CN115550357A - Multi-agent multi-task cooperative unloading method - Google Patents

Abstract

The application discloses a multi-agent multi-task cooperative unloading method, which comprises the following steps: setting a task and a service node thereof; then determining the node type, and then representing the tasks generated by any task vehicle by using triples; then, local computation and multi-service node edge unloading computation are selected for task unloading; on the basis of selecting edge unloading calculation, judging the idle state of a service node of a task vehicle TaV, wherein the service node meets the requirement of serving as the relay node; then, based on the result judgment of the idle state, executing the unloading task in a single-hop unloading mode; and if the idle state does not meet the single-hop unloading condition, unloading the calculation unloading task into an idle service node of the relay node for execution in two hops. According to the technical scheme, in a dynamic vehicle-mounted environment, the problem of multi-vehicle multi-task cooperative unloading conflict can be solved by using limited computing resources of task vehicles and service resources of surrounding spaces, and minimum system average time delay is realized.

Description

A multi-agent multi-task cooperative unloading method

technical field

The present application relates to the technical field of vehicle networking communication, and more specifically, relates to a multi-agent and multi-task cooperative unloading method.

Background technique

The Internet of Vehicles is a typical Industrial IoT technology in which ubiquitous information can be exchanged and shared between vehicles without human intervention. In the Internet of Vehicles environment, driving vehicles generate massive amounts of sensor data per second. In order to have an intelligent vision in a complex driving environment, it is necessary to complete a large number of data transmission, storage, and processing operations in a relatively short period of time. The multi-task scene of the real 3D winding mountain road is a road section with a high incidence of traffic accidents. Due to the limitation of terrain conditions and occlusion environment, the vehicle speed is slow, and there are blind spots in the RSU signal coverage. Therefore, task collaborative unloading is particularly important.

At present, there are many deficiencies in the existing research on the cooperative offloading scheme of vehicle networking tasks. In the unloading cooperative unloading model, most of the cooperative unloading modes are executed by a single-process unloading model. Such an unloading mode will bring higher latency and lower channel resource utilization; then in the task splitting method , the task types of collaborative unloading are currently mainly indivisible and equal splitting. Such a task splitting scheme often cannot fully utilize the computing resources on the service vehicle of the host vehicle; secondly, in terms of vehicle movement status, the current unloading research During the process, the vehicles are relatively in a static state, but for the real 3D road scene, the vehicle is in a moving state during the collaborative unloading process; and more critically: in dealing with conflicts, when there is a multi-task multi-host vehicle cooperative unloading scene In this case, multiple host vehicles will compete for a collaborative node, resulting in communication conflicts, reducing the efficiency of task offloading, increasing the delay of task offloading, and failing to provide users with good quality of service.

Therefore, how to provide a multi-agent multi-task cooperative unloading method, which can use the limited computing resources of the mission vehicle and the service resources of the surrounding space in a dynamic vehicle environment to solve the problem of multi-vehicle multi-task cooperative unloading conflicts and achieve the smallest system The average delay has become a technical problem to be solved urgently by those skilled in the art.

Contents of the invention

In order to solve the above technical problems, this application provides a multi-agent multi-task cooperative unloading method, which can solve the problem of multi-vehicle multi-task cooperative unloading conflicts by using the limited computing resources of task vehicles and the service resources of the surrounding space in a dynamic vehicle environment , to achieve the minimum average system delay.

The technical scheme that this application provides is as follows:

表示；其中D_i为卸载任务的大小，C_i为任务所需计算资源，Tmax_i为执行任务的最大可容限时延；S4、卸载方式选择：在定义节点状态后，选择本地计算和多服务节点边缘卸载计算进行任务卸载，其中所述边缘卸载计算包括单跳卸载与两跳卸载；S5、执行卸载任务：在选择边缘卸载计算基础上，判断任务车辆TaV中v_i的服务节点v_j的空闲状态，且所述服务节点v_j满足充当所述中继节点的要求；随后基于空闲状态的结果判断，通过所述单跳卸载方式执行卸载任务；若所述空闲状态不满足单跳卸载条件，则将计算卸载任务通过所述两跳卸载到所述中继节点v_j的空闲服务节点v_k中执行。The application provides a multi-agent and multi-task cooperative unloading method, comprising the following steps: S1, setting the vehicle task unloading scene: identifying the vehicle that generates the task as the task vehicle TaV; within the maximum tolerable time delay of the task vehicle, it is always in Within the communication range, configure a service node that provides unloading services for the task vehicle; S2, determine the node type: identify a service node that is subordinate to multiple task vehicles within the TaV communication range as a candidate service node; when the candidate service node is determined to be After the service node of vi in the task vehicle _TaV , the current candidate service node can act as a candidate relay node for the remaining task vehicles; then select the relay node that can perform two-hop offloading from the candidate relay nodes, and the relay node The service node of the node; S3, unloading task representation: the task generated by any task vehicle v _i is represented by a triplet:

where D _i is the size of the offloading task, C _i is the computing resource required by the task, and Tmax _i is the maximum tolerable delay in executing the task; S4. Selection of offloading method: after defining the node state, select local computing and multi-service Node edge offloading calculations perform task offloading, wherein the edge offloading calculations include single-hop offloading and two-hop offloading; S5, perform offloading tasks: on the basis of selecting edge offloading calculations, determine the service node vj of v _i in the task vehicle _TaV Idle state, and the service node v _j meets the requirements of acting as the relay node; then judge based on the result of the idle state, perform the offloading task through the single-hop offloading method; if the idle state does not meet the single-hop offloading condition , then the computation offloading task is offloaded to the idle service node v _k of the relay node v _j through the two hops for execution.

Further, in a preferred manner of the present invention, in the step S4, it also includes:

If the local computing is selected, that is, when TaV chooses to use its own local computing resources to process tasks, it will directly obtain the local computing delay

表示TaV的本地计算能力，

表示本地任务的计算量。in,

Indicates the local computing power of the TaV,

Indicates the calculation amount of the local task.

Further, in a preferred manner of the present invention, in the step S4, the definition of the node state is specifically: use the 0-1 variable Δ _{i, j} to represent the communication range of the task vehicle v _i in the single-hop unloading The status of the internal service node v _j :

Use the 0-1 variable β _{j, k} to represent the state of the service node v _k of the relay node v _j when performing two-hop offloading:

Further, in a preferred manner of the present invention, in the step S5, the step of executing the unloading task is specifically:

Judging the idle state of the service node v _j of v _i in the task vehicle TaV:

If the service node v _j of the task vehicle v _i is in an idle state, i.e. α _{i, j} =0; the task is directly unloaded to the service node v _j for calculation by single-hop unloading, and the single-hop calculation delay is obtained;

为任务车辆v_i向服务节点v_j分配的子计算任务；r_i，j为v_i传输计算任务到v_j的上传速率；μ₁为通信过程中的重叠因子，表示能被执行卸载到边缘节点数据比例；

表示服务节点v_j给任务车辆车辆v_i分配的算力；in,

is the subcomputing task assigned by the task vehicle v _i to the service node v _j ; r _i,j is the upload rate of vi _i transmitting the computing task to v _j ; μ ₁ is the overlap factor in the communication process, indicating that it can be executed and offloaded to the edge Node data ratio;

Indicates the computing power allocated by the service node v _j to the task vehicle v _i ;

If the service node v _j of the task vehicle v _i is in the occupied state, i.e. α _{i, j =} 1, it means that the service node v _j has been selected as the service node by the remaining TaV; Under the requirement, the node v _j as a relay node is offloaded to its idle service node v _k through two hops to perform unloaded calculation tasks, that is, β _{j, k} = 0; then obtain the two-hop calculation delay:

表示v_i将计算子任务

上传到中继节点v_j；

表示中继节点v_j将计算任务

两跳卸载到中继节点v_j的服务节点v_k；Vt表示为任务车辆TaV的集合；Vs表示为服务节点的集合。in,

Indicates that v _i will compute subtasks

Upload to relay node v _j ;

Indicates that the relay node v _j will calculate the task

Two hops are offloaded to the service node v _k of the relay node v _j ; Vt is represented as a collection of task vehicles TaV; Vs is represented as a collection of service nodes.

预估t+Δt时刻的位置信息

Further, in a preferred manner of the present invention, in step S2, after completing the determination of the node type, it also includes: constructing a vehicle motion model, and passing the position information of the task vehicle v _α at time t

Estimated position information at time t+Δt

的步骤具体包括：Further, in a preferred mode of the present invention, a vehicle motion model is constructed to estimate position information

The steps specifically include:

与x轴、y轴与z轴之夹角分别为θ_x，θ_y，θ_z∈[0，π]；S1. Obtain the speed of the task vehicle v _α at time t

The included angles with the x-axis, y-axis and z-axis are θ _x , θ _y , θ _z ∈ [0, π];

和夹角获取速度

在x轴、y轴与z轴的速度分量；S2. According to the speed

and the included angle to get the velocity

Velocity components on the x-axis, y-axis and z-axis;

S3. Then obtain the distance components of the task vehicle v _α traveling in the x-axis, y-axis and z-axis directions after the vehicle runs Δt through the direction cosine theorem;

According to the law of direction cosines, cos ² θ _x +cos ² θ _y +cos ² θ _z = 1, that is:

S4. Then use the L2 _norm to calculate the Euclidean distance from the task vehicle v _α to v _β at any time:

which is

Further, in a preferred manner of the present invention, in the step S2 determining the node type, it also includes: clustering the vehicle driving road based on the task vehicle, and there are N+1 edge servers, and the task vehicle Vt There are M, so as to cluster the service nodes of multi-tasking vehicles;

Wherein, the steps of clustering service nodes specifically include:

S1. Randomly select M mission vehicles from the vehicle set V={v ₀ , v ₁ , v ₂ ,..., v _N } as centroids;

S2. Sampling is performed in the service node, and the Euclidean distance from each sample to M centroids is calculated;

进入后续步骤；其中R为设于道路为所有设备提供服务的路边单元RSU的覆盖半径；S3. If the Euclidean distance from the sample to the centroid at any time Δt∈[0～Tmax _i ]

Enter the next step; where R is the coverage radius of the RSU located on the road to provide services for all equipment;

S4. If the sample is within the communication range of multiple centroids, α _{i, j} = 1, then use this sample as the selected service node; otherwise, α _{i, j} = 0, divide this sample into the cluster corresponding to this centroid;

S5. Finally output the clustering result to obtain the service node set Vs of the mission vehicle.

Further, in a preferred mode of the present invention, before the selection of the unloading mode in step S4, it also includes: constructing a communication model, and obtaining the rate of the communication upload link; the steps specifically include:

S1. Set the communication mode between the mission vehicle and the service node: short-range wireless communication; use orthogonal frequencies to reduce mutual influence during communication;

S2. Set the upload link in the communication process as a flat Rayleigh fading channel, regardless of channel interference;

S3. Then, according to the Shannon formula, the average transmission rate of the upload link from the mission vehicle v _i to the service node v _j can be calculated:

代表路径损耗指数；信道衰落系数用h表示。Among them, * can be v2v or v2r, W _v2v and W _v2r represent the channel bandwidth of V2V and V2R respectively; Pi is the transmission power of the mission vehicle v _i , ρ ₀ represents the white Gaussian noise power, and d _{i, j} are v _i to v _j transmission distance;

Represents the path loss index; the channel fading coefficient is represented by h.

Further, in a preferred manner of the present invention, before the task representation is unloaded in step S3, task assignment is also included: determining the unequal task set C assigned to the service node by the task vehicle.

Further, in a preferred manner of the present invention, the specific steps of task assignment include:

S1. Setting an objective function, the objective function aims to solve the calculation time delay problem of multi-vehicle multi-task unloading;

最小化问题，即

越靠近0代表并行程度越高；转化后目标函数为：S2. Transform the objective function into the absolute value of edge node unloading and local calculation parallel difference

The minimization problem is

The closer to 0, the higher the degree of parallelism; the transformed objective function is:

Among them, C represents the computing subtasks that are not equally split; the constraint condition C ₁ represents the constraint boundary of the local and offloaded computing subtasks; C ₂ represents the distance between the task vehicle v _i and the service node v _j within the task maximum tolerance time limit Tmax _i The relative Euclidean distance of is less than or equal to the communication range R of _vi ; C ₃ means that the sum of the local and multi-service node offloaded computing subtasks is equal to the total computing task of the task vehicle _vi

S3. Then use the differential evolution algorithm to assign the offloading task to the service node.

表示；其中D_i为卸载任务的大小，C_i为任务所需计算资源，Tmax_i为执行任务的最大可容限时延；S4、卸载方式选择：在定义节点状态后，选择本地计算和多服务节点边缘卸载计算进行任务卸载，其中所述边缘卸载计算包括单跳卸载与两跳卸载；S5、执行卸载任务：在选择边缘卸载计算基础上，判断任务车辆TaV中v_i的服务节点v_j的空闲状态，且所述服务节点v_j满足充当所述中继节点的要求；随后基于空闲状态的结果判断，通过所述单跳卸载方式执行卸载任务；若所述空闲状态不满足单跳卸载条件，则将计算卸载任务通过所述两跳卸载到所述中继节点v_j的空闲服务节点v_k中执行。利用本申请提供的多智能体多任务协同卸载方法，当出现有限计算资源和用户需要体验之间的矛盾，通过本地执行和多服务节点多跳分布式卸载执行并行计算任务，在保障用户服务需求的前提下，充分利用了任务车辆本身的计算资源和周围空间的服务资源，实现最小系统整体的计算时延。在任务车辆数较大的情况时，采用多跳分布式卸载相比传统的单跳单节点卸载系统整体时延更优。本申请公开的技术方案，其能够利用任务车辆有限计算资源和周围空间的服务资源，解决多车辆多任务协同卸载冲突问题，实现最小的系统平均时延。Compared with the prior art, a multi-agent and multi-task cooperative unloading method provided by the present invention includes the following steps: S1, setting a vehicle task unloading scene: identifying the vehicle that generates the task as a task vehicle TaV; The maximum tolerable time delay is always within the communication range, and configured as a service node that provides unloading services for the task vehicle; S2. Determine the node type: identify service nodes that belong to multiple task vehicles within the communication range of the TaV as candidate services node; when the candidate service node is determined to be the service node of vi in the task vehicle _TaV , the current candidate service node can act as a candidate relay node for the remaining task vehicles; then the intermediate node that can perform two-hop unloading is selected from the candidate relay nodes Relay node, and the service node of the relay node; S3, unloading task representation: the task that any task vehicle v _i produces is triplet:

where D _i is the size of the offloading task, C _i is the computing resource required by the task, and Tmax _i is the maximum tolerable delay in executing the task; S4. Selection of offloading method: after defining the node state, select local computing and multi-service Node edge offloading calculations perform task offloading, wherein the edge offloading calculations include single-hop offloading and two-hop offloading; S5, perform offloading tasks: on the basis of selecting edge offloading calculations, determine the service node vj of v _i in the task vehicle _TaV Idle state, and the service node v _j meets the requirements of acting as the relay node; then judge based on the result of the idle state, perform the offloading task through the single-hop offloading method; if the idle state does not meet the single-hop offloading condition , then the computation offloading task is offloaded to the idle service node v _k of the relay node v _j through the two hops for execution. Using the multi-agent multi-task cooperative unloading method provided by this application, when there is a contradiction between limited computing resources and user needs, parallel computing tasks are executed through local execution and multi-hop distributed offloading of multi-service nodes, ensuring user service needs Under the premise of the system, the computing resources of the task vehicle itself and the service resources of the surrounding space are fully utilized to achieve the minimum overall system computing delay. When the number of mission vehicles is large, the overall delay of the multi-hop distributed offloading system is better than that of the traditional single-hop single-node offloading system. The technical solution disclosed in this application can utilize the limited computing resources of mission vehicles and the service resources of the surrounding space to solve the problem of multi-vehicle multi-tasking cooperative unloading conflicts and achieve the minimum average system delay.

Beneficial effect:

1. The multi-agent multi-task cooperative unloading method provided by this application can adopt multi-hop serial unloading distributed execution unloading strategy for multi-task and multi-node, and solve the problem of multi-vehicle multi-task cooperative unloading conflict;

2. The multi-agent and multi-task cooperative unloading method provided by this application has obvious low-latency characteristics in solving the problem of multi-task optimization unloading, and realizes the prediction of vehicle movement trajectories in three-dimensional road scenes;

3. The multi-agent and multi-task cooperative unloading method provided by this application distributes the unloading tasks of the task vehicles to the service nodes based on the differential evolution algorithm, and has good convergence when solving the high-dimensional nonlinear problem of unequal splitting of tasks. .

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of the steps of the multi-agent multi-task cooperative unloading method provided by the embodiment of the present invention;

Fig. 2 is a framework flowchart of a clustering algorithm solving service node provided by an embodiment of the present invention;

FIG. 3 is a schematic diagram of a first clustering structure of a service node set provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of a second clustering structure of a service node set provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of a third clustering structure of a service node set provided by an embodiment of the present invention;

FIG. 6 is a comparison chart of task offloading calculation delays provided by an embodiment of the present invention.

detailed description

In order to enable those skilled in the art to better understand the technical solution in the application, the technical solution in the embodiment of the application will be clearly and completely described below in conjunction with the drawings in the embodiment of the application. Obviously, the described implementation Examples are only some of the embodiments of the present application, but not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

It should be noted that when an element is referred to as being "fixed" or "disposed on" another element, it may be directly disposed on another element or indirectly disposed on another element; when an element is referred to as being "connected" It may be directly connected to another element or indirectly connected to another element.

It is to be understood that the terms "length", "width", "upper", "lower", "front", "rear", "first", "second", "vertical", "horizontal", " The orientations or positional relationships indicated by "top", "bottom", "inner", "outer", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the application and simplifying the description, rather than indicating or implying the It should not be construed as limiting the application to indicate that a device or element must have a particular orientation, be constructed, and operate in a particular orientation.

In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, a feature defined as "first" and "second" may explicitly or implicitly include one or more of these features. In the description of the present application, the meanings of "plurality" and "several" are two or more, unless otherwise specifically defined.

It should be noted that the structures, proportions, sizes, etc. shown in the drawings of this specification are only used to match the content disclosed in the specification, for those who are familiar with this technology to understand and read, and are not used to limit the conditions that this application can implement , so it has no technical substantive meaning, and any modification of structure, change of proportional relationship or adjustment of size shall still fall within the scope of the disclosure disclosed in this application without affecting the effect and purpose of this application. The technical content must be within the scope covered.

表示；其中D_i为卸载任务的大小，C_i为任务所需计算资源，Tmax_i为执行任务的最大可容限时延；S4、卸载方式选择：在定义节点状态后，选择本地计算和多服务节点边缘卸载计算进行任务卸载，其中所述边缘卸载计算包括单跳卸载与两跳卸载；S5、执行卸载任务：在选择边缘卸载计算基础上，判断任务车辆TaV中v_i的服务节点v_j的空闲状态，且所述服务节点v_j满足充当所述中继节点的要求；随后基于空闲状态的结果判断，通过所述单跳卸载方式执行卸载任务；若所述空闲状态不满足单跳卸载条件，则将计算卸载任务通过所述两跳卸载到所述中继节点v_j的空闲服务节点v_k中执行。本发明提供一种多智能体多任务协同卸载方法，其能够利用任务车辆有限计算资源和周围空间的服务资源，解决多车辆多任务协同卸载冲突问题，实现最小的系统平均时延。A multi-agent and multi-task cooperative unloading method provided by the present invention includes the following steps: S1, setting a vehicle task unloading scene: identifying the vehicle that generates the task as a task vehicle TaV; always within the maximum tolerable time delay of the task vehicle Within the communication range, configure a service node that provides unloading services for the task vehicle; S2, determine the node type: identify a service node that is subordinate to a plurality of task vehicles within the TaV communication range as a candidate service node; when the candidate service node is determined After serving as the service node of vi in the task vehicle _TaV , the current candidate service node can act as the candidate relay node of the remaining task vehicles; then select the relay node that can perform two-hop offloading from the candidate relay nodes, and the The service node of the successor node; S3, unloading task representation: the task generated by any task vehicle v _i is represented by a triplet:

where D _i is the size of the offloading task, C _i is the computing resource required by the task, and Tmax _i is the maximum tolerable delay in executing the task; S4. Selection of offloading method: after defining the node state, select local computing and multi-service Node edge offloading calculations perform task offloading, wherein the edge offloading calculations include single-hop offloading and two-hop offloading; S5, perform offloading tasks: on the basis of selecting edge offloading calculations, determine the service node vj of v _i in the task vehicle _TaV Idle state, and the service node v _j meets the requirements of acting as the relay node; then judge based on the result of the idle state, perform the offloading task through the single-hop offloading method; if the idle state does not meet the single-hop offloading condition , then the computation offloading task is offloaded to the idle service node v _k of the relay node v _j through the two hops for execution. The invention provides a multi-agent and multi-task cooperative unloading method, which can solve the problem of multi-vehicle multi-task cooperative unloading conflicts by utilizing the limited computing resources of task vehicles and service resources in the surrounding space, and realize the minimum average system delay.

As shown in FIG. 1 to FIG. 6 , the steps and flow of a multi-agent and multi-task cooperative offloading method disclosed in the present application will be described in detail below in conjunction with specific embodiments.

This application takes Enshi, Hubei as a specific example, and proposes a multi-agent multi-task cooperative unloading method, which specifically includes the following steps:

S1. Setting the vehicle task offloading scenario: identify the vehicle that generates the task as the task vehicle TaV; the task vehicle is always within the communication range within the maximum tolerable time delay, and is configured as a service node that provides offloading services for the task vehicle.

TaV通信范围内且可提供服务的车辆和RSU，统称为服务节点(ServiceNodes，SN)，用集合

表示。集合

标示道路所有车辆的实时轨迹信息，第α辆车在第t时刻的轨迹信息为

其中

为速度，

为第α辆车的位置坐标。Among them, in the embodiment of this application, the ten-turn road is used as the unloading scene; firstly, a Road Side Unit (Road Side Unit, RSU) with a coverage radius of R and which can provide services for all devices at the same time is deployed on the road. The MEC server is deployed on the RSU, marked v ₀ . Use the set V={v ₁ , v ₂ ,...,v _N } to represent N vehicle-mounted terminals with a Poisson distribution within the coverage of the RSU, and each vehicle-mounted terminal is equipped with a Beidou satellite navigation system (BDS), etc. Positioning equipment, through which the trajectory information of the vehicle can be obtained in real time; the vehicle that generates the task is called a task vehicle (Task Vehicle, TaV), expressed as a set

Vehicles and RSUs that are within the communication range of TaV and can provide services are collectively referred to as service nodes (ServiceNodes, SN).

express. gather

The real-time trajectory information of all vehicles on the marked road, the trajectory information of the αth vehicle at the time t is

in

for speed,

is the position coordinate of the αth car.

S2. Determine the node type: identify the service nodes that belong to the communication range of multiple task vehicles TaV as candidate service nodes; when the candidate service node is determined to be the service node of vi in the task vehicle _TaV , the current candidate service node can act as Candidate relay nodes for the remaining task vehicles; then select relay nodes capable of performing two-hop offloading from the candidate relay nodes, and the service nodes of the relay nodes.

当候选服务节点已确定为TaV v_i的服务节点后，当前候选服务节点可充当剩余任务车辆(Vs-{v_i})的候选中继节点(Candidate relay Nodes，CrN)，用集合

随后通过从CrN中进一步筛选出可执行两跳卸载的中继节点(Relay Nodes，RN)，用集合

中继节点的服务节点(Relay service Nodes，RsN)，用集合

Among them, when the SN belongs to the communication range of multiple task vehicles TaV, the intersection vehicles are collectively referred to as candidate service nodes (Candidate service Nodes, CsN).

When the candidate service node has been determined as the service node of TaV v _i , the current candidate service node can act as the candidate relay node (Candidate relay Nodes, CrN) of the remaining mission vehicles (Vs-{v _i }), using the set

Then, by further screening out the relay nodes (Relay Nodes, RN) that can perform two-hop offloading from CrN, use the set

The service node of the relay node (Relay service Nodes, RsN), uses the collection

预估t+Δt时刻的位置信息

Specifically, in an embodiment of the present invention, the multi-agent multi-task cooperative unloading method further includes: after completing the determination of the node type, it also includes: constructing a vehicle motion model, through the task vehicle v _α at time t location information

Estimated position information at time t+Δt

的步骤具体包括：S1、获取任务车辆v_α在t时刻速度

与x轴、y轴与z轴之夹角分别为θ_x，θ_y，θ_z∈[0，π]；Specifically, in the embodiment of the present invention, a vehicle motion model is constructed to estimate position information

The steps specifically include: S1. Obtain the speed of task vehicle v _α at time t

The included angles with the x-axis, y-axis and z-axis are θ _x , θ _y , θ _z ∈ [0, π];

和夹角获取速度

在x轴、y轴与z轴的速度分量；S2. According to the speed

and the included angle to get the velocity

Velocity components on the x-axis, y-axis and z-axis;

S3. Then obtain the distance components of the task vehicle v _α traveling in the x-axis, y-axis and z-axis directions after the vehicle runs Δt through the direction cosine theorem;

According to the law of direction cosines, cos ² θ _x +cos ² θ _y +cos ² θ _z = 1,

which is:

S4. Then use the L2 _norm to calculate the Euclidean distance from the task vehicle v _α to v _β at any time:

表示；其中D_i为卸载任务的大小，C_i为任务所需计算资源，Tmax_i为执行任务的最大可容限时延。S3. Unloading task representation: the task generated by any task vehicle v _i is represented by a triplet:

where D _i is the size of the unloaded task, C _i is the computing resource required by the task, and Tmax _i is the maximum tolerable time delay for executing the task.

Specifically, in the embodiment of the present invention, the multi-agent multi-task cooperative unloading method further includes: constructing a communication model, and obtaining the rate of the communication upload link; the steps specifically include:

S1. Set the communication mode between the mission vehicle and the service node: short-range wireless communication; use orthogonal frequencies to reduce mutual influence during communication;

S2. Set the upload link in the communication process as a flat Rayleigh fading channel, regardless of channel interference;

S3. Then, according to the Shannon formula, the average transmission rate of the upload link from the mission vehicle v _i to the service node v _j can be calculated:

代表路径损耗指数；信道衰落系数用h表示。Among them, * can be v2v or v2r, W _v2v and W _v2r represent the channel bandwidth of V2V and V2R respectively; P _i is the transmission power of task vehicle v _i , ρ ₀ represents the white Gaussian noise power, and d _{i, j} are v _i to v The transmission distance of _j ;

Represents the path loss index; the channel fading coefficient is represented by h.

In the embodiment, the communication between the vehicle and the vehicle and the RSU adopts the IEEE802.11 protocol in the short-range wireless communication mode. The use of orthogonal frequencies in the communication process can reduce the mutual influence in the communication process. The upload link in the communication process is set as a flat Rayleigh fading channel, regardless of channel interference.

S4. Selection of offloading mode: After defining the node state, select local computing and multi-service node edge offloading computing to offload tasks, wherein the edge offloading computing includes single-hop offloading and two-hop offloading.

In the embodiment of this application, there are four ways to process tasks: local processing, single-hop V2I (Vehicle to Infrastructure) way to offload to the MEC server of RUS, and single-hop V2V (Vehicle to Vehicle) way to offload to adjacent idle service vehicles and two-hop methods to offload to the idle service vehicle of the relay node; the next three offloading methods are edge offloading calculations.

Specifically, in an embodiment of the present invention, the definition of the node state is specifically: using the 0-1 variable αi _,j to represent the state of the service node v _j within the communication range of the task vehicle v _i in the single-hop unloading:

Use the 0-1 variable β _{j, k} to represent the state of the service node v _k of the relay node v _j when performing two-hop offloading:

S5. Execute unloading task: on the basis of selecting the edge offloading calculation, judge the idle state of the service node v _j of vi in the task vehicle _{TaV, and the service node v j meets} the requirements of acting as the relay node; then based on the idle state Judging the result of the state, the unloading task is executed through the single-hop offloading method; if the idle state does not meet the single-hop offloading condition, the calculation offloading task is offloaded to the idle service of the relay node v _j through the two-hop Executed in node v _k .

时刻内始终处于TaV通信范围内的SN。可以通过本地计算和边缘卸载计算2种方式处理任务。In this embodiment, in order to make full use of the computing resources of the vehicle, the task vehicle TaV can not only process tasks locally, but also use

The SN is always within the communication range of TaV at all times. Tasks can be processed in two ways: local computing and edge offloading computing.

1) Use local computing to offload computing:

When TaV chooses to use its own local computing resources to process tasks, the local computing delay is:

表示本地任务的计算量。Among them, the local computing power of TaV is

Indicates the calculation amount of the local task.

2) Offloading computing at the edge

When TaV executes service node offload calculation, it includes the following stages: task upload, task calculation and result feedback. Since the downlink transmission speed of the task in the result feedback stage is much higher than the uplink speed, the time delay of the result feedback is ignored in this embodiment. Assuming that u _i of TaV chooses the offloaded service node set V _i , the time delay of different stages is analyzed separately:

Case 1: when the service node v _j of the task vehicle v _i is in an idle state, that is, α _i,j =0. In this case, offloading the task to the service node can directly calculate the task:

为任务车辆v_i向服务节点v_j分配的子计算任务；r_i，j为v_i传输计算任务到v_j的上传速率；μ₁为通信过程中的重叠因子，表示能被执行卸载到边缘节点数据比例；

表示服务节点v_j给任务车辆车辆v_i分配的算力。in,

is the subcomputing task assigned by the task vehicle v _i to the service node v _j ; r _i,j is the upload rate of vi _i transmitting the computing task to v _j ; μ ₁ is the overlap factor in the communication process, indicating that it can be executed and offloaded to the edge Node data ratio;

Indicates the computing power allocated by the service node v _j to the task vehicle v _i .

则v_j车辆可作为中继节点通过两跳卸载方式执行卸载计算任务，即β_j，k＝0。Case 2: when the service node v _j of the task vehicle v _i is in the occupied state, that is, α _i,j =1. The reason for this is that the service node v _j has been selected as a service node by other TaVs, and in this case v _j can only be the candidate relay node CrN of the mission vehicle v _i . Suppose v _k is always within the communication range of v _j and

Then vehicle v _j can be used as a relay node to perform unloaded computing tasks in a two-hop offloading manner, that is, β _j,k =0.

表示v_i将计算子任务

上传到中继节点v_j；

表示中继节点v_j将计算任务

两跳卸载到中继节点v_j的服务节点v_k。in,

Indicates that v _i will compute subtasks

Upload to relay node v _j ;

Indicates that the relay node v _j will calculate the task

Two-hop offload to service node v _k of relay node v _j .

Specifically, in the embodiment of the present invention, the total expression of multi-task offloading is:

Based on this, when there is a contradiction between limited computing resources and user experience, parallel computing tasks are performed through local execution and multi-hop distributed offloading of multi-service nodes. Compared with traditional single-hop single-node offloading systems, multi-hop distributed offloading Overall latency is better. The design objective function is as follows:

Among them, C represents the computing subtasks that are split differently by different Vs, Vrn and Vrs.

Constraint condition C ₁ represents the constraint boundary of local and offloaded computing subtasks; C ₂ represents the communication range R where the relative Euclidean distance between task vehicle v _i and service node v _j within the task maximum tolerance time limit Tmax _i is less than or equal to v _i ; C ₃ means that the sum of local and multi-service node offloading computing subtasks is equal to the total computing tasks of task vehicle v _i

In step S2, in view of the fact that the position of the vehicle is constantly changing dynamically, the network topology is also constantly changing, and the communication range of the task vehicle is also limited, not all vehicles on the three-dimensional road can be used as service nodes of the task vehicle. Therefore, in this application, a clustering algorithm is used to determine service node sets, relay nodes, and two-hop service nodes of different mission vehicles.

Specifically, in the real-time of the present invention, in the step S2 determining the node type, it also includes: clustering the vehicle driving road based on the task vehicle, with N+1 edge servers, and M task vehicles Vt , so as to cluster the service nodes of multi-tasking vehicles;

Wherein, the steps of clustering service nodes specifically include:

S1. Randomly select M mission vehicles from the vehicle set V={v ₀ , v ₁ , v ₂ ,..., v _N } as centroids;

S2. Sampling is performed in the service node, and the Euclidean distance from each sample to M centroids is calculated;

进入后续步骤；其中R为设于道路为所有设备提供服务的路边单元RSU的覆盖半径；S3. If the Euclidean distance from the sample to the centroid at any time Δt∈[0～Tmax _i ]

Enter the next step; where R is the coverage radius of the RSU located on the road to provide services for all equipment;

S4. If the sample is within the communication range of multiple centroids, α _{i, j} = 1, then use this sample as the selected service node; otherwise, α _{i, j} = 0, divide this sample into the cluster corresponding to this centroid;

S5. Finally output the clustering result to obtain the service node set Vs of the mission vehicle.

即任务车辆与服务节点的距离越小表示二者的相似性越大，反之则相似性越小。Among them, in order to verify the quality of the degree of clustering, the similarity measure is chosen as the reciprocal of the Euclidean distance

That is, the smaller the distance between the task vehicle and the service node, the greater the similarity between the two, and vice versa.

The objective function optimized in this application is the problem of minimizing the average delay of the system. If a candidate service node has been used as a service node of a task vehicle, this node can act as a candidate relay node for other Tavs within the communication range. In order to ensure the minimum system delay, it is determined by comparing the number of clusters after TaV clustering, that is, the selected service node will be used as a TaV service node with a small number of clusters.

In order to express more clearly, we assume that V ₀ and V ₄ are task vehicles, A and B are their service node sets respectively, and C=A∩B is a candidate service node. By giving examples to illustrate that the elements in C should be TaV in different cases Service nodes are also elected relay nodes. The specific situation is analyzed as follows:

Case 1: As shown in Figure 3, when the first clustering structure of the service node set is: |A|>|B|, then V ₃ is the service node of V ₄ and the selected relay node of V ₀ .

Case 2: As shown in FIG. 4 , when the second clustering structure of the set of service nodes is |A|<|B|, then V ₃ is the service node of V ₀ and the selected relay node of V ₄ .

则V₃为V₀的服务节点，为V₄的获选中继节点，反之依然。Case 3: As shown in Figure 5, the third clustering structure of the service node set: |A|=|B|, if

Then V ₃ is the serving node of V ₀ and the selected relay node of V ₄ , and vice versa.

Next, RN and RsN need to be determined from the candidate relay node CrN, and the adopted strategy is whether there is an idle vehicle in the communication range of the candidate relay node, that is, β _j,k =0. If β _{i, k} = 0, this node is selected as a relay node, otherwise it cannot be used as a relay node for two-hop offloading. The sets Vrn and Vrs can be calculated by the above method.

Specifically, in the embodiment of the present invention, before the task representation is unloaded in the step S3, task allocation is further included: determining the unequal task set C allocated to the service node by the task vehicle.

最小化问题，即

越靠近0代表并行程度越高；转化后的目标函数如下：In order to minimize the average system delay, in this embodiment, the optimization objective function is transformed into the absolute value of edge node offloading and local calculation parallel difference

The minimization problem is

The closer to 0, the higher the degree of parallelism; the transformed objective function is as follows:

Then use the differential evolution algorithm (DE algorithm) to assign the offloading task to the service node:

The DE algorithm generates population individuals by encoding with floating-point vectors. In the optimization process of the DE algorithm, firstly, select two individuals from the parent individuals to perform vector difference to generate a difference vector; secondly, select another individual and sum the difference vector to generate an experimental individual; then, the parent individual and the corresponding The experimental individuals are cross-operated to generate new offspring individuals; finally, the selection operation is performed between the parent individual and the offspring individual, and the individuals that meet the requirements are saved to the next generation group.

The unloading of multi-tasking vehicles will only affect the unloading strategy, and will not affect the assignment of tasks. Only the task vehicle v _i is taken as an example to illustrate the splitting process of unequal tasks. C ₂ only affects the unloading strategy, and the unequal splitting set C of computing tasks in this section does not consider its impact. The mathematical model of the optimization problem is simplified as:

分别表示计算子任务取值范围的上界和下界。DE算法流程如下：Among them, D is the dimension of the solution space, x ₁ , x ₂ ,..., x _D respectively represent the amount of assigned calculation subtasks,

Respectively represent the upper bound and lower bound of the value range of the calculation subtask. The DE algorithm flow is as follows:

随机产生：1) Initial population: initial population

Generate randomly:

Among them, x _i (0) represents the i-th "chromosome" (or individual) of the 0th generation in the population; x _{j, i} (0) represents the j-th "gene" of the i-th "chromosome" in the 0th generation ; NP represents the population size, and rand(0, 1) represents a random number uniformly distributed in the interval (0, 1).

2) Mutation operation: DE realizes individual variation through the difference strategy, and the difference strategy is to randomly select two different individuals in the population, scale their vector difference and then perform vector synthesis with the individual to be mutated, that is

表示第g代种群中第i个个体。Among them, F is the scaling factor,

Represents the i-th individual in the g-th generation population.

In the process of evolution, in order to ensure the validity of the solution, it must be judged that each "gene" in the "chromosome" is regenerated by a random method (the same method as the generation of the initial population).

通过变异后，产生一个中间体

g generation population

After mutation, an intermediate

3) Crossover operation: Perform inter-individual crossover operation on the g-th generation population { _xi (g)} and the mutated intermediate {h _i (g+1)}:

Among them, CR is the crossover probability, and j _rand is a random integer in [1, 2, ..., D].

In order to ensure that each "chromosome" of the variant intermediate {h _i (g+1)} has at least one "gene" inherited to the next generation. The gene of the first crossover operation is to randomly select the j _rand "gene" in h _i (g+1) as the j _rand allele "gene" of the "chromosome" u _i (g+1) after crossover. In the subsequent crossover operation process, _xi (g) is selected through the crossover probability CR, and h _i (g+1) is used as the allele of u _i (g+1).

3) Selection operation: DE uses a greedy algorithm to select individuals to enter the next generation population:

Its specific evolution process is as follows:

(1) Determine the control parameters of the differential evolution algorithm and determine the fitness function. The control parameters of differential evolution algorithm include: population size NP, scaling factor F and hybridization probability CR.

(2) Randomly generate the initial population.

(3) Evaluate the initial population, that is, calculate the fitness value of each individual in the initial population.

(4) Judging whether the termination condition is reached or the evolution algebra reaches the maximum. If yes, terminate the evolution, and output the best individual as the optimal solution; if not, continue.

(5) Perform mutation and crossover operations to obtain intermediate populations.

(6) Select individuals from the original population and the intermediate population to obtain a new generation population.

(7) Evolution algebra g=g+1, go to step (4).

表示；其中D_i为卸载任务的大小，C_i为任务所需计算资源，Tmaxi为执行任务的最大可容限时延；S4、卸载方式选择：在定义节点状态后，选择本地计算和多服务节点边缘卸载计算进行任务卸载，其中所述边缘卸载计算包括单跳卸载与两跳卸载；S5、执行卸载任务：在选择边缘卸载计算基础上，判断任务车辆TaV中v_i的服务节点v_j的空闲状态，且所述服务节点v_j满足充当所述中继节点的要求；随后基于空闲状态的结果判断，通过所述单跳卸载方式执行卸载任务；若所述空闲状态不满足单跳卸载条件，则将计算卸载任务通过所述两跳卸载到所述中继节点v_j的空闲服务节点v_k中执行。本发明提供一种多智能体多任务协同卸载方法，其能够利用任务车辆有限计算资源和周围空间的服务资源，解决多车辆多任务协同卸载冲突问题，实现最小的系统平均时延。From the above, the multi-agent and multi-task cooperative unloading method involved in the embodiment of the present invention includes: S1, setting the vehicle task unloading scene: identifying the vehicle that generates the task as the task vehicle TaV; when the task vehicle is at its maximum tolerable Yannei is always within the communication range, configured as a service node that provides unloading services for the task vehicle; S2, determining the node type: identifying a service node that is within the communication range of multiple task vehicles TaV as a candidate service node; when the candidate After the service node is determined as the service node of vi in the task vehicle _TaV , the current candidate service node can act as a candidate relay node for the remaining task vehicles; then select the relay node that can perform two-hop offloading from the candidate relay nodes, and The service node of the relay node; S3, unloading task representation: the task generated by any task vehicle v is _triplet :

where D _i is the size of the offloading task, C _i is the computing resource required by the task, and Tmaxi is the maximum tolerable delay in executing the task; S4. Selection of offloading mode: after defining the node status, select local computing and multi-service nodes The edge offloading calculation performs task offloading, wherein the edge offloading calculation includes single-hop offloading and two-hop offloading; S5. Execute the offloading task: on the basis of selecting the edge offloading calculation, determine the idleness of the service node v _j of v _i in the task vehicle TaV state, and the service node v _j meets the requirements of acting as the relay node; then judge based on the result of the idle state, and perform the offloading task through the single-hop offloading method; if the idle state does not meet the single-hop offloading condition, Then the calculation offloading task is offloaded to the idle service node v _k of the relay node v _j through the two hops for execution. The invention provides a multi-agent and multi-task cooperative unloading method, which can solve the problem of multi-vehicle multi-task cooperative unloading conflicts by utilizing the limited computing resources of task vehicles and service resources in the surrounding space, and realize the minimum average system delay.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this application may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention will not be limited to these embodiments shown in this application, but will conform to the widest scope consistent with the principles and novel features disclosed in this application.

Claims (10)

1. A multi-agent multi-task cooperative unloading method is characterized by comprising the following steps:

s1, setting a vehicle task unloading scene: identifying the vehicle producing the mission as mission vehicle TaV; the task vehicle is always within a communication range within the maximum tolerable time delay, and a service node for providing unloading service for the task vehicle is configured;

s2, determining the node type: service nodes belonging to the communication range of the plurality of task vehicles TaV are determined as candidate service nodes; when the candidate service node is determined to be v in task vehicle TaV _i After the service node is obtained, the current candidate service node can be used as a candidate relay node of the rest task vehicles; then, screening out relay nodes capable of executing two-hop unloading and service nodes of the relay nodes from the candidate relay nodes;

s3, unloading task representation: will arbitrarily task the vehicle v _i Generated task-use triplets:

represents; wherein D _i To unload the size of a task, C _i Tmax, the computational resources required for a task _i Maximum tolerable latency for executing the task;

s4, unloading mode selection: after defining the node state, selecting local computation and multi-service node edge unloading computation for task unloading, wherein the edge unloading computation comprises single-hop unloading and two-hop unloading;

s5, executing an unloading task: on the basis of selecting edge unloading calculation, judging v in task vehicle TaV _i Service node v _j And said serving node v, and _j satisfying a requirement to act as the relay node; then, based on the result judgment of the idle state, executing the unloading task in the single-hop unloading mode;

if the idle state does not meet the single-hop unloading condition, the calculation unloading task is unloaded to the relay through the two hopsNode v _j Idle service node v _k Is executed.

2. The multi-agent multi-tasking cooperative offloading method of claim 1, wherein in step S4, further comprising:

if the local computation is selected, namely TaV selects to use local computing resources to process tasks, the local computation time delay is directly obtained

Wherein,

representing the local computing power of TaV,

representing the computational load of the local task.

3. The multi-agent multi-task cooperative offloading method of claim 2, wherein in the step S4, the defining node state is specifically: using a variable alpha of 0-1 _i,j Representing said mission vehicle v in single hop unloading _i Of a serving node v within communication range _j The state of (2):

using a variable beta of 0-1 _j,k Indicating a relay node v when performing two-hop offloading _j Service node v _k The state of (1):

4. the multi-agent multi-task cooperative offloading method of claim 3, wherein in the step S5, the step of performing an offloading task is specifically:

judging v in task vehicle TaV _i Service node v _j Idle state of (2):

if the task vehicle v _i Service node v of _j In an idle state, i.e. alpha _i,j =0; offloading tasks directly to service node v through single hop offloading _j Calculating to obtain single-hop calculation time delay;

wherein,

for task vehicles v _i To the service node v _j A distributed sub-computation task; r is _i,j Is v is _i Transmitting a computing task to v _j The upload rate of (d); mu.s ₁ The overlapping factor in the communication process represents the proportion of data which can be unloaded to the edge node;

representing a service node v _j For task vehicle v _i The calculated force of the distribution;

if the task vehicle v _i Service node v _j In an occupied state, i.e. alpha _i,j =1, description serving node v _j Has been selected as a serving node by the remaining TaV; at a service node v _j Under the condition of meeting the requirement of serving as a relay node, the node v _j Serving node v offloaded to its idle as a relay node over two hops _k Performing off-load computing tasks, i.e. beta _j,k =0; then obtain twoJump calculation delay:

wherein,

denotes v _i Will calculate the subtask

Upload to relay node v _j ；

Indicating a relay node v _j Will calculate the task

Two-hop offload to relay node v _j Service node v _k (ii) a Vt is represented as a set of mission vehicles TaV; vs is represented as a collection of service nodes.

5. The multi-agent multi-task cooperative offloading method of claim 3, wherein in step S2, after completing the determining the node type, further comprising: constructing a vehicle motion model, passing a mission vehicle v _α Position information at time t

Predicting position information at t + delta t moment

6. The multi-agent multi-task collaborative offloading method of claim 5, wherein a vehicle motion model is constructed and location information is predicted

The method specifically comprises the following steps:

s1, obtaining task vehicle v _α Velocity at time t

The included angles with the x-axis, the y-axis and the z-axis are theta _x ,θ _y ,θ _z ∈[0,π]；

S2, according to the speed

And included angle acquisition speed

Velocity components in the x, y and z axes;

s3, obtaining a vehicle running delta t through the direction cosine law, and then obtaining a task vehicle v _α Distance components traveled in the x, y and z directions;

cos is obtained by the direction cosine theorem ² θ _x +cos ² θ _y +cos ² θ _z =1, i.e.:

s4, subsequent application of L ₂ Calculating norm to obtain task vehicle v _α To v _β Euclidean distance at any instant:

namely that

7. The multi-agent multitask coordinated offload method according to claim 6, wherein in said step S2 determining node type, further comprising: clustering the running roads of the vehicles based on the task vehicles, wherein N +1 edge servers are provided, and M task vehicles Vt are provided, so that service node clustering of the multi-task vehicles is performed;

wherein the step of clustering the service nodes specifically comprises:

s1, from a vehicle set V = { V = ₀ ,v ₁ ,v ₂ ,…,v _N Randomly selecting M task vehicles as mass centers;

s2, sampling in the service node, and calculating Euclidean distances from each sample to M centroids;

s3, if the sample is within the range of delta t epsilon from 0 to Tmax _i ]Euclidean distance from any time to centroid

Entering the subsequent step; wherein R is the coverage radius of a roadside unit RSU which is arranged on a road and provides service for all equipment;

s4, if the sample is in the communication range of the centroids, alpha _i,j =1 using the sample as the selected serving node; otherwise alpha _i,j =0, the sample is divided into clusters corresponding to the centroid;

and S5, finally, outputting a clustering result to obtain a service node set Vs of the task vehicle.

8. The multi-agent multi-task cooperative offloading method of claim 1, further comprising, before the offloading mode selection in step S4: constructing a communication model, and acquiring the rate of a communication uploading link; the method comprises the following steps:

s1, setting a communication mode between a task vehicle and a service node: a short-range wireless communication mode; orthogonal frequencies are adopted to reduce the mutual influence in the communication process;

s2, setting an uplink in the communication process as a flat Rayleigh fading channel without considering channel interference;

s3, calculating the task vehicle v according to the Shannon formula _i To the service node v _j Average transmission rate of the uplink:

wherein, v2v or v2r, W _v2v And W _v2r Respectively representing V2V and V2R channel bandwidths; p is _i Is a task vehicle v _i Of the transmitted power, p ₀ Representing the white Gaussian noise power, d _i,j Is v _i To v _j The transmission distance of (a); θ represents a path loss exponent; the channel fading coefficient is denoted by h.

9. The multi-agent multi-task collaborative offload method according to claim 6, further comprising, before offloading the task representation in the step S3: and determining an unequal task set C of the task vehicle distributed to the service nodes.

10. The multi-agent multi-task cooperative work method according to claim 9, wherein the specific step of task assignment comprises:

s1, setting an objective function, wherein the objective function aims to solve a calculation time delay problem of multi-vehicle multi-task unloading;

s2, converting the target function into an absolute value of a parallel difference between unloading of the edge node and local computation

Minimization problem, i.e.

Closer to 0 represents higher parallelism; the transformed objective function is:

st.

C ₁ :

C ₂ :

C ₃ :

wherein, C represents a computation subtask with unequal splitting; constraint C ₁ Representing constraint boundaries for local and offload computation subtasks; c ₂ Indicating the maximum tolerance time limit Tmax of the task _i Inner mission vehicle v _i To the service node v _j Relative Euclidean distance therebetween is less than or equal to v _i The communication range R of (2); c ₃ Indicating that the sum of local and multi-service node offload computation subtasks equals task vehicle v _i Total calculation task of

And S3, distributing the unloading task to the service node by utilizing a differential evolution algorithm.

Images