Internet of vehicles task offloading method, device, equipment, medium and program product

Abstract

The application relates to the Internet of Vehicles technical field and provides an Internet of Vehicles task offloading method, device and equipment, a storage medium and a program product, the method comprising the following steps: according to the task type of a to-be-calculated task of a target vehicle, determining a task offloading strategy corresponding to the to-be-calculated task; the task offloading strategy contains an offloading object of the to-be-calculated task; acquiring the motion feature of the target vehicle, calculating the task offloading proportion of the offloading object based on the task offloading strategy and the motion feature; and offloading the to-be-calculated task to the offloading object according to the task offloading proportion. By dividing Internet of Vehicles tasks into different types, different task offloading strategies are adopted for task offloading of different types of tasks, which is beneficial to meeting the task offloading requirements of diversified vehicle-mounted applications. Meanwhile, the task offloading proportion is determined in combination with the motion feature of the vehicle, the mobility of the vehicle is considered, the rationality and effectiveness of the selection of the offloading object are ensured, and the smooth progress of task offloading is facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Vehicles, and in particular to an Internet of Vehicles task offloading method, device, equipment, storage medium and program product. BACKGROUND

[0002] With the vigorous development of the Internet of Vehicles field, the emergence of computationally intensive vehicle-mounted applications has brought great challenges to the computing power of vehicles and the latency performance of systems. Internet of Vehicles task offloading based on cloud-edge collaboration mainly focuses on the integration of cloud computing and edge computing. By optimizing task offloading strategies, the performance and efficiency of the Internet of Vehicles are improved. For example, by prioritizing tasks and designing a cloud-edge collaborative task offloading decision mechanism, the cloud and edge nodes work collaboratively to reduce data transmission delay. Alternatively, machine learning and other technologies are introduced to build a task offloading model to intelligently optimize offloading decisions and improve the efficiency and accuracy of task offloading. However, with the continuous development of Internet of Vehicles technology and the continuous expansion of application scenarios, the types of vehicle-mounted applications tend to diversify. The existing Internet of Vehicles task offloading methods have a single offloading mode, which cannot meet the diversified offloading needs. SUMMARY

[0003] The embodiments of the present application provide an Internet of Vehicles task offloading method, device, equipment, storage medium and program product to solve the defect that the existing Internet of Vehicles task offloading mode is single and cannot meet the task offloading needs of diversified vehicle-mounted applications.

[0004] The embodiments of the present application provide an Internet of Vehicles task offloading method, comprising:

[0005] According to the task type of the to-be-computed task of the target vehicle, a task offloading strategy corresponding to the to-be-computed task is determined; the task offloading strategy contains an offloading object of the to-be-computed task;

[0006] Obtain the motion characteristics of the target vehicle, and calculate the task offloading proportion of the offloading object based on the task offloading strategy and the motion characteristics;

[0007] According to the task offloading proportion, the to-be-computed task is offloaded to the offloading object.

[0008] In one embodiment, the calculation of the task offloading proportion of the offloading object based on the task offloading strategy and the motion characteristics comprises:

[0009] According to the motion characteristics, the target residence duration of the target vehicle in the current road side unit is calculated;

[0010] determine a maximum tolerable time delay of the to-be-computed task according to the target residence duration, and calculate a task offloading ratio of the offloading object based on the task offloading strategy and the maximum tolerable time delay.

[0011] In one embodiment, if the to-be-computed task includes a traffic safety task, the offloading object includes an auxiliary vehicle and an edge server corresponding to the current road side unit; and the calculating of the task offloading ratio of the offloading object based on the task offloading strategy and the maximum tolerable time delay includes:

[0012] obtaining a candidate auxiliary vehicle within a preset distance range around the target vehicle, and screening the candidate auxiliary vehicle based on a preset mobility constraint condition to obtain an auxiliary vehicle of the target vehicle; wherein the candidate auxiliary vehicle and the target vehicle are both within a coverage range of the current road side unit, and the mobility constraint condition includes that a residence duration within the current road side unit is greater than or equal to the target residence duration, and a relative distance with the target vehicle within the target residence duration is less than the preset distance;

[0013] calculating a first task offloading ratio of the auxiliary vehicle and a second task offloading ratio of the edge server corresponding to the current road side unit based on an energy consumption optimization principle with the maximum tolerable time delay as a constraint.

[0014] In one embodiment, a first offloading time delay corresponding to the first task offloading ratio is a sum of a first transmission time delay and a first computation time delay, and a second offloading time delay corresponding to the second task offloading ratio is a sum of a second transmission time delay and a second computation time delay.

[0015] The first transmission time delay is a time delay of data transmission between the target vehicle and the auxiliary vehicle, and the first computation time delay is a time delay required for the auxiliary vehicle to compute a task amount of the first task offloading ratio.

[0016] The second transmission time delay is a time delay of data transmission between the target vehicle and the edge server corresponding to the current road side unit, and the second computation time delay is a time delay required for the edge server corresponding to the current road side unit to compute a task amount of the second task offloading ratio.

[0017] A task total time delay corresponding to the traffic safety task in the to-be-computed task is a maximum value among a third computation time delay, the first offloading time delay and the second offloading time delay, and the third computation time delay is a time delay required for local computation of the target vehicle.

[0018] The task energy consumption corresponding to the traffic safety type task in the to-be-computed task is a sum of a first offloading energy consumption, a second offloading energy consumption, and a first computation energy consumption. The first offloading energy consumption is energy consumption corresponding to the first offloading time delay. The second offloading energy consumption is energy consumption corresponding to the second offloading time delay. The first computation energy consumption is energy consumption corresponding to the third computation time delay.

[0019] In one embodiment, if the to-be-computed task includes an entertainment information type task, the offloading object includes a cloud server and an edge server corresponding to the current road side unit; and the task offloading ratio of the offloading object is calculated based on the task offloading strategy and the maximum tolerable time delay, including:

[0020] The number of target road side units under a current base station coverage of the target vehicle is obtained, as well as a remaining task amount of the to-be-computed task. The target road side units are road side units after the current road side unit along a driving direction of the target vehicle under the current base station coverage.

[0021] A third task offloading ratio of the cloud server is calculated based on an energy consumption optimization principle with the maximum tolerable time delay as a constraint, and a fourth task offloading ratio of the edge server corresponding to the current road side unit is calculated according to the number of target road side units and the remaining task amount.

[0022] In one embodiment, a third offloading time delay corresponding to the third task offloading ratio is a sum of a third transmission time delay and a fourth computation time delay, and a fourth offloading time delay corresponding to the fourth task offloading ratio is a sum of a fourth transmission time delay and a fifth computation time delay.

[0023] The third transmission time delay is a time delay of data transmission between the target vehicle and the current base station, and between the current base station and the cloud server. The fourth computation time delay is a time delay required by the cloud server for computing a task amount of the third task offloading ratio.

[0024] The fourth transmission time delay is a time delay of data transmission between the target vehicle and the edge server corresponding to the current road side unit. The fifth computation time delay is a time delay required by the edge server corresponding to the current road side unit for computing a task amount of the fourth task offloading ratio.

[0025] A total task time delay corresponding to the entertainment information type task in the to-be-computed task is a maximum value among a sixth computation time delay, the third offloading time delay, and the fourth offloading time delay. The sixth computation time delay is a time delay required by the target vehicle for local computation.

[0026] The entertainment information type task in the to-be-computed task corresponds to a task energy consumption, which is a sum of a third offloading energy consumption, a fourth offloading energy consumption and a second computation energy consumption, the third offloading energy consumption is an energy consumption corresponding to the third offloading time delay, the fourth offloading energy consumption is an energy consumption corresponding to the fourth offloading time delay, and the second computation energy consumption is an energy consumption corresponding to the sixth computation time delay.

[0027] The application further provides a vehicle networking task offloading device, comprising:

[0028] A policy determination module is configured to determine a task offloading policy corresponding to the to-be-computed task according to a task type of the to-be-computed task of the target vehicle, wherein the task offloading policy comprises an offloading object of the to-be-computed task.

[0029] A proportion calculation module is configured to acquire a motion feature of the target vehicle, and calculate a task offloading proportion of the offloading object based on the task offloading policy and the motion feature.

[0030] A task offloading module is configured to offload the to-be-computed task to the offloading object according to the task offloading proportion.

[0031] The application further provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the vehicle networking task offloading method according to any one of the above embodiments when executing the program.

[0032] The application further provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program implements the steps of the vehicle networking task offloading method according to any one of the above embodiments when executed by a processor.

[0033] The application further provides a computer program product, which comprises a computer program, and the computer program implements the steps of the vehicle networking task offloading method according to any one of the above embodiments when executed by a processor.

[0034] The vehicle networking task offloading method, device, equipment, storage medium and program product provided by the application divide vehicle networking tasks into different types, adopt different task offloading strategies for different types of tasks for task offloading, which is conducive to meeting the task offloading needs of diversified vehicle-mounted applications. Meanwhile, the task offloading proportion is determined in combination with the motion feature of the vehicle, the to-be-computed task is offloaded to the offloading object according to the task offloading proportion, the mobility of the vehicle is considered during task offloading, the rationality and effectiveness of the selection of the offloading object are ensured, and the task offloading is facilitated. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0036] Figure 1 is a flowchart of a vehicle networking task offloading method provided by an embodiment of the application;

[0037] Figure 2 is a system architecture diagram of a vehicle networking task offloading method provided by an embodiment of the application;

[0038] Figure 3 is a structure diagram of a vehicle networking task offloading device provided by an embodiment of the application;

[0039] Figure 4 is a structure diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the application clearer, the technical solutions in the application will be described clearly and completely in the following with reference to the drawings in the application. Obviously, the described embodiments are some embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0041] Figure 1 is a flowchart of a vehicle networking task offloading method provided by an embodiment of the application, as shown in Figure 1 , the method comprises the following steps:

[0042] Step 100, according to the task type of the to-be-computed task of the target vehicle, determining the task offloading strategy corresponding to the to-be-computed task; the task offloading strategy contains the offloading object of the to-be-computed task;

[0043] Step 200, obtaining the motion feature of the target vehicle, and based on the task offloading strategy and the motion feature, calculating the task offloading proportion of the offloading object;

[0044] Step 300, according to the task offloading proportion, offloading the to-be-computed task to the offloading object.

[0045] It should be noted that the vehicle networking task offloading method provided by the embodiments of the present application is applied to a target vehicle, which is any vehicle that needs to perform vehicle networking task offloading. According to the task type of the to-be-computed task of the target vehicle, a task offloading strategy corresponding to the to-be-computed task is determined, and the task offloading strategy contains an offloading object of the to-be-computed task, and the offloading object is an object for carrying out the computing work of the offloaded task of the target vehicle.

[0046] Optionally, the to-be-computed task of the target vehicle can include one or more tasks, and the task types of the multiple tasks can be the same or different. The task type of the to-be-computed task can be determined according to the application type of the application that initiates the to-be-computed task, which is not limited here.

[0047] Optionally, different task types correspond to different offloading strategies, different offloading strategies contain different offloading objects, and the offloading objects contained in the same offloading strategy can include one or more.

[0048] The motion characteristics of the target vehicle are obtained, which include but are not limited to the speed, trajectory and movement trend of the target vehicle, the task offloading proportion of the offloading object is calculated based on the task offloading strategy and the motion characteristics, and the to-be-computed task is offloaded to the offloading object according to the task offloading proportion. When there are multiple offloading objects, the task offloading proportions of different offloading objects can be the same or different.

[0049] Optionally, when there are multiple offloading objects, the multiple offloading objects can include multiple types, the task offloading proportions of offloading objects of different types can be different, and the task offloading proportions of different offloading objects of the same type can also be different.

[0050] Optionally, the offloading object includes at least one of other vehicles, cloud servers and edge servers corresponding to road side units (RSUs). Taking the RSU as an example, since the coverage range of an RSU is limited, some to-be-computed tasks can cross the RSU, and some to-be-computed tasks need to be completed within the coverage range of the same RSU. Therefore, based on the motion characteristics of the target vehicle, the amount of tasks that can be offloaded by the target vehicle in different RSU coverage ranges can be calculated, so as to obtain the task offloading proportion of the edge server corresponding to the RSU. The edge server corresponding to the RSU can also be referred to as a mobile edge computing (MEC) server.

[0051] In the embodiment, by dividing the vehicle networking task into different types, different task offloading strategies are adopted for different types of tasks for task offloading, which is beneficial to meet the task offloading needs of diversified vehicle-mounted applications. Meanwhile, the task offloading proportion is determined in combination with the motion characteristics of the vehicle, and the to-be-computed task is offloaded to the offloading object according to the task offloading proportion. The mobility of the vehicle is considered during task offloading, which ensures the rationality and effectiveness of the offloading object selection, and is beneficial to the smooth progress of task offloading.

[0052] In one embodiment, the vehicle networking task offloading method provided by the embodiment of the application is applied to a vehicle networking task offloading system. The system has a multi-level architecture based on cloud edge collaboration, can divide vehicle networking tasks into types, and adopts different task offloading strategies for different types of tasks. The multi-level architecture is as shown in Figure 2 , and includes a cloud, a base station, a road side unit (RSU) and a vehicle. The cloud server is connected with the base station through a fiber backbone network. The coverage radius of the base station is R, and the vertical distance between the base station and the road is D. A straight road passes through the cell covered by the base station. There are S RSUs distributed along the road in the cell covered by the base station. Each RSU is located at the center of the corresponding coverage range. The coverage ranges of the S RSUs are different and do not overlap. The coverage ranges of two adjacent RSUs can be continuous or discontinuous. The coverage ranges of the S RSUs can be represented by . S segments with different lengths are divided on the straight road, and the length of each segment is represented by a set . Each RSU is deployed with a corresponding MEC server. The MEC servers are connected through optical fibers and can be used to provide local distribution, computing task offloading and information sharing for vehicles. That is, as shown in Figure 2 , the vehicles communicate with each other based on V2V link, the vehicles and the RSUs communicate with each other through V2I / V2N link, and the RSUs and the base station communicate with each other through optical fiber link. The following is based on Figure 2 The vehicle networking task offloading method provided by the embodiment of the application is described in detail.

[0053] In one exemplary application scenario, it is assumed that Figure 2The straight road in the figure is a double-lane same-direction, variable-speed free-moving state, the vehicles in the two lanes are independently distributed, each vehicle is equipped with a positioning device such as a GPS (Global Positioning System), and since the coverage range of the RSU is generally small, it can be considered that the same vehicle is uniformly moving within the coverage range of the same RSU. The vehicles in the road communicate with the RSU and the base station in the V2I / V2N manner, and can also communicate with the nearby vehicles in the V2V manner. Further, the tasks of the Internet of Vehicles are divided into traffic safety type tasks and information entertainment type tasks according to the types, wherein the traffic safety type is the application task of real-time road condition information, the task quantity is small, and the maximum tolerable delay is low; the information entertainment type is the vehicle-mounted entertainment application task, the task quantity is large, and the maximum tolerable delay is relatively high.

[0054] It can be understood that different types of tasks have different data quantity and delay characteristics, considering the task interruption problem caused by the movement of the vehicle, a differentiated task offloading strategy is adopted according to different types of tasks, so as to reasonably and effectively select the offloading object. Specifically, in step 200, based on the task offloading strategy and the motion characteristics of the target vehicle, the task offloading proportion of the offloading object is calculated, including:

[0055] In step 210, according to the motion characteristics, the target residence duration of the target vehicle in the current road side unit is calculated.

[0056] In step 220, the maximum tolerable delay of the task to be calculated is determined according to the target residence duration, and the task offloading proportion of the offloading object is calculated based on the task offloading strategy and the maximum tolerable delay.

[0057] According to the motion characteristics of the target vehicle, the target residence duration of the target vehicle in the current road side unit is calculated, according to the target residence duration, the maximum tolerable delay of the task to be calculated is determined, and based on the maximum tolerable delay and the task offloading strategy, the task offloading proportion of the offloading object is calculated.

[0058] Optionally, for some offloading objects, the residence duration of the target vehicle in the current road side unit can be taken as the maximum tolerable delay, and the task quantity that can be processed by the offloading object within the time duration corresponding to the maximum tolerable delay is the task offloading proportion corresponding thereto. For other offloading objects, the residence duration of the target vehicle in the current road side unit can be estimated, the total residence duration of the target vehicle in the current road side unit and the subsequent remaining road side units is taken as the maximum tolerable delay, and the total task quantity that can be processed by the offloading object within the time duration corresponding to the maximum tolerable delay is averaged according to the number of road side units to obtain the task offloading proportion corresponding thereto.

[0059] It is assumed that at a certain moment a target vehicle in a road a certain type of vehicle computing task, which can be represented as wherein, 0 represents a traffic safety type task, and 1 represents an information entertainment type task; represents a computing resource of the target vehicle , which can be specifically a number of CPU operation cycles required for computing 1 bit of task; is a data size required for the computing task, in bits; is a maximum tolerable time delay of the computing task. In order to maximize the offloading efficiency and time delay performance, it is assumed that the computing task can be divided, i.e., the computing task can be completed by the target vehicle locally, a nearby vehicle, an MEC server at the RSU side, and a cloud. However, for different types of vehicle networking tasks, due to their different sensitivities to time delay, the selected offloading objects for the task computing are also different.

[0060] In an embodiment, the vehicle networking task at least includes a traffic safety type task and an entertainment information type task, and the to-be-computed task includes at least one of the traffic safety type task and the entertainment information type task. If the to-be-computed task includes the traffic safety type task, the offloading objects thereof include an auxiliary vehicle and an MEC server corresponding to a current road side unit. Since the traffic safety type task has a small amount of traffic safety type business and a low maximum tolerable time delay, such a computing task can be completed by offloading between the target vehicle locally, a nearby vehicle, and an MEC server at the RSU side, without the participation of the cloud. In addition, in order to ensure the real-time performance of task completion and avoid task interruption, such a task cannot be transmitted across RSUs, i.e., a computing task generated within the current RSU coverage range must be completed before the vehicle enters the next RSU range.

[0061] For the traffic safety type task, based on the residence time of the target vehicle within the current RSU coverage range, and the distance and mobility between vehicles, an optimal auxiliary vehicle is determined, and based on a task offloading ratio, a time delay and energy consumption of the target vehicle for V2V (Vehicle to Vehicle) and V2I (Vehicle to Infrastructure, where the infrastructure refers to an MEC server) collaborative offloading within the current RSU coverage range are obtained. Therefore, based on the energy consumption optimization principle and the maximum tolerable time delay as a constraint, the task offloading ratio of the target vehicle within the current RSU coverage range can be calculated. Based on this, in step 220, based on the task offloading strategy and the maximum tolerable time delay, the task offloading ratio of the offloading object is calculated, including:

[0062] In step 221, a candidate auxiliary vehicle within a preset distance range around the target vehicle is obtained, and the candidate auxiliary vehicle is screened based on a preset mobility constraint condition to obtain an auxiliary vehicle of the target vehicle; wherein the candidate auxiliary vehicle and the target vehicle are both within the coverage range of the current road side unit, and the mobility constraint condition includes that the residence time within the current road side unit is greater than or equal to the target residence time, and the relative distance with the target vehicle within the target residence time is less than the preset distance.

[0063] In step 222, a first task offloading ratio of the auxiliary vehicle and a second task offloading ratio of the edge server corresponding to the current road side unit are calculated based on the energy optimization principle with the maximum tolerable time delay as the constraint.

[0064] A candidate auxiliary vehicle within a preset distance range around the target vehicle is obtained, and the candidate auxiliary vehicle is screened based on a preset mobility constraint condition to obtain an auxiliary vehicle of the target vehicle, wherein the candidate auxiliary vehicle and the target vehicle are both within the coverage range of the target vehicle current RSU, that is, the candidate auxiliary vehicle and the target vehicle are within the coverage range of the same RSU. The mobility constraint condition specifically refers to that the residence time of the candidate auxiliary vehicle within the current RSU coverage range is greater than or equal to the residence time of the target vehicle, and within the residence time of the target vehicle, the relative distance between the candidate auxiliary vehicle and the target vehicle is less than the preset distance, so the candidate auxiliary strategy is the auxiliary vehicle that meets the mobility constraint condition and can be used as the offloading object of the target vehicle.

[0065] Further, the task offloading ratio of the offloading object of the traffic safety type task includes a first task offloading ratio of the auxiliary vehicle and a second task offloading ratio of the MEC server corresponding to the current RSU. The first task offloading ratio of the auxiliary vehicle and the second task offloading ratio of the MEC server corresponding to the current RSU are calculated based on the energy optimization principle with the maximum tolerable time delay as the preset.

[0066] For The corresponding to-be-calculated task can be divided into several main parts, including a target vehicle local calculation part, a part offloaded to an auxiliary vehicle for calculation, and a part offloaded to an RSU side MEC server for calculation. The target vehicle generating the calculation task can be referred to as a task vehicle, and the nearby vehicle assisting the task vehicle in performing the calculation task offloading is an auxiliary vehicle. Offloading the calculation task to the auxiliary vehicle and the RSU side MEC server for calculation will generate task time delay. For the traffic safety type calculation task that cannot cross the RSU, it is necessary to complete the task offloading and receive the feedback results of each offloading object within the residence time of the target vehicle within the current RSU coverage range.

[0067] Furthermore, regarding task latency, for traffic safety tasks, the corresponding task latency includes the latency generated by the target vehicle performing local calculations, and the latency of offloading the calculation task to the auxiliary vehicle and MEC server for calculation. This includes data transmission latency, calculation latency, and feedback latency. After receiving the task data uploaded by the task vehicle, the auxiliary vehicle and MEC server begin to perform calculation operations and feed back the completed calculation results to the task vehicle. Since the amount of data in the calculation results is usually very small, the feedback latency of the calculation results can generally be ignored.

[0068] Specifically, if the task to be calculated includes traffic safety tasks, the latency of the task to be calculated includes the first unloading latency corresponding to the first unloading ratio and the second unloading latency corresponding to the second task unloading ratio. The first unloading latency is the sum of the first transmission latency and the first calculation latency, and the second unloading latency is the sum of the second transmission latency and the second calculation latency. The first transmission latency is the latency of data transmission between the target vehicle and the auxiliary vehicle, and the first calculation latency is the latency required for the auxiliary vehicle to calculate the task load of the first task unloading ratio; the second transmission latency is the latency of data transmission between the target vehicle and the MEC server corresponding to the current RSU, and the second calculation latency is the latency required for the MEC server corresponding to the current RSU to calculate the task load of the second task unloading ratio. Furthermore, since the target vehicle unloads tasks for different unloading objects, and the calculation and feedback of unloading tasks by each unloading object and the target vehicle's local calculation are executed in parallel, the total latency of the traffic safety tasks in the task to be calculated is the maximum value among the third calculation latency, the first unloading latency, and the second unloading latency. The third calculation latency is the latency required for the target vehicle's local calculation.

[0069] Furthermore, the energy consumption of the traffic safety category task in the task to be calculated is the sum of the first unloading energy consumption, the second unloading energy consumption, and the first calculation energy consumption. The first unloading energy consumption is the energy consumption corresponding to the first unloading delay, the second unloading energy consumption is the energy consumption corresponding to the second unloading delay, and the first calculation energy consumption is the energy consumption corresponding to the third calculation delay required for local calculation of the target vehicle.

[0070] In one embodiment, within the coverage area of ​​the same RSU, when vehicles approach each other, the mission vehicle can be at a distance... It initiates a connection request to any surrounding vehicle within its range, thereby establishing a data transmission channel based on V2V communication. Therefore, a preset distance is required around the mission vehicle. Vehicles within the specified range are candidate auxiliary vehicles for the mission vehicle. Coverage of RSUk At that time, the set of candidate auxiliary vehicles can be represented as:

[0071] (1)

[0072] wherein, is a set of vehicles within denotes a task vehicle a position coordinate within

[0073] Since the speed of a vehicle is variable, in order to ensure that the task vehicle can complete the V2V-based computation task offloading within time, the candidate auxiliary vehicle needs to satisfy the mobility constraint condition with the task vehicle In an embodiment, the mobility constraint condition includes two conditions, which are represented by sets and auxiliary vehicles satisfying the two constraint conditions, as follows:

[0074] (2)

[0075] (3)

[0076] wherein, denotes the speed of the task vehicle denotes a set of vehicles in the candidate auxiliary vehicles that still satisfy the upper limit of the V2V communication range after the elapsed time , that is, the distance from the task vehicle is less than the preset distance, and the V2V communication process will not be interrupted due to the change of the relative displacement of the task vehicle and the auxiliary vehicle, denotes a set of vehicles in the candidate auxiliary vehicles that are still within the coverage range after the elapsed time As mentioned earlier, the computation offloading process of the traffic safety type task cannot be performed across the RSU, so the maximum delay of the task vehicle performing collaborative computation with the auxiliary vehicle cannot exceed the residence time of the task vehicle

[0077] within , that is: (4)

[0078] (5)

[0079] Further, by substituting formula 4 into formulas 2 and 3, the auxiliary vehicle of the task vehicle is represented in the form of a set as follows:

[0080] (5)

[0081] ​​For Formulas 4 and 5, under certain conditions, such as low road vehicle density leading to a lack of candidate auxiliary vehicles that meet the conditions, or candidate auxiliary vehicles not meeting the vehicle mobility constraints, the task vehicle may not have an auxiliary vehicle. Based on this, parameters are defined. For mission vehicles The auxiliary vehicle binary presence indication is shown in Formula 6 below:

[0082] (6)

[0083] Only when At that time, it indicates the mission vehicle. There are available auxiliary vehicles capable of offloading V2V-based computational tasks. At this point, the task vehicle is determined based on the principle of optimal vehicle distance. The optimal auxiliary vehicle is the closest vehicle. As shown in Formula 7 below:

[0084] (7)

[0085] Since traffic safety-related computational tasks can be completed collaboratively by the task vehicle's local machine, nearby vehicles, and the MEC server on the RSU side, based on the above analysis, the task vehicle can be determined. exist Internal auxiliary vehicles. Definition ,in Indicates the mission vehicle Unload to auxiliary vehicle The percentage of tasks and the percentage of tasks uninstalled as the first task. Indicates the mission vehicle Uninstall to The task ratio of the internal MEC server, which is also the second task offloading ratio. Further, the task vehicles... Traffic safety tasks The latency and energy consumption generated during the unloading and computation process include the following three parts:

[0086] mission vehicles The latency and energy consumption incurred by performing computation locally are expressed as follows:

[0087] (8)

[0088] (9)

[0089] in, For mission vehicles computing power For mission vehicles CPU energy efficiency.

[0090] the latency and energy consumption of offloading the task to the auxiliary vehicle the communication latency and energy consumption of uploading the task to the auxiliary vehicle, the latency and energy consumption of the auxiliary vehicle performing the task computation, and the latency and energy consumption of the auxiliary vehicle feeding back the computation result to the task vehicle the latency and energy consumption of the auxiliary vehicle feeding back the computation result to the task vehicle can be negligible.

[0091] Under the V2V communication mode, the task vehicle uploads the task to the auxiliary vehicle The data transmission rate can be characterized by the Shannon channel capacity:

[0092] ; (10)

[0093] wherein, the bandwidth of the task vehicle communicating with the auxiliary vehicle via V2V, the transmission power of the task vehicle , the channel gain between the task vehicle and the auxiliary vehicle, and the known Gaussian white noise power. Based on this, the communication latency of the task vehicle sending the task data to the auxiliary vehicle is:

[0094] ; (11)

[0095] Further, after receiving the task data uploaded by the task vehicle , the auxiliary vehicle starts to perform the computation operation, and the computation latency of the auxiliary vehicle is:

[0096] ; (12)

[0097] wherein, and are the computation resource and computation capability of the auxiliary vehicle , respectively.

[0098] After the auxiliary vehicle completes the computation, it feeds back the result to the task vehicle. Since the data volume of the computation result is small, the feedback latency can be ignored, and the required latency of the task vehicle offloading to the auxiliary vehicle is the sum of formula 11 and formula 12, which can be expressed as:

[0099] (13)

[0100] Similarly, the task vehicle unloads the computing task to the auxiliary vehicle The energy consumption of the computing can be expressed as:

[0101] (14)

[0102] wherein, is the CPU energy coefficient of the auxiliary vehicle .

[0103] Similarly, the partial task offloaded to the MEC server on the RSU side for computing generates the latency and energy consumption, including the communication latency and energy consumption of uploading the task data of the task to be offloaded to the MEC server on the RSU side, and the latency and energy consumption of the MEC server on the RSU side performing the task computing.

[0104] Under the V2I communication mode, the task vehicle uploads the data to the RSUkwith the data transmission rate of:

[0105] (15)

[0106] wherein, is the bandwidth of the V2I communication between the task vehicle and the MEC server on the RSU side, is the channel gain between the task vehicle and the RSU.

[0107] After the RSU receives the data uploaded by the task vehicle , the RSU transmits the data to the corresponding MEC server for computing operation, and the MEC server feeds back the result to the task vehicle after the task computing is completed. Here, the data transmission time between the RSU and the MEC server and the time of the computing result feedback can also be ignored, and the task vehicle requires the latency and energy consumption for offloading the computing task to the MEC server on the RSU side for computing, which can be expressed as follows, respectively:

[0108] (16)

[0109] (17)

[0110] wherein, and correspond to the computing resource and the computing capacity of the MEC server, respectively, is the CPU energy coefficient of the MEC server.

[0111] As described above, for traffic safety type tasks, the task vehicle The time delay and energy consumption generated by the task unloading process are represented as follows:

[0112] ; (18)

[0113] + + ; (19)

[0114] Based on formula 18 and formula 19, taking the residence duration of the target vehicle within the RSU coverage range as the maximum tolerable time delay of task offloading, based on the energy consumption optimization principle, the task offloading ratio of the auxiliary vehicle to the RSU side MEC server can be calculated, and the to-be-computed task is offloaded to the auxiliary vehicle and the RSU side MEC server for calculation according to the offloading ratio, realizing the cooperative calculation of the auxiliary vehicle, the RSU side MEC server and the target vehicle for the traffic safety type vehicle networking task.

[0115] In another embodiment, the to-be-computed task can also include an entertainment information type task, at this time, the offloading object includes a cloud server and a current RSU side MEC server. For the information entertainment type task, according to the residence duration of the target vehicle in the cross-RSU driving process, the amount of task to be offloaded by each RSU is determined one by one, and based on the corresponding task offloading ratio, the time delay and energy consumption of the target vehicle in the base station coverage range for V2I and V2N cooperative offloading are obtained. Since the information entertainment type business task amount is large, and the time delay tolerance degree is relatively high, this type of computing task can be offloaded across the RSU, and considering the uncertainty of V2V communication in the cross-RSU coverage range, and the limited computing power of the vehicle terminal, etc. Factors, the computing task of the information entertainment type business does not offload the task between V2V, but is completed by the target vehicle, the MEC server on the RSU side and the cloud server, which can ensure the time delay performance while making the distribution of the computing offloading process more reasonable and efficient.

[0116] Specifically, when the to-be-computed task includes an entertainment information type task, in step 220, based on the task offloading strategy and the maximum tolerable time delay, the task offloading ratio of the offloading object is calculated, including:

[0117] Step 223, acquiring the number of target road side units under the current base station coverage of the target vehicle, and the remaining task amount of the to-be-computed task; the target road side unit is a road side unit behind the current road side unit along the driving direction of the target vehicle under the current base station coverage;

[0118] In step 224, based on the energy consumption optimization principle, the third task offloading ratio of the cloud server is calculated with the maximum tolerable delay as a constraint, and the fourth task offloading ratio of the edge server corresponding to the current road side unit is calculated according to the number of target road side units of the target vehicle and the remaining task quantity of the task to be calculated.

[0119] Firstly, the number of target road side units of the target vehicle under the current base station coverage and the remaining task quantity of the task to be calculated are obtained, and the target road side unit is the road side unit after the current road side unit along the driving direction of the target vehicle under the current base station coverage. That is, during the driving of the target vehicle across the RSU, the task offloading ratio of the MEC server on the current RSU side is calculated according to the number of the current RSU and the subsequent RSU and the remaining task quantity of the task to be calculated with the maximum tolerable delay as a constraint based on the energy consumption optimization principle, to obtain the fourth task offloading ratio, and the task offloading ratio of the cloud server is calculated to obtain the third task offloading ratio.

[0120] Further, the third offloading delay corresponding to the third task offloading ratio is the sum of the third transmission delay and the fourth calculation delay, and the fourth offloading delay corresponding to the fourth task offloading ratio is the sum of the fourth transmission delay and the fifth calculation delay. The third transmission delay is the delay of data transmission between the target vehicle and the current base station and between the current base station and the cloud server, and the fourth calculation delay is the delay required for the cloud server to calculate the task quantity of the third task offloading ratio. The fourth transmission delay is the delay of data transmission between the target vehicle and the edge server corresponding to the current road side unit, and the fifth calculation delay is the delay required for the edge server corresponding to the current road side unit to calculate the task quantity of the fourth task offloading ratio. The total task delay corresponding to the entertainment information task in the task to be calculated is the maximum value among the sixth calculation delay, the third offloading delay and the fourth offloading delay, and the sixth calculation delay is the delay required for local calculation of the target vehicle. The task energy consumption corresponding to the entertainment information task in the task to be calculated is the sum of the third offloading energy consumption, the fourth offloading energy consumption and the second calculation energy consumption, the third offloading energy consumption is the energy consumption corresponding to the third offloading delay, the fourth offloading energy consumption is the energy consumption corresponding to the fourth offloading delay, and the second calculation energy consumption is the energy consumption corresponding to the sixth calculation delay.

[0121] For the entertainment information task, since the data quantity is large and the delay tolerance degree is relatively high, this type of computing task can be offloaded across the RSU and can be completed by the target vehicle, the MEC server on the RSU side and the cloud server. For the The computing task can also be divided into several main parts, i.e., a target vehicle local computing part, a part offloaded to the MEC server computing of each RSU side, and a part offloaded to the cloud server computing. In terms of task latency, the latency involved in collaborative computing also includes computing latency, transmission latency, and feedback latency. However, unlike traffic safety tasks, in addition to the latency generated by the target vehicle performing local computing during cross-RSU travel and the latency of offloading the computing task to the MEC server of the RSU side for computing, the latency of uploading part of the task data to the cloud server for computing is also involved. The process of offloading the computing task to the cloud server is to first upload the task data to the base station through V2N, and then transmit the task data to the cloud server for computing through the optical fiber backbone network by the base station. The computing latency of offloading the task to the cloud includes the communication latency of data transmission between the task vehicle and the base station, the latency of data transmission between the base station and the cloud server through the optical fiber backbone network, and the computing latency of the cloud server. Here, the feedback latency of the computing results of the MEC server and the cloud after computing can also be ignored.

[0122] Similarly, define wherein and represent the task vehicle offloading ratio of the MEC server and the cloud server of the RSU side. Further, the task vehicle the offloading and computing process of the information entertainment task produces the latency and energy consumption as follows:

[0123] First, the latency and energy consumption generated by the task vehicle performing local computing are represented as:

[0124] ; (20)

[0125] ; (21)

[0126] wherein is the computing capability of the task vehicle , is the CPU energy coefficient of the task vehicle .

[0127] The latency and energy consumption generated by the task vehicle offloading part of the task to the MEC server of the current RSU side for computing during cross-RSU travel include the communication latency and energy consumption of the task vehicle uploading task data when passing through each RSU, and the latency and energy consumption of the MEC server of each RSU side performing task computing.

[0128] It should be noted that the task vehicle In the process of crossing the RSU, the MEC server on each RSU side will feed back the task data and the calculation result of the completed calculation in the current MEC server to the task vehicle before the vehicle leaves the current RSU coverage At the same time, the task vehicle will record the total amount of tasks that have been completed in advance, and when entering the coverage area of the next RSU, the remaining tasks will be uploaded to the MEC server on the next RSU side for calculation. In this process, through information sharing between the MEC servers on the RSU side, the maximum data amount of the task vehicle performing task calculation in the current RSU coverage is dynamically controlled to ensure that the task vehicle unloads the task amount in each RSU can complete the calculation before leaving the coverage range of the RSU, thereby avoiding task interruption until the total task amount unloaded to the MEC server is calculated.

[0129] Suppose that the calculation task is generated in the coverage range of the first RSU and is completed by the MEC server on the first RSU side, the task vehicle is relative to the abscissa of the coverage range of the first RSU , and the task vehicle continuously performs V2I-based task offloading in the process of crossing the RSU, then the maximum tolerable delay of the task vehicle performing task calculation in is the residence time of the task vehicle in , which is shown in the following formula 22:

[0130] ; (22)

[0131] wherein is the speed of the task vehicle in .

[0132] Further, the delay and energy consumption of the task vehicle in can be expressed as:

[0133] ; (23)

[0134] ​​​ ; (24)

[0135] wherein, is the task vehicle the task offloading ratio corresponding to the task amount completed in , and is the task vehicle the data transmission rate when uploading the task data to the RSUk. and are the computing resource and the computing capability of the MEC server respectively, is the CPU energy coefficient of the MEC server.

[0136] According to the formula 22 and the formula 23, the computing task amount completed by the task vehicle in respectively can be determined as follows:

[0137] ; (25)

[0138] Based on the above analysis, the task amount offloaded to the corresponding MEC server for computing by the task vehicle in the coverage range of the last RSU can be obtained as follows:

[0139] ; (26)

[0140] That is, for the entertainment information type task, firstly, the task offloading ratio offloaded to the RSU side MEC server and the cloud server is determined, and for the task amount offloaded to the RSU side MEC server, the task offloading ratio of the current RSU side MEC server is further determined when the task vehicle drives into the coverage range of each RSU, so as to obtain the task offloading ratio of each RSU side MEC server. Further, the delay and the energy consumption of the task vehicle in for completing the computing task can be represented as follows:

[0141] = = ; (27)

[0142] ; (28)

[0143] wherein, is the task vehicle the data transmission rate when uploading the task data to the RSU .

[0144] In summary, the task vehicle The latency and energy consumption of the MEC server on the RSU side during the task vehicle's travel across the RSU are:

[0145] ; (29)

[0146] ; (30)

[0147] The latency and energy consumption of the cloud server during the partial task offloading are: The communication latency between the task vehicle and the base station, the data transmission latency between the base station and the cloud server through the fiber backbone network, and the computing latency of the cloud server.

[0148] Under the V2N communication mode, the task vehicle The data transmission rate of the task vehicle uploading task data to the base station is:

[0149] ; (31)

[0150] wherein, The bandwidth of the task vehicle communicating with the base station through V2N, The channel gain between the task vehicle and the base station.

[0151] After the base station receives the task data uploaded by the task vehicle , it transmits the task data to the cloud server through the fiber backbone network, and the cloud server feeds back the result to the task vehicle after the task computation is completed. Ignoring the time of the computation result feedback, the latency and energy consumption required for the partial task offloading are:

[0152] ; (32)

[0153] ; (33)

[0154] In formula 32 and formula 33, The data transmission rate of the task vehicle uploading task data to the base station, The base station transmission power, The latency of the base station transmitting data to the cloud server through the fiber backbone network per unit data; The computing resources and computing capacity of the cloud server are and respectively, The CPU energy coefficient of the cloud server.

[0155] For entertainment-related tasks, the task vehicle... For the task The time delay and energy consumption generated during the unloading process are as follows:

[0156] (34)

[0157] + + (35)

[0158] It should be noted that, because the vehicle needs to upload some tasks to the cloud for auxiliary calculations via base stations while traveling across RSUs, in order to ensure that this data is not interrupted during the calculation process, the task vehicle... The task needs to be unloaded before leaving the coverage area of ​​the base station. Therefore, for infotainment tasks, the upper limit of the delay required for the vehicle to calculate the unloading time is specified as the vehicle's dwell time within the cell covered by the current base station, i.e.:

[0159] (36)

[0160] in, This represents the number of RSUs currently covered by the base station.

[0161] In some embodiments, the task to be computed may include both traffic safety tasks and entertainment information tasks. Therefore, when unloading the task, it is necessary to coordinate the traffic safety tasks and entertainment information tasks. On the basis of meeting the latency constraints of the computing tasks, an energy consumption optimization model for the task vehicle should be constructed to minimize the energy consumption of the task vehicle in completing the unloading of the computing tasks.

[0162] Since traffic safety tasks and infotainment tasks have different unloading targets, a binary parameter is defined to unify the two. , means as follows:

[0163] (37)

[0164] Based on this, for Mission vehicles The latency and energy consumption required to complete the unloading of the computing task can be converted into:

[0165] (38)

[0166] (39)

[0167] Furthermore, in order to enable the mission vehicles When the delay constraint is met, the energy consumption is minimized, and the task offloading of the Internet of Vehicles can be modeled as an energy optimization problem, as shown in the following formulas:

[0168] P1: ; (40)

[0169] St. ; (41)

[0170] ; (42)

[0171] ; (43)

[0172] ; (44)

[0173] ; (45)

[0174] ; (46)

[0175] ; (47)

[0176] ; (48)

[0177] ; (49)

[0178] ; (50)

[0179] In the above formulas 40-50, formula 41 indicates that the task offloading delay cannot exceed the maximum tolerable delay of the computing task, and formulas 42-45 limit the computing time of the tasks to be calculated by the task vehicle When the local, auxiliary vehicle, MEC server on the RSU side and cloud server complete the offloading cooperatively, the delay of each part cannot exceed the residence time of the task vehicle in the current RSU coverage.

[0180] Based on the above analysis and the energy consumption optimization model shown in formulas 40-50, when the to-be-computed task is obtained, the residence duration of the task vehicle in the current RSU coverage range is determined according to the motion characteristics of the task vehicle, so as to determine the maximum tolerable delay. With the maximum tolerable delay as a constraint, the task offloading ratio is solved according to the energy consumption optimization principle according to the above formulas 40-50, and the task offloading ratio of each offloading object is calculated while selecting the offloading object, so that the to-be-computed task is offloaded to each offloading object according to the task offloading ratio, and the task computation is completed by the task vehicle, the auxiliary vehicle, the RSU-side MEC server and the cloud server. For each part of the delay and energy consumption shown in formulas 40-50, the calculation is performed according to the manner shown in the above formulas 1-39 based on the computing resources, computing capacity, CPU capacity coefficient, communication bandwidth, data transmission rate and channel gain of the target vehicle, the auxiliary vehicle SU-side MEC server and the cloud server.

[0181] In the embodiment, by constructing the energy consumption optimization model, the offloading computation of different types of vehicle networking tasks can be planned as a whole, and the task offloading ratio with optimal energy consumption is solved based on the delay constraint, so that the task offloading is ensured while the task offloading energy consumption is reduced.

[0182] The vehicle networking task offloading device provided by the embodiment of the application is described below, and the vehicle networking task offloading device described below can be correspondingly referred to the vehicle networking task offloading method described above.

[0183] Referring to Figure 3 The vehicle networking task offloading device provided by the embodiment of the application includes:

[0184] The strategy determination module 10 is configured to determine a task offloading strategy corresponding to the to-be-computed task according to the task type of the to-be-computed task of the target vehicle, and the task offloading strategy includes an offloading object of the to-be-computed task.

[0185] The ratio calculation module 20 is configured to obtain the motion characteristics of the target vehicle, and calculate the task offloading ratio of the offloading object based on the task offloading strategy and the motion characteristics.

[0186] The task offloading module 30 is configured to offload the to-be-computed task to the offloading object according to the task offloading ratio.

[0187] In one embodiment, the ratio calculation module 20 is further configured to:

[0188] calculate a target residence duration of the target vehicle in the current road side unit according to the motion characteristics;

[0189] determine a maximum tolerable time delay of the to-be-computed task according to the target residence duration, and calculate a task offloading ratio of the offloading object based on the task offloading strategy and the maximum tolerable time delay.

[0190] In one embodiment, if the to-be-computed task includes a traffic safety task, the offloading object includes an auxiliary vehicle and an edge server corresponding to the current road side unit; the ratio calculation module 20 is further configured to:

[0191] obtain a candidate auxiliary vehicle within a preset distance range around the target vehicle, and screen the candidate auxiliary vehicle based on a preset mobility constraint condition to obtain an auxiliary vehicle of the target vehicle; wherein the candidate auxiliary vehicle and the target vehicle are both within a coverage range of the current road side unit, and the mobility constraint condition includes that a residence duration within the current road side unit is greater than or equal to the target residence duration, and a relative distance with the target vehicle within the target residence duration is less than the preset distance;

[0192] based on an energy consumption optimization principle, calculate a first task offloading ratio of the auxiliary vehicle and a second task offloading ratio of the edge server corresponding to the current road side unit, with the maximum tolerable time delay as a constraint.

[0193] In one embodiment, a first offloading time delay corresponding to the first task offloading ratio is a sum of a first transmission time delay and a first computation time delay, and a second offloading time delay corresponding to the second task offloading ratio is a sum of a second transmission time delay and a second computation time delay.

[0194] The first transmission time delay is a time delay of data transmission between the target vehicle and the auxiliary vehicle, and the first computation time delay is a time delay required for the auxiliary vehicle to compute a task amount of the first task offloading ratio;

[0195] The second transmission time delay is a time delay of data transmission between the target vehicle and the edge server corresponding to the current road side unit, and the second computation time delay is a time delay required for the edge server corresponding to the current road side unit to compute a task amount of the second task offloading ratio;

[0196] A total task time delay corresponding to the traffic safety task in the to-be-computed task is a maximum value among a third computation time delay, the first offloading time delay and the second offloading time delay, and the third computation time delay is a time delay required for local computation of the target vehicle;

[0197] The task energy consumption corresponding to the traffic safety type task in the to-be-computed task is the sum of a first offloading energy consumption, a second offloading energy consumption, and a first computation energy consumption. The first offloading energy consumption is energy consumption corresponding to the first offloading time delay. The second offloading energy consumption is energy consumption corresponding to the second offloading time delay. The first computation energy consumption is energy consumption corresponding to the third computation time delay.

[0198] In one embodiment, if the to-be-computed task includes an entertainment information type task, the offloading object includes a cloud server and an edge server corresponding to the current road side unit; and the proportion calculation module 20 is further configured to:

[0199] obtain a quantity of target road side units under a target base station currently covered by the target vehicle and a remaining task amount of the to-be-computed task. The target road side units are road side units after the current road side unit along a driving direction of the target vehicle under the current base station coverage.

[0200] based on an energy consumption optimization principle, calculate a third task offloading proportion of the cloud server, and calculate a fourth task offloading proportion of the edge server corresponding to the current road side unit according to the quantity of target road side units and the remaining task amount, with the maximum tolerable time delay as a constraint.

[0201] In one embodiment, a third offloading time delay corresponding to the third task offloading proportion is the sum of a third transmission time delay and a fourth computation time delay. A fourth task offloading time delay corresponding to the fourth task offloading proportion is the sum of a fourth transmission time delay and a fifth computation time delay.

[0202] The third transmission time delay is a time delay for data transmission between the target vehicle and the current base station and between the current base station and the cloud server. The fourth computation time delay is a time delay required for the cloud server to compute a task amount of the third task offloading proportion.

[0203] The fourth transmission time delay is a time delay for data transmission between the target vehicle and the edge server corresponding to the current road side unit. The fifth computation time delay is a time delay required for the edge server corresponding to the current road side unit to compute a task amount of the fourth task offloading proportion.

[0204] A total task time delay corresponding to the entertainment information type task in the to-be-computed task is the maximum value among a sixth computation time delay, the third offloading time delay, and the fourth offloading time delay. The sixth computation time delay is a time delay required for local computation of the target vehicle.

[0205] The energy consumption of the entertainment information type task in the task to be calculated is the sum of the third unloading energy consumption, the fourth unloading energy consumption, and the second calculation energy consumption. The third unloading energy consumption is the energy consumption corresponding to the third unloading delay, the fourth unloading energy consumption is the energy consumption corresponding to the fourth unloading delay, and the second calculation energy consumption is the energy consumption corresponding to the sixth calculation delay.

[0206] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communications interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute the steps of the vehicle networking task offloading method, which includes:

[0207] Based on the task type of the task to be calculated for the target vehicle, a task unloading strategy corresponding to the task to be calculated is determined; the task unloading strategy includes the unloading object of the task to be calculated.

[0208] The motion characteristics of the target vehicle are obtained, and the task unloading ratio of the unloading object is calculated based on the task unloading strategy and the motion characteristics.

[0209] The task to be calculated is unloaded to the unloading object according to the task unloading ratio.

[0210] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0211] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the vehicle networking task unloading method provided in the above embodiments, the method including:

[0212] Based on the task type of the task to be calculated for the target vehicle, a task unloading strategy corresponding to the task to be calculated is determined; the task unloading strategy includes the unloading object of the task to be calculated.

[0213] The motion characteristics of the target vehicle are obtained, and the task unloading ratio of the unloading object is calculated based on the task unloading strategy and the motion characteristics.

[0214] The task to be calculated is unloaded to the unloading object according to the task unloading ratio.

[0215] In another aspect, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the vehicle networking task offloading method provided in the above embodiments, the method comprising:

[0216] Based on the task type of the task to be calculated for the target vehicle, a task unloading strategy corresponding to the task to be calculated is determined; the task unloading strategy includes the unloading object of the task to be calculated.

[0217] The motion characteristics of the target vehicle are obtained, and the task unloading ratio of the unloading object is calculated based on the task unloading strategy and the motion characteristics.

[0218] The task to be calculated is unloaded to the unloading object according to the task unloading ratio.

[0219] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0220] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0221] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for unloading tasks in a vehicle-to-everything (V2X) network, characterized in that, include: Based on the task type of the task to be calculated for the target vehicle, determine the task unloading strategy corresponding to the task to be calculated. The task unloading strategy includes the unloading objects of the task to be calculated; The motion characteristics of the target vehicle are obtained, and the target dwell time of the target vehicle in the current roadside unit is calculated based on the motion characteristics. The maximum tolerable latency of the task to be calculated is determined based on the target dwell time, and the task offloading ratio of the offloading object is calculated based on the task offloading strategy and the maximum tolerable latency. The task to be calculated is unloaded to the unloading object according to the task unloading ratio; If the task to be calculated includes traffic safety tasks, the offloading objects include auxiliary vehicles and the edge server corresponding to the current roadside unit; the calculation of the task offloading ratio of the offloading objects based on the task offloading strategy and the maximum tolerable latency includes: Candidate auxiliary vehicles within a preset distance range around the target vehicle are acquired, and the candidate auxiliary vehicles are filtered based on preset mobility constraints to obtain the auxiliary vehicle of the target vehicle; wherein, both the candidate auxiliary vehicles and the target vehicle are within the coverage area of ​​the current roadside unit, and the mobility constraints include that the dwell time in the current roadside unit is greater than or equal to the target dwell time, and the relative distance between the candidate auxiliary vehicles and the target vehicle within the target dwell time is less than the preset distance; Using the maximum tolerable latency as a constraint and based on the principle of optimal energy consumption, the first task offloading ratio of the auxiliary vehicle and the second task offloading ratio of the edge server corresponding to the current roadside unit are calculated.

2. The vehicle networking task offloading method according to claim 1, characterized in that, The first unloading delay corresponding to the first task unloading ratio is the sum of the first transmission delay and the first calculation delay, and the second unloading delay corresponding to the second task unloading ratio is the sum of the second transmission delay and the second calculation delay; The first transmission delay is the delay for data transmission between the target vehicle and the auxiliary vehicle, and the first calculation delay is the delay required for the auxiliary vehicle to calculate the workload of the first task unloading ratio. The second transmission delay is the data transmission delay between the target vehicle and the edge server corresponding to the current roadside unit, and the second calculation delay is the delay required for the edge server corresponding to the current roadside unit to calculate the task load of the second task offloading ratio. The total task latency corresponding to the traffic safety category task in the task to be calculated is the maximum value among the third calculation latency, the first unloading latency, and the second unloading latency, where the third calculation latency is the latency required for the target vehicle to perform local calculations. The energy consumption of the traffic safety task in the task to be calculated is the sum of the first unloading energy consumption, the second unloading energy consumption, and the first calculation energy consumption. The first unloading energy consumption is the energy consumption corresponding to the first unloading delay, the second unloading energy consumption is the energy consumption corresponding to the second unloading delay, and the first calculation energy consumption is the energy consumption corresponding to the third calculation delay.

3. The vehicle networking task unloading method according to claim 1, characterized in that, If the task to be calculated includes entertainment information tasks, the unloading objects include cloud servers and edge servers corresponding to the current roadside unit; the calculation of the task unloading ratio of the unloading objects based on the task unloading strategy and the maximum tolerable latency includes: The number of target roadside units under the current base station coverage of the target vehicle and the remaining task volume of the task to be calculated are obtained; the target roadside unit is the roadside unit following the current roadside unit along the driving direction of the target vehicle under the current base station coverage. Based on the maximum tolerable latency as a constraint and the principle of optimal energy consumption, the third task offloading ratio of the cloud server is calculated, and the fourth task offloading ratio of the edge server corresponding to the current roadside unit is calculated based on the number of target roadside units and the remaining task volume.

4. The vehicle networking task unloading method according to claim 3, characterized in that, The third unloading delay corresponding to the third task unloading ratio is the sum of the third transmission delay and the fourth calculation delay, and the fourth unloading delay corresponding to the fourth task unloading ratio is the sum of the fourth transmission delay and the fifth calculation delay. The third transmission delay is the data transmission delay between the target vehicle and the current base station, and between the current base station and the cloud server; the fourth calculation delay is the delay required for the cloud server to calculate the workload of the third task offloading ratio. The fourth transmission delay is the data transmission delay between the target vehicle and the edge server corresponding to the current roadside unit, and the fifth calculation delay is the delay required for the edge server corresponding to the current roadside unit to calculate the task load of the fourth task unloading ratio. The total task latency corresponding to the entertainment information type task in the task to be calculated is the maximum value among the sixth calculation latency, the third unloading latency, and the fourth unloading latency, where the sixth calculation latency is the latency required for local calculation by the target vehicle. The energy consumption of the entertainment information type task in the task to be calculated is the sum of the third unloading energy consumption, the fourth unloading energy consumption, and the second calculation energy consumption. The third unloading energy consumption is the energy consumption corresponding to the third unloading delay, the fourth unloading energy consumption is the energy consumption corresponding to the fourth unloading delay, and the second calculation energy consumption is the energy consumption corresponding to the sixth calculation delay.

5. A vehicle networking task offloading device, characterized in that, include: The strategy determination module is used to determine the task unloading strategy corresponding to the task to be calculated based on the task type of the task to be calculated for the target vehicle. The task unloading strategy includes the unloading objects of the task to be calculated; The ratio calculation module is used to obtain the motion characteristics of the target vehicle and calculate the target dwell time of the target vehicle in the current roadside unit based on the motion characteristics. The maximum tolerable latency of the task to be calculated is determined based on the target dwell time, and the task offloading ratio of the offloading object is calculated based on the task offloading strategy and the maximum tolerable latency. The task unloading module is used to unload the task to be calculated to the unloading object according to the task unloading ratio. If the task to be calculated includes traffic safety tasks, and the unloading objects include auxiliary vehicles and the edge server corresponding to the current roadside unit, the ratio calculation module is specifically used for: Candidate auxiliary vehicles within a preset distance range around the target vehicle are acquired, and the candidate auxiliary vehicles are filtered based on preset mobility constraints to obtain the auxiliary vehicle of the target vehicle; wherein, both the candidate auxiliary vehicles and the target vehicle are within the coverage area of ​​the current roadside unit, and the mobility constraints include that the dwell time in the current roadside unit is greater than or equal to the target dwell time, and the relative distance between the candidate auxiliary vehicles and the target vehicle within the target dwell time is less than the preset distance; Using the maximum tolerable latency as a constraint and based on the principle of optimal energy consumption, the first task offloading ratio of the auxiliary vehicle and the second task offloading ratio of the edge server corresponding to the current roadside unit are calculated.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the vehicle networking task unloading method as described in any one of claims 1 to 5.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the vehicle networking task unloading method as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the vehicle networking task unloading method as described in any one of claims 1 to 5.

Images