task offloading and resource allocation method based on cooperative multi-access edge computing of unmanned aerial vehicles

Technical Field

The present disclosure relates to the field of mobile communication technology, and in particular to a method for unloading tasks and allocating resources based on unmanned aerial vehicle (UAV) collaborative multi-access edge computing.

Background Art

The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

In recent years, with the development and popularization of mobile communication technology, many new applications such as online video, map navigation, mobile payment, and face recognition have emerged. Subsequently, the surge in networked smart devices has led to an explosive growth of data. At the same time, in emergencies such as various infectious diseases, face-to-face communication between people has become difficult, and the reliance on online medical care, online learning, and remote work has increased significantly. The above applications are usually delay-sensitive and require a lot of communication and computing resources. In order to support a large number of smart devices and process a large amount of data in a timely manner, multi-access edge computing, formerly known as mobile edge computing, has become a key technology in the next generation of wireless networks. By deploying mobile edge computing servers on the edge side of the communication network (such as ground cellular infrastructure), user devices can offload data to the edge to improve the service experience.

However, there are also many problems with current edge computing systems. Ground mobile edge computing servers at fixed locations cannot be adjusted according to the needs of terminals. Due to non-line-of-sight links, their channel quality may be poor, resulting in limited communication rates. And, some user devices may abandon mobile edge computing services due to severe obstructions or damage caused by natural disasters. Recently, drones have become a promising technology to improve wireless connectivity and provide wide coverage in mobile edge computing networks due to their flexible deployment and low-cost advantages. Generally, there are two technologies for drone-assisted mobile edge computing networks, in which drones act as air relays and air mobile edge computing servers. In addition, with the rapid growth of user devices, a single or even multiple drones may not be able to meet the needs of a large number of computationally intensive and delay-sensitive applications such as virtual reality and smart transportation, and cannot solve the service delay and fairness problems, that is, it cannot maximize the service experience ratio of user devices.

Summary of the invention

In order to solve the above problems, the present disclosure proposes a method for unloading tasks and allocating resources based on drone-coordinated multi-access edge computing, which utilizes drones to collaborate on computing and caching resources for mobile edge computing services, minimizes service delays, and ensures service fairness among user devices.

According to some embodiments, the present disclosure adopts the following technical solutions:

The method for unloading tasks and allocating resources based on UAV collaborative multi-access edge computing includes:

Initialize the collaborative computing task offloading environment, the base station and all drones collaborate to provide mobile edge computing services for user devices, obtain the task set, and divide the task cycle into multiple time slots with equal duration;

The input data size of the computing task requested by the user in each time slot is obtained. According to the service experience requirements of each user device, an optimization model is constructed with the goal of minimizing service delay while ensuring service fairness among user devices. Based on the Dinkelbach method and convex optimization theory, the problem model is simplified, and a four-stage alternating iterative optimization algorithm is proposed. The optimization objective is decomposed into four sub-optimization objectives: UAV trajectory decision, task offloading decision, service cache decision, and resource allocation decision. The alternating solution iterative algorithm is used to solve the calculation until the objective converges, and the task offloading and resource allocation decisions in the UAV collaborative multi-access edge computing network are obtained and executed.

Furthermore, the satisfaction-based task offloading decision optimization includes: fixing UAV trajectories, bandwidth resource allocation decisions, service cache decisions, computing resource allocation decisions and auxiliary variables to optimize task offloading decisions, and defining the optimization objective formula of the task offloading subproblem.

In the optimization process of task offloading decision, a set of several UAVs and tasks is defined. Each user device sends a task offloading request to its associated UAV at the beginning of the time slot, and a suitable offloading location is selected for the task based on the user's satisfaction. Under the current task offloading decision, the value of the objective function corresponding to each user device is calculated. If the maximum delay tolerance of all user devices is met and the computing resources and energy consumption limits of each UAV are not exceeded, then the task offloading decision at this time is appropriate.

In the set, each task has different satisfaction with different unloading locations. The satisfaction value is related to the task processing delay and fairness. The greater the task processing delay and the lower the fairness, the smaller the satisfaction value.

Furthermore, the optimization of service cache decision is to fix the trajectory of UAV, bandwidth resource allocation decision, task offloading decision, computing resource allocation decision and auxiliary variables to optimize the service cache decision, and define the optimization formula of service cache decision sub-problem.

Considering the cache space utilization of the drone, the services required by tasks with high service cache decision values are cached with high priority on the drone until the upper limit of the drone's cache space is reached.

Furthermore, the optimization of the UAV trajectory includes fixed task offloading decisions, bandwidth resource allocation decisions, service cache decisions, computing resource allocation decisions and auxiliary variables to optimize the UAV trajectory and define the formula of the UAV trajectory optimization sub-problem.

In the optimization of the UAV trajectory, trajectory planning is used as an optimization variable, which consists of the coordinate position of the UAV in each time slot during the entire mission cycle.

Furthermore, the optimization of computing resource allocation decisions includes defining an optimization formula for a computing resource allocation subproblem given a UAV trajectory, a task offloading decision, a service cache decision, and auxiliary variables. The computing resource allocation subproblem is a convex problem, and convex optimization is used to obtain the optimal solution for bandwidth resource allocation and computing resource allocation.

Furthermore, task offloading, service caching, trajectory planning and resource allocation are jointly optimized to maximize the service experience ratio. A four-stage alternating iterative optimization is proposed to solve the original problem. Task offloading decisions, service caching decisions and UAV trajectory planning are iteratively optimized respectively until the target value converges. After each round of the four-stage liberalization iteration, the parameters of the Dinkelbach method are updated.

Compared with the prior art, the beneficial effects of the present invention are:

The present disclosure proposes a method for unloading and allocating resources based on cooperative multi-access edge computing tasks of drones, whereby drones can effectively utilize computing and cache resources to collaborate on mobile edge computing services, aiming to minimize service delays while ensuring service fairness among user devices.

In order to improve the service experience, the present invention considers the joint optimization of task offloading, resource allocation, trajectory planning and service cache placement under the constraints of drone energy budget and delay requirements, and expresses it as a service experience ratio maximization problem. Since the original problem is a mixed integer non-convex programming problem with a fractional structure, it is difficult to solve in polynomial time. Based on the Dinkelbach method and convex optimization theory, the present invention simplifies the problem model and proposes a four-stage alternating iterative service ratio maximization algorithm to solve the problem. Numerical results show that compared with other benchmark algorithms, the algorithm proposed in the present invention can reduce service delay by 78.2% while improving fairness among all user devices by 53.0%.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

FIG1 is a schematic diagram of a multi-UAV assisted mobile edge computing scenario according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram showing a decomposition of an optimization method according to an embodiment of the present disclosure.

DETAILED DESCRIPTION

The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and/or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and/or combinations thereof.

Example 1

In one embodiment of the present disclosure, a method for unloading tasks and allocating resources based on drone-coordinated multi-access edge computing is provided, including:

Step 1: Initialize the collaborative computing task offloading environment. The base station and all drones collaborate to provide mobile edge computing services for user devices, obtain the task set, and divide the task cycle into multiple time slots with equal duration;

Step 2: Obtain the input data size of the computing task of the user request service in each time slot, and build an optimization model based on the service experience requirements of each user device to minimize service delay while ensuring service fairness among user devices;

Step 3: Based on the Dinkelbach method and convex optimization theory, the problem model is simplified, and a four-stage alternating iterative optimization algorithm is proposed. The optimization objective is decomposed into four sub-optimization objectives: UAV trajectory decision, task offloading decision, service cache decision, and resource allocation decision. The alternating iterative algorithm is used to solve the calculation until the objective converges, and the task offloading and resource allocation decisions in the UAV collaborative multi-access edge computing network are obtained and executed.

As an embodiment, the present disclosure is a method for task offloading and resource allocation in a caching drone collaborative multi-access edge computing network based on service experience, which solves the problems of service delay and fairness, that is, maximizes the service experience ratio of user equipment. In order to solve the above technical purpose, the implementation process specifically includes the following:

Step 1: For the UAV-supported mobile edge computing network, a multi-UAV-assisted mobile edge computing problem is proposed to maximize the service experience ratio of user devices.

Initialize the collaborative computing task offloading environment, the base station and all drones collaborate to provide mobile edge computing services for user devices, obtain the task set, and divide the task cycle into multiple time slots with equal duration;

In which, the base station and all drones collaborate to provide mobile edge computing services for M user devices. One macro base station, U drones and M user devices are represented as b, Y = {1, 2, ..., U}, M = {1, 2, ..., M} respectively. The set of all services that the macro base station can provide is represented as ∑ = {1, 2, ..., S}. Because the connection between the user device and the drone can remain stable in a sufficiently short period of time. For ease of representation, the task cycle N is divided into T time slots with equal duration _Δt , and T = {1, 2, ..., T}. Each user device has only one delay-sensitive task in a time slot, which can be offloaded to a drone or macro base station for processing, and each task is atomic and indivisible. In time slot t, a delay-sensitive task generated by user m requesting service s It can be represented by a 3-tuple

Assumptions represents the input data size of the computational task of user m requesting service s in time slot t. Let Represents the computational intensity of the computational task of user m requesting service s in time slot t. is the processing delay tolerance of the task requesting service s. Beyond this limit, the result is invalid for user m, and each user device has different requirements in terms of service experience; the proposed goal is to maximize the service experience ratio of the user device. This can be achieved by jointly optimizing task offloading, service caching, drone trajectory and resource allocation. The details are as follows:

Among them, _C1 indicates that in order to offload the task of user m requesting service s to drone u, service s needs to be cached in drone u. _C2 indicates the spectrum resource allocation constraint of user equipment associated with the same drone. _C3 indicates the computing resource constraint of a single drone. _C4 indicates that the total storage space occupied by the services stored on each drone shall not exceed the total storage space of the drone, which is represented by K _u . _C5 indicates that the ground user should be within the coverage of the associated drone. _C6 indicates the position change constraint of the drone between any two time slots. _C7 indicates that the minimum safety distance should be maintained between any two drones to ensure that they avoid collision within time slot t. _C8 indicates that the flight trajectories of all drones should be within the target area. In order to avoid high communication delay of drones, _C9 indicates that the horizontal distance between the associated drone and the drone selected as a relay shall not exceed R _uav . _C10 indicates the energy upper limit E _th of each drone within time slot t. _C11 indicates that the completion delay of each task cannot exceed the processing delay tolerance of the task. _C12 indicates that the task generated by each user equipment is allowed to be accurately offloaded to a nearby drone or macro base station. Constraint C ₁₃ and C ₁₄ indicate that the service cache decision and task offloading decision variables are binary, respectively, while constraint C ₁₅ indicates that the bandwidth and computing resource allocation variables are continuous.

Step 2: Obtain the input data size of the computing task of the user request service in each time slot. According to the service experience requirements of each user device, an optimization model is constructed with the goal of minimizing service delay while ensuring service fairness among user devices. Based on the Dinkelbach method and convex optimization theory, the problem model is simplified, and a four-stage alternating iterative optimization algorithm is proposed. The optimization goal is decomposed into four sub-optimization goals: drone trajectory decision, task offloading decision, service cache decision, and resource allocation decision. The alternating solution iterative algorithm is used to solve the calculation until the goal converges, and the task offloading and resource allocation decisions in the drone collaborative multi-access edge computing network are obtained and executed.

Specifically, in order to maximize the service experience ratio, the optimization objective is decomposed into four sub-optimization objectives: drone trajectory, task offloading decision, service cache decision, and resource allocation decision. After each round of these four liberalization iterations, the parameters of the Dinkelbach method are updated. In order to decouple this non-convex objective, it is decomposed into different sub-objectives, and an alternating solution iterative algorithm is proposed, which includes the following process:

S1: First, we optimize the task offloading decision based on satisfaction, fix the drone trajectory Q, bandwidth resource allocation decision B, service cache decision A, computing resource allocation decision F and auxiliary variable η to optimize the task offloading decision. The task offloading optimization sub-problem is formulated as:

stC ₁ , C ₃ , C ₅ , C ₉ -C ₁₂ , C ₁₄

In order to better describe the optimization process of task offloading, several sets of drones and tasks are defined. Let the set of drones that cache the service s required by task m be defined as Each user equipment sends a task offloading request to its associated UAV at the beginning of the time slot, and the set of task offloading requests received by the associated UAV u is expressed as It contains the tasks that are offloaded by the associated user equipment and the cooperative drone. If the associated drone u belongs to the set Then the drone hits the service s required by task m. Then, the hit task is added to Add the unhit ones to When the task offloading decision is initialized, it is assumed that the set All tasks in can be performed by drone u. The tasks in are further offloaded to collections The cooperative drone i or base station with the largest Π _3-1 value. The tasks in are further offloaded to collections The collaborative drone or base station in the network.

Select the appropriate offloading location for the task based on user satisfaction. Under the current task offloading decision, calculate the value of the Π _3-1 objective function corresponding to each user device. If the maximum delay tolerance of all user devices is met and the computing resources and energy consumption limits of each drone are not exceeded, then the task offloading decision at this time is appropriate. Then, calculate the satisfaction Π _3-1 of each task and select the set in turn. The task with the smallest satisfaction value is further unloaded. Then it is removed from Move to Until the collection All tasks in the task satisfy the maximum delay tolerance and CSS resource and energy consumption constraints.

In the collection In , each task has different satisfaction with different unloading locations. The satisfaction value is related to task processing delay and fairness. The larger the task processing delay and the lower the fairness, the smaller the satisfaction value. Then, the task m rejected by the associated UAV u and needs to be further unloaded has a satisfaction value for the cooperative UAV i, which can be expressed as:

The task m of requesting service s that is rejected by the associated UAV u has a satisfaction ratio for the macro base station b, which can be expressed as:

In the task m of the associated drone's request service s, the unloading request is sent to the location with high satisfaction first. If the requested location is a macro base station, the unloading request will be accepted directly. If the requested location is a cooperative drone, then the cooperative drone needs to allow it. If the unloading request is rejected, then it will be sent to the next best unloading location in the next iteration until it is accepted, allowing Repeat the above process until all tasks are found.

S2: Optimization of service cache decision The service cache decision is optimized by fixing the trajectory of the UAV, bandwidth resource allocation decision, task offloading decision, computing resource allocation decision and auxiliary variables, and the optimization formula of the service cache decision sub-problem is defined.

Specifically, the fixed drone trajectory Q, bandwidth resource allocation decision B, task offloading decision X, computing resource allocation decision F and auxiliary variable η are used to optimize the service cache decision A. The service cache decision subproblem is formulated as:

stC ₁ , C ₄ , C ₁₀ , C ₁₁ , C ₁₃

Since the drone has limited cache space, it is not possible to cache all programs. The service cache decision will be optimized to minimize task processing delay and ensure fairness. In order to improve the utilization of cache space, it is considered that the services required by tasks with higher Π _3-2 values will have a higher priority in caching on the drone until the drone's cache space limit is reached. and |M _u | respectively denote the set and number of tasks offloaded to drone u. Accordingly, let ∑ _u and |∑ _u | respectively denote the set and number of services required by tasks offloaded to drone u. In general, since multiple tasks may request the same service, |∑ _u |＜|M _u |. The service s required by task m is cached on drone u with a priority value, which can be expressed as:

Tasks requesting the same service are grouped into Sort the values in descending order, and then The task with the largest value is stored in the set ∑ _u . in Then the elements in the set ∑ _u are sorted by Sort the values in descending order. The services with larger values are cached in turn until the upper limit of the cache space of drone u is reached. We further set ∑′ _u = {s ₁ , s ₂ , ..., s _J-1 } where

S3: The optimization of UAV trajectory includes fixed task offloading decision, bandwidth resource allocation decision, service cache decision, computing resource allocation decision and auxiliary variables to optimize the UAV trajectory and define the formula of the UAV trajectory optimization sub-problem.

Specifically, the fixed task offloading decision X, bandwidth resource allocation decision B, service cache decision A, computing resource allocation decision F and auxiliary variable η are used to optimize the trajectory Q of the drone. The drone trajectory subproblem is formulated as:

stC ₅ , C ₈ -C ₁₁

The trajectory planning of the UAV is an optimization variable consisting of the coordinate position of the UAV in each time slot during the entire mission cycle. can be simplified to:

Note that due to the existence of It can be seen that problem P _3-3 is non-convex. The left-hand side of constraints C ₁₀ and C ₁₁ is non-convex with respect to the flight trajectory Q of the drone. Constraint C ₇ is non-convex because the domain of a convex function is a non-empty convex set. Therefore, solving non-convex problems is challenging.

Next, in order to deal with the non-convexity problem, the successive convex approximation method is used to achieve the local optimal solution of problem P _3-3 . The key idea of the successive convex approximation method is to approximate the non-convex function into a convex function in an iterative manner.

definition is the available spectrum efficiency from user equipment m to drone u, which can be written as:

It is not difficult to see About is a convex function. Therefore, it can be obtained by having The first-order Taylor expansion of achieves the global lower bound. Given the k-th iteration of the UAV flight trajectory The lower bound of can be calculated as:

in and are the available spectrum efficiency from user equipment m to drone u at the kth iteration and about The first-order derivatives of are given as follows:

definition is the available spectrum efficiency from UAV u to UAV i, which can be written as:

About is a convex function. Therefore, it can be obtained by having The first-order Taylor expansion of achieves the global lower bound. Given the k-th iteration of the UAV flight trajectory and The lower bound of can be calculated as

in and are the available spectrum efficiency from UAV u to UAV i at the kth iteration and about The first-order derivatives of are given as follows:

definition is the available spectrum efficiency from UAV u to macro base station b, which can be written as:

About is a convex function. Therefore, it can be obtained by having The first-order Taylor expansion of achieves the global lower bound. Given the k-th iteration of the UAV flight trajectory The lower bound of can be calculated as:

in and are the available spectrum efficiency from UAV u to macro base station b at the kth iteration and about The first-order derivatives of are given as follows:

In constraint C ₇ , because The flight trajectory of the UAV is convex, and we use the successive convex approximation method to relax the constraints. and Applying the first-order Taylor expansion, we get the following inequality:

therefore, The lower bound of can be calculated as:

In addition, for the flight power in constraint C ₁₀ The first and third items are about speed A convex function of . A continuous slack variable is introduced To deal with the second term in the propulsion power formula, it becomes:

Simplifying the above formula we can get:

Given the flight speed of the drone at the kth iteration and The right side of the above inequality is approximated by applying a first-order Taylor expansion as follows:

Then, you can pass The upper bound of is approximated as:

Based on the above discussion, all non-convexities in problem Π _3-3 are solved, and the original problem in the kth iteration can be restated as the following approximate problem P′ _3-3 (k).

stC ₅ , C ₆ , C ₈ , C ₉

After proving the convexity of the problem, the optimal solution for the UAV trajectory planning can be effectively obtained through convex optimization tools. It is worth noting that the optimal solution obtained from the approximate problem Π′ _3-3 is a lower bound of the problem Π _3-3 .

S4: The optimization of computing resource allocation decision includes defining the optimization formula of the computing resource allocation subproblem given the UAV trajectory, task offloading decision, service cache decision and auxiliary variables. The computing resource allocation subproblem is a convex problem, and convex optimization is used to obtain the optimal solution for bandwidth resource allocation and computing resource allocation.

Specifically, given the UAV trajectory Q, the task offloading decision X, the service cache decision A and the auxiliary variable η, the computational resource allocation subproblem is formulated as:

stC ₂ , C ₃ , C ₁₀ , C ₁₁ , C ₁₅

Because problem Π _3-4 is a convex problem, convex optimization tools are used to obtain the optimal solution for bandwidth resource allocation and computing resource allocation.

This disclosure proposes an alternating optimization to solve the original problem P ₁ . The key idea is to iteratively optimize task offloading decisions, service cache decisions, and drone trajectory planning respectively until the target value converges.

The present disclosure is described with reference to the flowchart and/or block diagram of the method, device (system), and computer program product according to the embodiment of the present disclosure. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the process and/or box in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Technical personnel in the relevant field should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims (10)

1. A method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing, characterized in that it includes: Initialize the collaborative computing task offloading environment, the base station and all UAVs cooperate to provide mobile edge computing services for user equipment, obtain task sets, and divide the task cycle into multiple time slots with equal duration; Obtain the input data size of the computing task requested by the user in each time slot, and build an optimization model with the goal of minimizing service delay and ensuring service fairness between user devices according to the service experience requirements of each user device. Based on the Dinkelbach method and convex optimization theory, the problem model is simplified, and a four-stage alternate iterative optimization algorithm is proposed. The optimization objective is decomposed into four sub-optimization objectives: UAV trajectory decision, task offload decision, service cache decision, and resource allocation decision. The iterative algorithm for solving solves and calculates until the target converges, and obtains and executes the task offloading and resource allocation decisions in the cooperative multi-access edge computing network of UAVs. 2. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 1, wherein the optimization of task unloading decision based on satisfaction degree includes: fixed UAV trajectory, bandwidth resource allocation decision , service caching decision, computing resource allocation decision and auxiliary variables to optimize the task offloading decision, and define the optimization objective formulation of the task offloading subproblem. 3. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 2, wherein, in the optimization process of task unloading decision-making, a set of several unmanned aerial vehicles and tasks are defined, each A user device sends a task offloading request to its associated UAV at the beginning of the time slot, and selects an appropriate offloading location for the task based on user satisfaction, and calculates the objective function corresponding to each user device under the current task offloading decision. If the maximum delay tolerance of all user devices is satisfied, and the computing resources and energy consumption constraints of each UAV are not exceeded, then the task offloading decision at this time is appropriate. 4. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 3, wherein, in the set, each task has different satisfaction levels for different unloading positions, and the satisfaction level is The value of is related to task processing delay and fairness, the greater the task processing delay and the lower the fairness, the smaller the value of satisfaction. 5. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 1, wherein the optimization of the service cache decision is the trajectory of the fixed unmanned aerial vehicle, bandwidth resource allocation decision, Task offloading decision, computing resource allocation decision and auxiliary variables are used to optimize the service caching decision, and the optimization formula of the service caching decision subproblem is defined. 6. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 5, characterized in that, considering the utilization rate of the cache space of the unmanned aerial vehicle, the task required for the high cache decision value of the service The priority of the service cache on the drone is high until the upper limit of the cache space of the drone is reached. 7. The method of task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 1, the optimization of the trajectory of the unmanned aerial vehicle includes a fixed task unloading decision, a bandwidth resource allocation decision, a service cache decision, and a computing resource allocation Decision and auxiliary variables are used to optimize the UAV trajectory, and the formulation of the UAV trajectory optimization subproblem is defined. 8. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 7, in the optimization of the trajectory of the unmanned aerial vehicle, the trajectory planning is used as an optimization variable, and the unmanned aerial vehicle is used in the entire The coordinate position of each time slot in the task cycle. 9. The method for task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 1, the optimization of computing resource allocation decision includes given unmanned aerial vehicle trajectory, task unloading decision, service cache decision and auxiliary variable , defining an optimization formula for the sub-problem of computing resource allocation, the sub-problem of computing resource allocation is a convex problem, and the optimal solution of bandwidth resource allocation and computing resource allocation is obtained by using convex optimization. 10. The method of task offloading and resource allocation based on unmanned aerial vehicle cooperative multi-access edge computing as claimed in claim 1, which jointly optimizes task offloading, service caching, trajectory planning and resource allocation, and maximizes the service experience ratio, proposing a Four stages of alternate iterative optimization to solve the original problem, iteratively optimize task offloading decision, service cache decision and UAV trajectory planning respectively, until the target value converges, after each round of the four stages of liberalization iteration, the parameters of the Dinkelbach method to update.