Cooperative task processing system for vehicle-road collaborative vehicle networking

Abstract

The invention provides a cooperative task processing system for a vehicle-road cooperation vehicle networking, which is characterized by comprising a core cloud, an MEC terminal, a plurality of task vehicle terminals and a plurality of service vehicle terminals, and is characterized in that the MEC terminal is in communication connection with the core cloud, each task vehicle terminal and each service vehicle terminal; each task vehicle end is in communication connection with each service vehicle end, the task vehicle ends generate unloading task data according to tasks needing to be unloaded, and the unloading task data are processed through the MEC end or the task vehicle ends or the service vehicle ends. In a word, the method can realize reasonable allocation of computing resources in the Internet of Vehicles and maximization of the task processing quantity.

Description

Technical Field

[0001] The present invention belongs to the technical field of task offloading, and specifically relates to a cooperative task processing system for a vehicle-road cooperative Internet of Vehicles. Background Art

[0002] As traffic safety issues and traffic congestion become increasingly serious, intelligent connected vehicles (ICVs) have become a new potential solution. ICVs integrate advanced driver assistance systems and Internet of Vehicles technology to achieve vehicle intelligence and connectivity. Advanced driver assistance systems enable vehicles to autonomously perceive their surroundings by deploying a large number of sensors such as cameras, microwave radars, and lidars on the vehicle side. Internet of Vehicles technology, on the other hand, enables information exchange between vehicles and their surroundings by deploying communication equipment on vehicles and surrounding facilities, greatly expanding the vehicle's perception range and breaking through the limitations of traditional perception components. The above-mentioned applications usually require complex deep learning operations for image detection, processing, and control decision-making. However, due to the space and power consumption limitations of vehicles, the computing power of vehicles is often limited, resulting in application processing delays that are difficult to meet actual needs, which seriously affects the user experience.

[0003] In this context, based on mobile edge computing technology and Internet of Vehicles communication technology, sufficient computing resources are deployed on the roadside and in the cloud, and the application's pending tasks are offloaded to specific objects through the Internet of Vehicles link, thereby processing tasks that the vehicle itself cannot complete.

[0004] However, the existing mobile edge computing technology still has problems such as long task processing delay, inability to reasonably arrange task processing objects, and inability to respond in time. In short, the existing mobile edge computing technology cannot achieve the reasonable allocation of computing resources. Summary of the invention

[0005] The present invention is made to solve the above-mentioned problems, and its purpose is to provide a cooperative task processing system for vehicle-road cooperative vehicle networks.

[0006] The present invention provides a cooperative task processing system for a vehicle-road cooperative Internet of Vehicles, which has the following characteristics: a core cloud, an MEC end, multiple task vehicle ends and multiple service vehicle ends, wherein the MEC end is respectively communicated with the core cloud, each task vehicle end and each service vehicle end, each task vehicle end is respectively communicated with each service vehicle end, the task vehicle end generates unloading task data according to the task to be unloaded, and processes the unloading task data through the MEC end or the task vehicle end or the service vehicle end, the core cloud includes: a core cloud computing power data generation module for generating core cloud computing power resource data; a core cloud task processing module for processing received unloading task data, the MEC end includes: a MEC computing power data generation module for generating MEC computing power resource data according to the core cloud computing power resource data sent by the core cloud; an unloading task matching module for generating unloading tasks according to the core cloud computing power resource data sent by the core cloud, the MEC computing power resource data, the service vehicle computing power resource data sent by the service vehicle end and the task vehicle computing power resource data sent by the task vehicle end Matching result, MEC task processing module, used to process the received offloading task data; the service vehicle side includes: service computing power data generation module, used to generate service vehicle computing power resource data according to MEC computing power resource data and core cloud computing power resource data sent by MEC side; service vehicle task processing module, used to process the received offloading task data, the task vehicle side includes: task side data transmission module; auxiliary computing data generation module, used to generate auxiliary computing data corresponding to the task; task computing power data generation module, used to generate task vehicle computing power resource data according to the auxiliary computing data, as well as the core cloud computing power resource data, MEC computing power resource data and service vehicle computing power resource data sent by MEC side; task side control module, used to control the task side data transmission module to send the offloading task data to the target side when the offloading task matching result is successful, and control the task computing power data generation module and the task computing power data generation module to regenerate the task vehicle computing power resource data when the offloading task matching result is failed, and the target side is the core cloud or MEC side or service vehicle side matched in the offloading task matching result.

[0007] In the cooperative task processing system for the vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: wherein, the auxiliary computing data includes the task computing power resource size, the core cloud computing power resource data includes the core cloud computing power resource size and the core cloud quotation, the MEC computing power resource data includes the MEC computing power resource size and the MEC quotation, the service vehicle computing power resource data includes the service vehicle computing power size and the service vehicle quotation, and the task vehicle computing power resource data includes the task computing power resource size and the task vehicle quotation.

[0008] The cooperative task processing system for vehicle-road cooperative vehicle network provided by the present invention may also have the following features: wherein, in the MEC computing power data generation module, the calculation expression of the MEC quotation is: price MEC =θprice CLOUD , where price MEC is the MEC quotation, θ is the preset priority task offloading value, price CLOUD For the core cloud quotation, in the service computing power data generation module, the calculation expression for the service vehicle quotation is: price SV =γlog(10-s MEC / s sv )·price CLOUD , where price SV is the price quote for the service vehicle, γ is the price coefficient, s MEC is the size of MEC computing resources, s sv is the computing power of the service vehicle. When the computing power resource size of the task is greater than the computing power resource size of the MEC and greater than the computing power size of any service vehicle, the expression for generating the quotation of the task vehicle by the task computing power data generation module is: price TV =price CLOUD , where price TV When the task computing resource size is less than or equal to the MEC computing resource size or the service vehicle computing resource size, the task computing data generation module generates the task vehicle quotation expression as follows: price TV =price MEC *λ TV ln(1+s j / s TV ), where λ TV is the profit coefficient, s TV is the computing power of the task vehicle, s i The computing power resource size of the task.

[0009] In the cooperative task processing system for the vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: when the number of service vehicle terminals is greater than the number of task vehicle terminals, the unloading task matching module generates the unloading task matching result according to the continuous matching principle; when the number of service vehicle terminals is less than the number of task vehicle terminals, the unloading task matching module generates the unloading task matching result according to the collective matching principle.

[0010] In the cooperative task processing system for the vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: wherein, the process in which the offloading task matching module generates the offloading task matching result according to the continuous matching principle includes the following steps: a service queue generation step, in which the offloading task matching module generates a service queue according to the MEC computing power resource data, the core cloud computing power resource data and the computing power resource data of each service vehicle received in real time; a demand queue generation step, in which the offloading task matching module generates a demand queue according to the computing power resource data of each task vehicle received in real time; a matching step, in which the offloading task matching module matches each task according to the service queue and the demand queue, and generates the offloading task matching result corresponding to each task.

[0011] In the cooperative task processing system for vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: wherein, the service queue is composed of various service vehicle terminals, MEC terminals and core cloud terminals, and is arranged in order from large to small according to the corresponding quotations; the demand queue is composed of various task vehicle terminals, and is arranged in order from large to small according to the corresponding task vehicle quotations; when the quotations in the service queue and the demand queue are the same, they are arranged according to the time order of receiving the quotations by the MEC terminal, and the unloading task matching module performs task matching according to the queue order.

[0012] In the cooperative task processing system for vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: wherein, the process in which the offloading task matching module generates the offloading task matching result according to the set matching principle includes the following steps: Step S1, the offloading task matching module takes all the core cloud computing power resource data, MEC computing power resource data, service vehicle computing power resource data and task vehicle computing power resource data received within a preset time period as centralized data; Step S2, the offloading task matching module solves the centralized data with KM algorithm to obtain the offloading task matching results of each task.

[0013] In the cooperative task processing system for vehicle-road cooperative vehicle network provided by the present invention, it can also have the following characteristics: wherein, step S2 includes the following sub-steps: step S2-1, taking the computing power resource data of each task vehicle in the centralized data as the demand side, and taking the computing power resource data of each service vehicle, MEC computing power resource data and core cloud computing power resource data in the centralized data as the service side; step S2-2, constructing the original bipartite graph according to the matching relationship between the computing power resource size and the quotation of all service parties and demand parties; step S2-3, assigning values ​​to each edge and vertex in the original bipartite graph according to the quotations of the demand party and the service party; step S2-4, for each vertex in the original bipartite graph, using the Hungarian algorithm to find out whether there is an augmented path in turn, if so, exchange the matched edges and unmatched edges, if not, calculate the path difference and modify the label value of each vertex, step S2-5, setting the unloading task matching result corresponding to the demand side corresponding to the edge in the original bipartite graph to success, and setting the other unloading task matching results to failure. When the unloading task matching result is successful, the target end corresponding to the demand side is the service side connected by the corresponding edge in the original bipartite graph.

[0014] Functions and Effects of the Invention

[0015] According to the cooperative task processing system for vehicle-road cooperative Internet of Vehicles involved in the present invention, when resources are abundant, the MEC end sorts the computing resource abundance of the core end, MEC end and each service vehicle end in real time according to the real-time core cloud computing power resource data, MEC computing power resource data and service vehicle computing power resource data, and sorts the importance of the tasks corresponding to each task vehicle end in real time according to the real-time task vehicle computing power resource data, and then matches them in real time according to the order; when resources are tight, the computing resource abundance of the core end, MEC end and each service vehicle end within a period of time, as well as the importance of the tasks corresponding to each task vehicle end are concentrated and a bipartite graph is constructed, and then the optimal task matching result is calculated. Therefore, the cooperative task processing system for vehicle-road cooperative Internet of Vehicles of the present invention can realize the reasonable allocation of computing resources in the Internet of Vehicles and maximize the number of task processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic diagram of a cooperative task processing system in an embodiment of the present invention;

[0017] Figure 2 is a block diagram of the core cloud in an embodiment of the present invention;

[0018] Figure 3 is a block diagram of an MEC terminal in an embodiment of the present invention;

[0019] Figure 4 is a schematic diagram of a process for generating an offloading task matching result according to a continuous matching principle in an embodiment of the present invention;

[0020] Figure 5 It is a schematic diagram of a process of generating an offloading task matching result according to a set matching principle in an embodiment of the present invention;

[0021] Figure 6 is a schematic diagram of performance comparison of various matching algorithms in an embodiment of the present invention;

[0022] Figure 7 is a block diagram of a service vehicle end in an embodiment of the present invention;

[0023] Figure 8 is a block diagram of a task vehicle end in an embodiment of the present invention;

[0024] Fig. 9 It is a schematic diagram of the actions of each terminal at each time point when the task is unloaded in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following embodiments and the accompanying drawings specifically illustrate the cooperative task processing system of the present invention for vehicle-road cooperative vehicle networks.

[0026] In this embodiment, a cooperative task processing system for a vehicle-road cooperative vehicle network is provided, hereinafter referred to as the cooperative task processing system, which is used to allocate computing resources to vehicles traveling in an urban road network.

[0027] Figure 1 Schematic diagram of a cooperative task processing system in an embodiment of the present invention.

[0028] like Figure 1 As shown, the cooperative task processing system 100 includes a core cloud 10, multiple MEC terminals 20, multiple service vehicle terminals 30 and multiple task vehicle terminals 40. In this embodiment, the core cloud 10 is a core cloud, the MEC terminal 20 is a mobile edge server MEC, the service vehicle terminal 30 is a vehicle with idle computing resources, and the task vehicle terminal 40 is a vehicle without idle computing resources and needs to have tasks to be unloaded.

[0029] Among them, the MEC end 20 is connected to the core cloud 10 for communication, and is connected to each task vehicle end 40 and each service vehicle end 30 within its service range. In this embodiment, each MEC end 20 is connected to the corresponding wireless network base station, is connected to the core cloud 10 for communication through the corresponding base station, and performs V2I communication with all task vehicle ends 40 and service vehicle ends 30 within the range of the base station. That is, each MEC end 20 performs task matching with the task vehicle end 40 and the service vehicle end 30 within its range.

[0030] Each task vehicle terminal 40 is also respectively connected to each service vehicle terminal 30. In this embodiment, the task vehicle terminal 40 and the service vehicle terminal 30 perform V2V communication.

[0031] Figure 2 It is a block diagram of the core cloud in an embodiment of the present invention.

[0032] like Figure 2 As shown, the core cloud 10 includes a core cloud computing power data generation module 101, a core cloud task processing module 102, and a core-side control module 103 for controlling the operation of the above modules. The core cloud 10 of this embodiment has the highest computing power in the cooperative task processing system 100.

[0033] The core cloud computing power data generation module 101 is used to generate core cloud computing power resource data.

[0034] Among them, the core cloud computing resource data includes the core cloud computing resource size and core cloud quotation.

[0035] The core cloud task processing module 102 is used to process the received offloading task data.

[0036] The core-side control module 103 stores a control program for controlling the operation of each module in the core cloud 10 .

[0037] Figure 3 It is a block diagram of the MEC terminal in an embodiment of the present invention.

[0038] like Figure 3 As shown, the MEC end 20 includes a MEC computing power data generation module 201, an offloading task matching module 202, a MEC task processing module 203 and a MEC side control module 204 for controlling the above modules.

[0039] The MEC computing power data generation module 201 is used to generate MEC computing power resource data based on the core cloud computing power resource data sent by the core cloud 10.

[0040] Among them, MEC computing power resource data includes MEC computing power resource size and MEC quotation.

[0041] Furthermore, in the MEC computing power data generation module 201, the calculation expression of the MEC quotation is:

[0042] price MEC =θprice CLOUD ,

[0043] Where price MEC is the MEC quotation, θ is the preset priority task offloading value, price CLOUD Quote for Core Cloud.

[0044] The offloading task matching module 202 is used to generate an offloading task matching result based on the core cloud computing power resource data, MEC computing power resource data sent by the core cloud end 10, the service vehicle computing power resource data sent by the service vehicle end 30, and the task vehicle computing power resource data sent by the task vehicle end 40.

[0045] When the number of the service vehicle terminals 30 is greater than the number of the task vehicle terminals 40 , the unloading task matching module 202 generates an unloading task matching result according to the continuous matching principle.

[0046] Figure 4 It is a flow chart of generating offloading task matching results according to the continuous matching principle in an embodiment of the present invention.

[0047] like Figure 4 As shown, the process of the offloading task matching module 202 generating the offloading task matching result according to the continuous matching principle includes the following steps:

[0048] Step T1 is the service queue generation step. The offload task matching module 202 generates a service queue based on the real-time received MEC computing resource data, core cloud computing resource data and computing resource data of each service vehicle. The service queue is composed of each service vehicle terminal 30, MEC terminal 20 and core cloud terminal 10, and is arranged in descending order according to the corresponding quotations. When the quotations are the same, they are arranged according to the time order in which the MEC terminal receives the quotations, that is, the terminal corresponding to the quotation received first is arranged in front of the service queue.

[0049] Step T2 is the demand queue generation step, where the offloading task matching module 202 generates a demand queue based on the computing power resource data of each task vehicle received in real time. The demand queue is composed of each task vehicle terminal 40, and is arranged in descending order according to the corresponding task vehicle quotations. When the quotations are the same, they are arranged according to the time sequence of the MEC terminal receiving the quotations, that is, the task vehicle terminal 40 corresponding to the quotation received first is arranged in front of the demand queue.

[0050] Step T3 is the matching step, in which the offloading task matching module 202 matches each task according to the service queue and the demand queue, and generates the offloading task matching results corresponding to each task. The offloading task matching module 202 matches tasks according to the queue order, that is, the demand queue and the front end are matched with the front end of the service queue in turn, and the first matched end in the service queue is used as the target end.

[0051] In this embodiment, step T1 is performed first, and then step T2 is performed. In other embodiments, step T2 may be performed first, and then step T1, or step T1 and step T2 may be performed together as parallel steps according to actual needs.

[0052] When the number of the service vehicle terminals 30 is less than the number of the task vehicle terminals 40 , the unloading task matching module 202 generates an unloading task matching result according to the set matching principle.

[0053] Figure 5 It is a flow chart of generating offloading task matching results according to the set matching principle in an embodiment of the present invention.

[0054] like Figure 5 As shown, the process of the offloading task matching module 202 generating the offloading task matching result according to the set matching principle includes the following steps:

[0055] Step S1, the offloading task matching module 202 takes all the core cloud computing power resource data, MEC computing power resource data, service vehicle computing power resource data and task vehicle computing power resource data received within a preset time period as centralized data.

[0056] Step S2, the offloading task matching module 202 performs KM algorithm to solve the centralized data to obtain the offloading task matching results of each task.

[0057] Wherein, step S2 includes the following sub-steps:

[0058] Step S2-1, taking the computing power resource data of each task vehicle in the centralized data as the demand side, and taking the computing power resource data of each service vehicle, MEC computing power resource data and core cloud computing power resource data in the centralized data as the service side.

[0059] Step S2-2, construct the original bipartite graph based on the matching relationship between the computing power resource size and quotations of all service providers and demand parties.

[0060] Step S2-3, assigning values ​​to the edges and vertices in the original bipartite graph according to the quotations of the demander and the service provider.

[0061] Step S2-4, for each vertex in the original bipartite graph, use the Hungarian algorithm to find out whether there is an augmented path. If so, exchange the matched edges and the unmatched edges. If not, calculate the path difference and modify the label value of each vertex.

[0062] Step S2-5, setting the matching result of the offloading task corresponding to the demand side corresponding to the edge in the original bipartite graph to success, and setting the matching results of the other offloading tasks to failure.

[0063] Among them, when the offloading task matching result is successful, the target end corresponding to the demander is the service party connected by the corresponding edge in the original bipartite graph.

[0064] In order to verify the set matching principle of this embodiment, namely KM, it is compared with the existing selfless matching algorithm SLMA, selfish matching algorithm SMA and random matching algorithm RMA.

[0065] Figure 6 4 is a schematic diagram for comparing the performance of various matching algorithms in the embodiments of the present invention.

[0066] like Figure 6 As shown in the figure, the horizontal axis is the total number of vehicles, and the vertical axis is the number of successfully matched tasks. It can be seen that KM has a better task matching success rate than the three existing algorithms.

[0067] The MEC task processing module 203 is used to process the received offloading task data.

[0068] The MEC side control module 204 stores a control program for controlling the operation of each module in the MEC end 20 .

[0069] Figure 7 It is a block diagram of a service vehicle end in an embodiment of the present invention.

[0070] like Figure 7 As shown, the service vehicle end 30 includes a service computing power data generation module 301, a service vehicle task processing module 302 and a service side control module 303 for controlling the above modules.

[0071] The service computing power data generation module 301 is used to generate service vehicle computing power resource data based on the MEC computing power resource data and core cloud computing power resource data sent by the MEC terminal 20.

[0072] Among them, the service vehicle computing power resource data includes the service vehicle computing power size and the service vehicle quotation.

[0073] Furthermore, in the service computing power data generation module 301, the calculation expression of the service vehicle quotation is:

[0074] price SV =γlog(10-s MEC / s sv )·price CLOUD ,

[0075] Where price SV is the price quote for the service vehicle, γ is the price coefficient, s MEC is the size of MEC computing resources, s sv To serve the vehicle computing power.

[0076] The service vehicle task processing module 302 is used to process the received unloading task data.

[0077] The service-side control module 303 stores a control program for controlling the operation of each module in the service vehicle end 30 .

[0078] Figure 8 It is a block diagram of the task vehicle end in an embodiment of the present invention.

[0079] like Figure 8 As shown, the task vehicle end 40 includes a task side data transmission module 401, an auxiliary computing data generation module 402, a task computing power data generation module 403 and a task side control module 404 that controls the above modules.

[0080] The auxiliary computing data generating module 402 is used to generate auxiliary computing data corresponding to the task.

[0081] Among them, the auxiliary computing data includes the size of task computing resources.

[0082] The task computing power data generation module 403 is used to generate task vehicle computing power resource data based on the auxiliary computing data, as well as the core cloud computing power resource data, MEC computing power resource data and service vehicle computing power resource data sent by the MEC terminal 20.

[0083] Among them, the task vehicle computing power resource data includes the task computing power resource size and the task vehicle quotation.

[0084] Furthermore, when the task computing power resource size is greater than the MEC computing power resource size and greater than the computing power size of any service vehicle, the task computing power data generation module 403 generates the task vehicle quotation expression as follows:

[0085] price TV =price CLoUD ,

[0086] Where price TV Get a quote for a mission vehicle.

[0087] When the task computing power resource size is less than or equal to the MEC computing power resource size or the service vehicle computing power size, the task computing power data generation module 403 generates the task vehicle quotation expression as follows:

[0088] price TV =price MEC *λ TV ln(1+s j / s TV ),

[0089] Where λ TV is the profit coefficient, s TV is the computing power of the task vehicle 40, s j The computing power resource size of the task.

[0090] The task side control module 404 stores a control program for controlling the operation of each module in the task vehicle end 40 .

[0091] Among them, when the offloading task matching result is successful, the task side control module 404 controls the task side data transmission module 401 to send the offloading task data to the target end. When the offloading task matching result is failed, the task side control module 404 controls the task computing power data generation module 402 and the task computing power data generation module 403 to regenerate the task vehicle computing power resource data. The target end is the core cloud end 10 or the MEC end 20 or the service vehicle end 30 matched in the offloading task matching result.

[0092] The following describes the actions of each terminal at each time point when the cooperative task processing system 100 is used to perform task offloading in conjunction with the accompanying drawings.

[0093] Fig. 9 It is a schematic diagram of the actions of each terminal at each time point when the task is unloaded in an embodiment of the present invention.

[0094] like Fig. 9 As shown, t0 is the initial time point, and the minimum executable unit of the time period is Δ t / 2.

[0095] At time t0, the core cloud 10, i.e., the core cloud, sends the core cloud computing resource data to the MEC end, i.e., MEC 20, to notify the MEC of the quotation. The MEC generates the MEC computing resource data (s MEC , N MEC , price MEC ) and sent to the service vehicle end 30, that is, MEC broadcasts the quotation, where s MEC is the computing resource size of MEC, N MEC is the number of cycles that can provide service, price MEC The price of occupying one unit of MEC time is the MEC quotation. After receiving the MEC computing power resource data, the service vehicle will provide the MEC with the service vehicle computing power resource data (s SV , N SV , prices V ), where s SV is the computing power of the service vehicle, N SV The number of service cycles that can be provided for the service vehicle. Its value depends on the time the service vehicle stays in the MEC service area. SV The price of occupying a service vehicle for one unit of time is the service vehicle quotation. In this embodiment, the MEC terminal 20 can process multiple tasks in parallel. If the number of tasks being processed by the MEC terminal 20 reaches the upper limit or the computing resources are all occupied, then N MEC =0.

[0096] From t0 to t0+Δ t At / 2, MEC receives the service vehicle computing resource data from the service vehicle.

[0097] At t0+Δ t At time / 2, MEC sends the core cloud computing resource data, MEC computing resource data and service vehicle computing resource data to the task vehicle terminal 40, i.e., the task vehicle, i.e., the MEC broadcast quotation. After receiving the data broadcast by MEC, the task vehicle sends the task vehicle computing resource data (C i , price TV ), where C i is the computing resources consumed by task i that needs to be offloaded, that is, the size of the task computing resources, price TV The quotation for the task vehicle corresponding to task i is the highest price.

[0098] At t0+Δ t From t / 2 to t1, MEC receives the computing power resource data of the task vehicle and selects different matching methods for task matching based on whether the computing resources of the service vehicle are surplus.

[0099] At time t1, MEC feeds back the unloading task matching result to the task vehicle and the service vehicle. If the unloading task matching result is successful, the corresponding task vehicle will unload the corresponding task to the service vehicle for processing, that is, task unloading. If the unloading task matching result is a failure, the corresponding task vehicle or service vehicle will provide the MEC with the regenerated task vehicle computing power resource data or service vehicle computing power resource data. MEC also regenerates MEC computing power resource data at this moment for MEC broadcast quotation.

[0100] From time t1 to time t2, the action from time t0 to time t1 is repeated, and this cycle repeats to achieve continuous matching and unloading of tasks in the Internet of Vehicles.

[0101] In this embodiment, the core cloud computing resource data, MEC computing resource data, service vehicle computing resource data and task vehicle computing resource data are processed by the MEC end 20, so as to match the objects of each task that needs to be unloaded. The quotation of this embodiment is used to quantify the abundance of computing resources of the core cloud end 10, the MEC end 20 and the service vehicle end 30, and the importance of the tasks that need to be unloaded by the task vehicle. Thus, the MEC end 20 can give priority to urgent tasks and maximize the number of task processing according to the computing resource data and the preset matching principle.

[0102] Furthermore, by calculating and organizing the final quotation values ​​of each end, the overall situation of computing resources in task processing in the entire Internet of Vehicles can be obtained, and then the investment in the core cloud 10, MEC end 20, service vehicle end 30 and task vehicle end 40 can be adjusted.

[0103] Functions and Effects of the Embodiments

[0104] According to the cooperative task processing system for vehicle-road cooperative Internet of Vehicles involved in this embodiment, when resources are abundant, the MEC end sorts the computing resource abundance of the core end, MEC end and each service vehicle end in real time according to the real-time core cloud computing power resource data, MEC computing power resource data and service vehicle computing power resource data, and sorts the importance of the tasks corresponding to each task vehicle end in real time according to the real-time task vehicle computing power resource data, and then matches them in real time according to the order; when resources are tight, the computing resource abundance of the core end, MEC end and each service vehicle end within a period of time, as well as the importance of the tasks corresponding to each task vehicle end are concentrated and a bipartite graph is constructed, and then the optimal task matching result is calculated. Thereby, the system can realize the priority processing of emergency tasks and maximize the total number of task processing. In short, this system can realize the reasonable allocation of computing resources in the Internet of Vehicles and maximize the number of task processing.

[0105] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A cooperative task processing system for vehicle-road cooperative Internet of Vehicles, characterized in that: include: Core cloud, MEC, multiple mission vehicles and multiple service vehicles. The MEC terminal is respectively connected to the core cloud, each of the task vehicles and each of the service vehicles in communication. Each of the task vehicles is connected to each of the service vehicles for communication. The task vehicle terminal generates unloading task data according to the task to be unloaded, and processes the unloading task data through the MEC terminal or the task vehicle terminal or the service vehicle terminal. The core cloud includes: The core cloud computing power data generation module is used to generate core cloud computing power resource data; A core cloud task processing module, configured to process the received offloading task data, The MEC terminal includes: An MEC computing power data generation module, configured to generate MEC computing power resource data according to the core cloud computing power resource data sent by the core cloud; An offloading task matching module, used to generate an offloading task matching result according to the core cloud computing power resource data sent by the core cloud, the MEC computing power resource data, the service vehicle computing power resource data sent by the service vehicle end, and the task vehicle computing power resource data sent by the task vehicle end; An MEC task processing module is used to process the received offloading task data, The service vehicle end includes: A service computing power data generation module, used to generate the service vehicle computing power resource data according to the MEC computing power resource data and the core cloud computing power resource data sent by the MEC terminal; a service vehicle task processing module, configured to process the received unloading task data, The mission vehicle end includes: Task side data transmission module; An auxiliary computing data generating module, used to generate auxiliary computing data corresponding to the task; A task computing power data generation module, used to generate the task vehicle computing power resource data according to the auxiliary computing data, and the core cloud computing power resource data, the MEC computing power resource data and the service vehicle computing power resource data sent by the MEC end; The task side control module is used to control the task side data transmission module to send the unloading task data to the target end when the unloading task matching result is successful, and to control the task computing power data generation module and the task computing power data generation module to regenerate the task vehicle computing power resource data when the unloading task matching result is failed. The target end is the core cloud end or the MEC end or the service vehicle end matched in the offloading task matching result.

2. The cooperative task processing system for vehicle-road cooperative Internet of Vehicles according to claim 1 is characterized by: in, The auxiliary computing data includes the task computing resource size, The core cloud computing resource data includes the size of the core cloud computing resource and the core cloud quotation. The MEC computing resource data includes the MEC computing resource size and MEC quotation. The service vehicle computing power resource data includes the service vehicle computing power size and the service vehicle quotation. The task vehicle computing power resource data includes the task computing power resource size and the task vehicle quotation.

3. The cooperative task processing system for vehicle-road cooperative Internet of Vehicles according to claim 2 is characterized by: in, In the MEC computing power data generation module, the calculation expression of the MEC quotation is: price MEC =θprice CLOUD , Where price MEC is the price of the MEC, θ is the preset priority task offloading value, price CLOUD Provide a quote for the core cloud. In the service computing power data generation module, the calculation expression of the service vehicle quotation is: price SV =γlog(10-s MEC / s sv )·price CLOUD , Where price SV is the price quote for the service vehicle, γ is the price coefficient, s MEC is the size of the MEC computing resources, s sv is the computing power of the service vehicle, When the task computing power resource size is greater than the MEC computing power resource size and greater than the computing power size of any of the service vehicles, the task computing power data generation module generates the task vehicle quotation expression as follows: price TV =price CLOUD , Where price TV Quote for the vehicle for the mission, When the task computing power resource size is less than or equal to the MEC computing power resource size or the service vehicle computing power size, the task computing power data generation module generates the task vehicle quotation expression as follows: price TV =price MEC *λ TV ln(1+s j / s TV ), Where λ TV is the profit coefficient, s TV is the computing power of the task vehicle, s j is the computing power resource size of the task.

4. The cooperative task processing system for vehicle-road cooperative Internet of Vehicles according to claim 2 is characterized by: in, When the number of the service vehicle terminals is greater than the number of the task vehicle terminals, the unloading task matching module generates the unloading task matching result according to the continuous matching principle. When the number of the service vehicle terminals is less than the number of the task vehicle terminals, the unloading task matching module generates the unloading task matching result according to the set matching principle.

5. The cooperative task processing system for vehicle-road cooperative vehicle network according to claim 4, Features: The process of generating the offloading task matching result according to the continuous matching principle by the offloading task matching module includes the following steps: A service queue generation step, wherein the offloading task matching module generates a service queue according to the MEC computing power resource data, the core cloud computing power resource data and the computing power resource data of each service vehicle received in real time; The demand queue generation step, wherein the unloading task matching module generates a demand queue according to the computing power resource data of each task vehicle received in real time; In the matching step, the offloading task matching module matches each of the tasks according to the service queue and the demand queue, and generates the offloading task matching results corresponding to each task.

6. The cooperative task processing system for vehicle-road cooperative Internet of Vehicles according to claim 5 is characterized by: in, The service queue is composed of the service vehicle end, the MEC end and the core cloud end, and is arranged in descending order according to the corresponding quotations. The demand queue is composed of each of the task vehicle terminals, and is arranged in descending order according to the corresponding task vehicle quotations. When the quotations in the service queue and the demand queue are the same, they are arranged according to the time sequence in which the MEC receives the quotations. The offloading task matching module performs task matching according to the queue order.

7. The cooperative task processing system for vehicle-road cooperative vehicle network according to claim 5, Features: The process of generating the offloading task matching result according to the set matching principle by the offloading task matching module includes the following steps: Step S1, the offloading task matching module uses all the core cloud computing resource data, the MEC computing resource data, the service vehicle computing resource data and the task vehicle computing resource data received within a preset time period as centralized data; Step S2, the offloading task matching module performs KM algorithm to solve the centralized data to obtain the offloading task matching results of each of the tasks.

8. The cooperative task processing system for vehicle-road cooperative vehicle network according to claim 7, Features: in, The step S2 comprises the following sub-steps: Step S2-1, taking the computing power resource data of each task vehicle in the centralized data as the demand side, and taking the computing power resource data of each service vehicle, the MEC computing power resource data and the core cloud computing power resource data in the centralized data as the service side; Step S2-2, constructing an original bipartite graph according to the matching relationship between the computing power resource sizes and quotations of all the service providers and demand parties; Step S2-3, assigning values ​​to the edges and vertices in the original bipartite graph according to the quotations of the demander and the service provider; Step S2-4, for each vertex in the original bipartite graph, use the Hungarian algorithm to find out whether there is an augmented path. If so, exchange the matched edge and the unmatched edge. If not, calculate the path difference and modify the label value of each vertex. Step S2-5, setting the matching result of the offloading task corresponding to the demand side corresponding to the edge in the original bipartite graph to success, and setting the matching results of the other offloading tasks to failure, When the offloading task matching result is successful, the target end corresponding to the demander is the service party connected by the corresponding edge in the original bipartite graph.

Images