Space-time crowdsourcing multi-objective optimization task allocation method and system based on intelligent vehicle

Abstract

The invention relates to the technical field of task allocation, in particular to a space-time crowdsourcing multi-objective optimization task allocation method and system based on an intelligent vehicle. The method comprises the steps that attribute sets of crowdsourcing vehicles and crowdsourcing tasks are acquired respectively, the attribute set of the crowdsourcing vehicles comprises social attributes, and the attribute set of the crowdsourcing tasks comprises space-time attributes and position information; performing trajectory prediction on the crowdsourcing vehicles according to the social attributes to obtain predicted trajectories of the crowdsourcing vehicles; formulating a task allocation strategy based on the prediction trajectory, the space-time attribute and the position information; and dynamically adjusting a task allocation strategy by using a multi-objective optimization algorithm to obtain a task allocation optimal solution. According to the method, crowdsourcing task allocation optimization can be realized.

Description

Technical Field

[0001] This application relates to the technical field of task allocation, and in particular, to a spatio-temporal crowdsourcing multi-objective optimization task allocation method and system based on intelligent vehicles. Background Art

[0002] With the rapid progress of mobile devices such as smart phones, mobile crowdsourcing sensing has become an efficient and cost-effective sensing paradigm. Intelligent vehicles upload various sensing information such as air quality, traffic conditions, and environmental images in real time through their devices. In this case, task allocation usually adopts two modes: one is that intelligent vehicles independently select tasks, but this method often leads to unbalanced allocation; the other is that intelligent vehicle companies actively select vehicles to perform sensing tasks and perform dynamic task allocation according to task characteristics and vehicle capabilities.

[0003] When performing tasks, intelligent vehicles usually integrate them into their daily activities to minimize interference with their daily trajectories and, in return, obtain corresponding rewards or incentives. Therefore, when designing dynamic task allocation strategies, intelligent vehicle companies must consider the location information and status changes of tasks and workers in real time to ensure the effective allocation of tasks.

[0004] With the development of technology, more and more vehicles participate in the crowdsourcing process. These vehicles move at different times and locations, gradually expanding the coverage of crowdsourcing services. Therefore, to effectively improve the efficiency and utility of dynamic task allocation, the following challenges still exist: in opportunistic crowdsourcing scenarios, the trajectories of vehicles are uncertain and tend to select tasks that match their daily routes. If task allocation does not consider the trajectories of vehicles, it may lead to increased company costs and reduced task completion rates. As the number of vehicles increases, the service scope continues to expand. Although the existing task allocation process ensures time constraints, it is difficult to meet the needs of remote tasks and cannot dynamically plan the activity paths of vehicles. Single-objective task allocation schemes limit the scope of solutions, are prone to falling into local optimal solutions, and are difficult to maximize long-term social welfare. Vehicles do not stay still waiting for tasks during dynamic movement, and their movement characteristics will affect task allocation results. Existing schemes have not been personalized for vehicle movement. The current task allocation scheme has no clear order, resulting in urgent tasks not being processed in a timely manner and affecting the responsiveness of task allocation. Summary of the Invention

[0005] To solve the problem that it is difficult to achieve optimal task allocation in the prior art, this application provides a spatio-temporal crowdsourcing multi-objective optimization task allocation method and system based on intelligent vehicles.

[0006] In a first aspect, this application provides a spatio-temporal crowdsourcing multi-objective optimization task allocation method based on intelligent vehicles, adopting the following technical solutions: A spatio-temporal crowdsourcing multi-objective optimization task allocation method based on intelligent vehicles, comprising: Obtain the attribute sets of crowdsourcing vehicles and crowdsourcing tasks respectively, where the attribute set of the crowdsourcing vehicles includes social attributes, and the attribute set of the crowdsourcing tasks includes spatio-temporal attributes and location information; Perform trajectory prediction on the crowdsourcing vehicles according to the social attributes to obtain the predicted trajectories of the crowdsourcing vehicles; Formulate a task allocation strategy based on the predicted trajectories, the spatio-temporal attributes, and the location information; Dynamically adjust the task allocation strategy using a multi-objective optimization algorithm to obtain the optimal solution for task allocation.

[0007] By adopting the above technical solution, comprehensively considering the social attributes of crowdsourcing vehicles and the spatio-temporal attributes and location information of crowdsourcing tasks, vehicle trajectory prediction is realized, and a task allocation strategy is formulated accordingly. Then, the strategy is dynamically adjusted using a multi-objective optimization algorithm to achieve the optimization of task allocation, effectively improving the efficiency and effect of crowdsourcing task execution.

[0008] In a preferred example of the present application, it can be further configured that: the dynamically adjusting the task allocation strategy using a multi-objective optimization algorithm to obtain the optimal solution for task allocation includes: Define optimization objectives and constraint conditions, where the optimization objectives include task service quality and vehicle company cost; Based on the optimization objectives and the constraint conditions, dynamically adjust the task allocation strategy using a MOEA multi-objective optimization algorithm based on PBI decomposition to obtain the optimal solution for task allocation.

[0009] By adopting the above technical solution, defining optimization objectives including task service quality and vehicle company cost and corresponding constraint conditions, and dynamically adjusting the task allocation strategy using a MOEA multi-objective optimization algorithm based on PBI decomposition can effectively balance service quality and cost, achieve the optimization of task allocation, and improve the overall operation efficiency and economic benefits.

[0010] In a preferred example of the present application, it can be further configured that: the task service quality is: Q = U w × r ek × f(t); where U w represents the vehicle's willingness to execute the task; r ek represents the vehicle's ability to complete the task; f(t) is a time function used to calculate the impact of task completion time on the task service quality; where j represents the vehicle number; d(l wj , ltj ) represents the vehicle position l wj and the task position l tj The distance between them; D max is a preset maximum distance threshold.

[0011] By adopting the above technical solution, comprehensively considering the vehicle's willingness to execute the task (based on the distance between the vehicle and the task position), the vehicle's ability to complete the task, and the impact of the task completion time on the service quality, the task service quality can be evaluated more accurately, thereby optimizing the task allocation strategy, ensuring that the task is executed by the most suitable vehicle, and improving the task completion efficiency and quality.

[0012] In a preferred example of the present application, it can be further configured that: the vehicle company cost is: where COST is the vehicle company cost; j represents the vehicle number; m is the total number of crowdsourcing vehicles executing crowdsourcing tasks; a 0 is the basic reward, indicating the basic remuneration obtained by each vehicle after completing the task; d(l wj ,l tj ) represents the vehicle position l wj and the task position l tj The distance between them; a is the scheduling reward coefficient per unit distance; c j is the cumulative reward for completing the task; β is the task difficulty coefficient, reflecting the influence degree of task difficulty on vehicle remuneration; d j is the task difficulty level; cost p is the fixed cost of the vehicle company.

[0013] By adopting the above technical solution, comprehensively considering the basic reward, distance-related scheduling reward, cumulative reward after task completion, the influence of task difficulty on remuneration, and the fixed cost of the vehicle company, accurately calculating the vehicle company cost helps to balance cost control and task execution efficiency during the task allocation process and achieve the maximization of economic benefits.

[0014] In a preferred example of the present application, it can be further configured that: the defined constraint conditions include: Define the crowdsourcing vehicle time constraint, crowdsourcing task time constraint, range constraint, ability constraint, and invariant constraint; the crowdsourcing vehicles include multiple vehicles; the crowdsourcing tasks include multiple tasks; Among them, the crowdsourcing vehicle time constraint is that each vehicle in the crowdsourcing vehicles is assigned a task after entering the vehicle company; The time constraint of the crowdsourcing tasks is as follows: Each task in the crowdsourcing tasks is assigned after entering the vehicle company, and each task has a corresponding time range, and the assigned task is required to be completed within the corresponding time range. The range constraint is as follows: Each vehicle in the crowdsourcing vehicles executes the assigned tasks within the corresponding daily route range. The capacity constraint is as follows: The number of tasks assigned to each vehicle in the crowdsourcing vehicles does not exceed the vehicle's capacity range. The invariant constraint is as follows: Any vehicle in the crowdsourcing vehicles forms a matching pair with the assigned task, and the matching pair does not change.

[0015] By adopting the above technical solution, detailed constraint conditions including the time constraint of the crowdsourcing vehicles, the time constraint of the crowdsourcing tasks, the range constraint, the capacity constraint, and the invariant constraint are defined, which can ensure the rationality and feasibility of the task assignment process, avoid vehicles from executing tasks overtime, beyond the scope, and overloading, and at the same time ensure the stable matching of tasks and vehicles, improving the reliability and efficiency of task execution.

[0016] In a preferred example, the present application can be further configured as follows: The trajectory prediction of the crowdsourcing vehicles according to the social attributes to obtain the predicted trajectories of the crowdsourcing vehicles includes: Performing a trajectory prediction step on each vehicle in the crowdsourcing vehicles according to the social attributes. Among them, the trajectory prediction step includes: performing spatio-temporal graph modeling based on the social attributes of the current vehicle; using a spatio-temporal graph neural network to extract vehicle trajectory features from the spatio-temporal graph modeling. Using a trajectory prediction convolutional neural network to predict the vehicle trajectory features through a trajectory prediction formula to obtain the predicted trajectory of the current vehicle. The trajectory prediction formula is: The value range of t is [1, T], representing each time point in the time series; p nt represents the actual position at time t. represents the predicted position at time t; T is the time length of the entire trajectory.

[0017] By adopting the above technical solution, spatio-temporal graph modeling is performed based on the social attributes of the crowdsourcing vehicles, and a spatio-temporal graph neural network and a trajectory prediction convolutional neural network are used to extract and predict vehicle trajectory features, which can accurately predict the future trajectories of the crowdsourcing vehicles, provide a reliable basis for formulating task assignment strategies, and thus improve the accuracy and execution efficiency of task assignment.

[0018] In a preferred example, the present application can be further configured as follows: The attribute set of the crowdsourcing vehicles is: Among them, represents the real-time position at the moment when the vehicle enters the vehicle company at this moment; represents the upper limit of the number of tasks that each vehicle can undertake at the same time; is the reputation value, representing the vehicle's credibility and work quality.

[0019] By adopting the above technical solution, comprehensively considering the social attributes such as the real-time position, task-bearing upper limit, and reputation value of the crowdsourcing vehicle, the ability and reliability of the vehicle can be more comprehensively evaluated, providing accurate data support for subsequent trajectory prediction and task allocation, and ensuring the efficient and high-quality completion of tasks.

[0020] In a preferred example of the present application, it can be further configured that: the attribute set of the crowdsourcing task is: Among them, is the starting position of the task; is the time required to complete the task; is the time when the task enters the vehicle company; C is the task budget; ψ is the task difficulty coefficient.

[0021] By adopting the above technical solution, comprehensively considering the attributes such as the starting position, required time, entry time, budget, and difficulty coefficient of the crowdsourcing task, the characteristics and requirements of the task can be more comprehensively evaluated, providing key information for the formulation of the task allocation strategy, and ensuring that the task can be reasonably and efficiently allocated to the appropriate vehicle, improving the overall task execution efficiency and quality.

[0022] In a preferred example of the present application, it can be further configured that: the social attributes include: the traffic conditions of the road where the crowdsourcing vehicle is located when performing the task and the route planning situation of the mutual influence between multiple vehicles.

[0023] By adopting the above technical solution, incorporating the traffic conditions of the road where the crowdsourcing vehicle is located when performing the task and the route planning situation of the mutual influence between vehicles into the social attributes can more accurately reflect the running state of the vehicle in the actual road environment, providing a more practical basis for trajectory prediction and task allocation, thereby enhancing the flexibility and efficiency of task execution.

[0024] In the second aspect, the present application provides a spatio-temporal crowdsourcing multi-objective optimization task allocation system based on intelligent vehicles, adopting the following technical solution: A spatio-temporal crowdsourcing multi-objective optimization task allocation system based on intelligent vehicles, comprising: An acquisition module, configured to acquire the attribute sets of crowdsourcing vehicles and crowdsourcing tasks respectively, where the attribute set of the crowdsourcing vehicle includes social attributes, and the attribute set of the crowdsourcing task includes spatio-temporal attributes and location information; A prediction module, configured to perform trajectory prediction on the crowdsourcing vehicle according to the social attributes to obtain the predicted trajectory of the crowdsourcing vehicle; An allocation module, configured to formulate a task allocation strategy based on the predicted trajectory, the spatio-temporal attributes, and the location information; an optimization module, configured to dynamically adjust the task allocation strategy by using a multi-objective optimization algorithm to obtain an optimal solution for task allocation.

[0025] In a third aspect, the present application provides an electronic device, adopting the following technical solution: One or more processors; A memory; At least one application program, where at least one application program is stored in the memory and is configured to be executed by at least one processor, and the at least one application program is configured to: execute the spatio-temporal crowdsourcing multi-objective optimization task allocation method based on an intelligent vehicle according to any item in the first aspect.

[0026] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution: A computer-readable storage medium, on which a computer program is stored. When the computer program is executed on a computer, the computer is made to execute the spatio-temporal crowdsourcing multi-objective optimization task allocation method based on an intelligent vehicle according to any item in the first aspect.

[0027] In a fifth aspect, the present application provides a computer program product, adopting the following technical solution: A computer program product, including a computer program. When the computer program is executed by a processor, it implements the spatio-temporal crowdsourcing multi-objective optimization task allocation method based on an intelligent vehicle according to any item in the first aspect.

[0028] In summary, the present application includes the following beneficial technical effects: By comprehensively considering the social attributes of crowdsourcing vehicles, the spatio-temporal attributes and location information of crowdsourcing tasks, the present application realizes vehicle trajectory prediction, formulates a task allocation strategy accordingly, and then dynamically adjusts the strategy by using a multi-objective optimization algorithm to achieve the optimization of task allocation, effectively improving the efficiency and effect of crowdsourcing task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a schematic diagram of a scenario of a spatio-temporal crowdsourcing multi-objective optimization task allocation method based on an intelligent vehicle provided by an embodiment of the present application; Figure 2It is a schematic flowchart of a spatio-temporal crowdsourcing multi-objective optimization task allocation method provided by an embodiment of the present application; Figure 3 It is a comparison diagram of the effects of a spatio-temporal crowdsourcing multi-objective optimization task allocation method provided by an embodiment of the present application; Figure 4 It is a schematic structural diagram of a spatio-temporal crowdsourcing multi-objective optimization task allocation system provided by an embodiment of the present application; Figure 5 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0030] The following will further describe the present application in detail with reference to the Figure 1 - attached Figure 5 drawings.

[0031] This specific embodiment is only an interpretation of the present application and does not limit the present application. After reading this specification, those skilled in the art can make modifications to this embodiment without creative contributions as needed, but as long as they are within the scope of the claims of the present application, they are protected by the patent law.

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts fall within the scope of protection of the present application.

[0033] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0034] It should be noted that in the optional embodiments of the present application, for relevant data such as object information, when the embodiments in the present application are applied to specific products or technologies, object permission or consent is required, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions. That is to say, if the embodiments in the present application involve data related to an object, it needs to be obtained under the authorization and consent of the object, the authorization and consent of relevant departments, and compliance with the relevant laws, regulations, and standards of the country and region. In the embodiments, if personal information is involved, the acquisition of all personal information requires the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained, and the embodiments also need to be implemented under the authorization and consent of the object.

[0035] See Figure 1 , which shows a schematic diagram of the scenario of a spatio-temporal crowdsourcing multi-objective optimization task allocation method based on intelligent vehicles provided by the embodiments of the present application. In the mobile crowdsourcing system, there are mainly three roles: crowdsourcing vehicles, crowdsourcing tasks, and crowdsourcing vehicle companies. The crowdsourcing tasks are issued by requesters, and the requesters need to provide a certain budget for the issued crowdsourcing tasks. The crowdsourcing vehicle company makes a reasonable task allocation according to the crowdsourcing tasks issued by the requesters and the future trajectories of the vehicles. When the crowdsourcing vehicle receives the perception task issued by the vehicle company, it correspondingly executes the perception task, and the vehicle company gives corresponding rewards according to the scheduling of the vehicle's task execution and the vehicle's reputation value.

[0036] The embodiments of the present application provide a spatio-temporal crowdsourcing multi-objective optimization task allocation method based on intelligent vehicles, as Figure 2 shown. In the method provided in the embodiments of the present application, it is executed by an electronic device, and the electronic device can be a server or a terminal device. Among them, the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication methods, and the embodiments of the present application do not limit this. The method includes step S201-step S204, where: S201. Respectively obtain the attribute sets of the crowdsourcing vehicle and the crowdsourcing task. The attribute set of the crowdsourcing vehicle includes social attributes, and the attribute set of the crowdsourcing task includes spatio-temporal attributes and location information.

[0037] Specifically, the attribute set of crowd-sourced vehicles includes the attribute set when each vehicle performs tasks and social attributes. For any vehicle, its attribute set includes: the real-time position of the vehicle at each moment after it enters the vehicle company, the upper limit of the number of tasks that the vehicle can undertake at the same time, and the reputation value, which represents the credit and work quality of the vehicle and can be determined by the vehicle's historical task execution situation. The social attributes of the vehicle include: the traffic conditions of the road where the vehicle is located when performing tasks and the route planning situation of the vehicle's interaction with multiple other vehicles.

[0038] The attribute set of crowd-sourced tasks includes the attribute set of each task. For any task, its attribute set includes: the starting position when the task is released, the time required to complete the task, the time when the task enters the vehicle company, the task budget of the task, and the task difficulty coefficient of the task.

[0039] S202. Perform trajectory prediction on crowd-sourced vehicles according to social attributes to obtain the predicted trajectories of crowd-sourced vehicles.

[0040] Specifically, for any vehicle, perform trajectory prediction on the vehicle based on the social attributes of the vehicle to obtain the predicted trajectory of the vehicle. The trajectory prediction process includes: performing spatio-temporal graph modeling based on the social attributes of the vehicle, using a spatio-temporal graph neural network to extract the vehicle trajectory features of the vehicle from the spatio-temporal graph modeling, and using a trajectory prediction convolutional neural network to predict the vehicle trajectory features through a trajectory prediction formula to obtain the predicted trajectory of the vehicle.

[0041] S203. Develop a task allocation strategy based on the predicted trajectory, spatio-temporal attributes, and position information.

[0042] In the task allocation strategy, a vehicle performs multiple sensing tasks in its daily route for multi-task allocation.

[0043] Among them, before task allocation, tasks can be divided into different priorities, which can include: high priority, medium priority, and low priority. Ensure that high-priority tasks (such as emergency monitoring and emergency response) can be preferentially matched with suitable vehicles. The system automatically adjusts the priority according to the importance and deadline of the task.

[0044] High-priority tasks (such as emergency environmental monitoring and emergency incident handling) are given priority to ensure that these tasks are executed within the shortest possible time. The system monitors the dynamic trajectories of vehicles in real time and matches the vehicles that best meet the task requirements, thus achieving a rapid response to high-priority tasks. For medium- and low-priority tasks, the system allocates them based on the daily movement trajectories of vehicles and the spatio-temporal attributes of the tasks to avoid resource waste. At the same time, the vehicle company dynamically adjusts the priorities during the task allocation process and optimizes the task allocation in real time to cope with changes in the task environment. This mechanism ensures a reasonable allocation of multi-level tasks with limited resources, improving the overall service quality and resource utilization efficiency.

[0045] S204. Dynamically adjust the task allocation strategy using a multi-objective optimization algorithm to obtain the optimal solution for task allocation.

[0046] Specifically, define the optimization objectives and constraints of the multi-objective optimization algorithm. The optimization objectives include task service quality and vehicle company cost, and the constraints include the maximum task load of the vehicle and the task time constraint, and perform multi-objective optimization. The optimal solution for task allocation includes an equilibrium solution for the conflicting indicators of task service quality and vehicle company cost.

[0047] The multi-objective optimization algorithm uses the MOEA multi-objective optimization algorithm based on PBI decomposition. Its main idea is to decompose a multi-objective problem into multiple scalar sub-problems for simultaneous solution. When there are two or more optimization objective functions, it is called multi-objective optimization. The solution of multi-objectives is usually a set of equilibrium solutions. In multi-objective optimization, there are multiple optimization objectives, so there is no longer a single optimal solution. Instead, there is a set of solutions called Pareto optimal solutions or non-dominated solution sets. The Pareto optimal solution is a solution where the optimization objectives cannot be improved without compromising other objectives. Its Pareto solution set is an integrated solution set considering both service quality and cost. Select the appropriate vehicle-task pair according to the different requirements of the requester.

[0048] In this embodiment, by comprehensively considering the social attributes of crowdsourcing vehicles, the spatio-temporal attributes and location information of crowdsourcing tasks, vehicle trajectory prediction is realized, and a task allocation strategy is formulated accordingly. Then, the multi-objective optimization algorithm is used to dynamically adjust the strategy to achieve the optimization of task allocation, effectively improving the efficiency and effect of crowdsourcing task execution.

[0049] A possible implementation manner of the embodiment of this application is to dynamically adjust the task allocation strategy using a multi-objective optimization algorithm to obtain the optimal solution for task allocation, including: Define the optimization objectives and constraints. The optimization objectives include task service quality and vehicle company cost; Based on the optimization objectives and constraints, the MOEA multi-objective optimization algorithm based on PBI decomposition is used to dynamically adjust the task allocation strategy to obtain the optimal solution of task allocation.

[0050] In this embodiment, by defining the optimization objectives including task service quality and vehicle company cost and the corresponding constraints, and using the MOEA multi-objective optimization algorithm based on PBI decomposition to dynamically adjust the task allocation strategy, it is possible to effectively balance service quality and cost, achieve the optimization of task allocation, and improve the overall operation efficiency and economic benefits.

[0051] A possible implementation manner of the embodiment of the present application, the task service quality is: Q = U w × r ek × f(t); Wherein, U w represents the willingness of the vehicle to execute the task; r ek represents the ability of the vehicle to complete the task; f(t) is a time function used to calculate the impact of task completion time on task service quality; Wherein, j represents the vehicle number; d(l wj , l tj ) represents the distance between the vehicle position l wj and the task position l tj ; D max is a preset maximum distance threshold.

[0052] In this embodiment, U w is related to the distance between the task and the vehicle position. The farther the distance, the lower the willingness of the vehicle to execute the task. Therefore, when calculating the willingness of the vehicle to execute the task, it is necessary to predict the movement trajectory of the vehicle. d(l wj , l tj ) can be the Euclidean distance, Manhattan distance or other suitable distance measurement methods between the vehicle position l wj and the task position l tj . This embodiment does not make a limitation. D max is used to normalize the distance.

[0053] In this embodiment, by comprehensively considering the willingness of the vehicle to execute the task (based on the distance between the vehicle and the task position), the ability of the vehicle to complete the task, and the impact of task completion time on service quality, it is possible to more accurately evaluate the task service quality, thereby optimizing the task allocation strategy, ensuring that the task is executed by the most suitable vehicle, and improving the task completion efficiency and quality.

[0054] A possible implementation manner of the embodiment of the present application, the vehicle company cost is: Among them, COST is the cost of the vehicle company; j represents the vehicle number; m is the total number of crowdsourcing vehicles performing crowdsourcing tasks; a 0 is the basic reward, indicating the basic remuneration obtained by each vehicle after completing the task; d(l wj ,l tj ) represents the distance between vehicle location l wj and task location l tj ; a is the scheduling reward coefficient per unit distance; c j is the cumulative reward for completing the task; β is the task difficulty coefficient, reflecting the impact of task difficulty on vehicle remuneration; d j is the task difficulty level; cost p is the fixed cost of the vehicle company.

[0055] In this embodiment, COST is the cost of the vehicle company, representing the total cost generated by the vehicle company during the task allocation process. a represents the additional distance reward obtained by the vehicle for moving to the task location. The farther the distance, the higher the additional distance reward. c j is the cumulative reward for completing the task, which can include additional rewards based on factors such as task quality and time. d j is the task difficulty level, which is an indicator used to measure the complexity of the task and the required skill level. cost p is the fixed cost of the vehicle company, including fixed expenses such as the operating cost and maintenance cost of the vehicle company, which does not change with specific tasks or the number of vehicles.

[0056] By comprehensively considering the basic reward, distance-related scheduling reward, cumulative reward after task completion, the impact of task difficulty on remuneration, and the fixed cost of the vehicle company, this embodiment accurately calculates the cost of the vehicle company, which helps to balance cost control and task execution efficiency during the task allocation process and achieve the maximization of economic benefits.

[0057] A possible implementation manner of the embodiment of the present application defines constraint conditions, including: Define the time constraint of the crowdsourcing vehicle, the time constraint of the crowdsourcing task, the range constraint, the ability constraint, and the invariant constraint; the crowdsourcing vehicle includes multiple vehicles; the crowdsourcing task includes multiple tasks; Among them, the time constraint of the crowdsourcing vehicle is that each vehicle in the crowdsourcing vehicle is assigned a task after entering the vehicle company; The time constraint of the crowdsourcing task is that each task in the crowdsourcing task is assigned after entering the vehicle company, and each task has a corresponding time range, and the assigned task is required to be completed within the corresponding time range; The range constraint is that each vehicle in the crowdsourcing vehicle performs the assigned task within the corresponding daily route range; The capacity constraint is that the number of tasks assigned to each vehicle in the crowdsourcing vehicles does not exceed the vehicle's capacity range; The invariant constraint is that any vehicle in the crowdsourcing vehicles forms a matching pair with the assigned task, and the matching pair does not change.

[0058] In this embodiment, trajectory prediction of the vehicles can be performed to achieve reasonable task allocation so that the vehicles meet the range constraint.

[0059] By defining detailed constraint conditions including the time constraint of crowdsourcing vehicles, the time constraint of crowdsourcing tasks, the range constraint, the capacity constraint, and the invariant constraint, this embodiment can ensure the rationality and feasibility of the task allocation process, avoid vehicles from executing tasks overtime, beyond the range, or overloading, and at the same time ensure the stable matching of tasks and vehicles, improving the reliability and efficiency of task execution.

[0060] A possible implementation manner of the embodiment of the present application is to perform trajectory prediction on the crowdsourcing vehicles according to their social attributes to obtain the predicted trajectories of the crowdsourcing vehicles, including: Performing a trajectory prediction step on each vehicle in the crowdsourcing vehicles according to its social attributes; Among them, the trajectory prediction step includes: performing spatio-temporal graph modeling based on the social attributes of the current vehicle; extracting vehicle trajectory features from the spatio-temporal graph modeling using a spatio-temporal graph neural network; Predicting the vehicle trajectory features through a trajectory prediction formula using a trajectory prediction convolutional neural network to obtain the predicted trajectory of the current vehicle; The trajectory prediction formula is: The value range of t is [1, T], representing each time point in the time series; p nt represents the actual position at time t; represents the predicted position at time t; T is the time length of the entire trajectory.

[0061] In this embodiment, taking the current vehicle as an example to illustrate the trajectory prediction step, spatio-temporal graph modeling is performed according to the social attributes of the current vehicle in the spatio-temporal region, and a spatio-temporal graph neural network (ST-GNN) is used to extract the trajectory features of the current vehicle; based on the vehicle trajectory features, an enhanced trajectory prediction convolutional neural network (TXP-CNN) is adopted to more accurately predict the future trajectories of the crowdsourcing vehicles.

[0062] This embodiment performs spatio-temporal graph modeling based on the social attributes of the crowdsourcing vehicles, and uses a spatio-temporal graph neural network and a trajectory prediction convolutional neural network to extract and predict vehicle trajectory features, which can accurately predict the future trajectories of the crowdsourcing vehicles, provide a reliable basis for formulating task allocation strategies, and thus improve the accuracy and execution efficiency of task allocation.

[0063] A possible implementation manner of the embodiment of the present application is that the attribute set of the crowdsourcing vehicles is: Among them, represents the real-time position at the moment when the vehicle enters the vehicle company This moment; represents the upper limit of the number of tasks that each vehicle can undertake at the same time; is the reputation value, representing the vehicle's credibility and work quality.

[0064] In this embodiment, will be updated as the vehicle moves. The reputation value represents the vehicle's credibility and work quality. A higher reputation value means that the vehicle can provide higher-quality services, and at the same time, it will also result in higher remuneration for the requester.

[0065] This embodiment comprehensively considers the social attributes such as the real-time position of the crowdsourcing vehicle, the upper limit of task bearing, and the reputation value, etc., can more comprehensively evaluate the vehicle's capabilities and reliability, provide accurate data support for subsequent trajectory prediction and task allocation, and ensure the efficient and high-quality completion of tasks.

[0066] A possible implementation manner of the embodiment of the present application, the attribute set of the crowdsourcing task is: Among them, is the starting position of the task; is the time required to complete the task; is the time when the task enters the vehicle company; C is the task budget; ψ is the task difficulty coefficient.

[0067] In this embodiment, is the starting position of the task, that is, a fixed point determined when the task is released. is the time required to complete the task. In order to ensure the quality of task completion, the task must be completed within the specified time. C is the task budget, which is set by the task publisher according to the specific requirements of the task. ψ is the task difficulty coefficient, which depends on the specific requirements of the task and the complexity of the equipment required to complete the task.

[0068] This embodiment can more comprehensively evaluate the characteristics and requirements of the task by comprehensively considering the attributes such as the starting position, required time, entry time, budget, and difficulty coefficient of the crowdsourcing task, provide key information for the formulation of the task allocation strategy, ensure that the task can be reasonably and efficiently allocated to the appropriate vehicle, and improve the overall task execution efficiency and quality.

[0069] A possible implementation manner of the embodiment of the present application, the social attributes include: the traffic conditions of the road where the crowdsourcing vehicle is located when performing the task and the route planning situation of the mutual influence between multiple vehicles.

[0070] In this embodiment, the traffic conditions of the roads where the crowdsourcing vehicles are located during task execution and the route planning situation of the mutual influence between vehicles are incorporated into the social attributes, which can more accurately reflect the running state of the vehicles in the actual road environment, provide a more practical basis for trajectory prediction and task allocation, and thus improve the flexibility and efficiency of task execution.

[0071] The multi-objective optimization task allocation method for spatio-temporal crowdsourcing based on intelligent vehicles provided in this embodiment aims to maximize the quality of service and minimize the costs of the intelligent vehicle company. The focus is on modeling the moving trajectories of the vehicles, representing the vehicle trajectories as spatio-temporal graphs, and using a graph convolutional neural network (GCN) to predict the vehicle trajectories. At the same time, combined with a real-time dynamic adjustment mechanism, an improved algorithm is used to perform multi-objective optimization of task allocation for sensing tasks, and a multi-level task priority strategy is introduced to ensure that high-priority tasks are processed in a timely manner.

[0072] See Figure 3 , which shows a comparison chart of the effects of this embodiment considering comprehensive evaluation indicators and other methods; The present invention predicts the vehicle trajectories in the spatio-temporal region through a real dataset: at the same time, it is optimized through a multi-objective optimization algorithm based on vehicle trajectory prediction. In the experiment, the different experimental effects of the single-objective optimization algorithm and the multi-objective algorithm MOEAD algorithm on the dataset are compared.

[0073] In the experiment, the influence of different numbers of vehicles and the same number of tasks on the algorithm is evaluated. First, the number of tasks is kept unchanged, and the practicability and matching number of the algorithm are evaluated by increasing the number of vehicles.

[0074] Figure 3 It shows the superiority of the algorithm proposed by the present invention compared with other algorithms in the task allocation stage under the comprehensive evaluation indicators of service quality and vehicle company costs on the premise of different numbers of vehicles. Compared with the single-objective static optimization algorithm, the present invention can realistically consider the optimization problems in the real scenario and thus can obtain an equilibrium solution that dynamically considers multiple objectives comprehensively.

[0075] The beneficial effects of a multi-objective optimization task allocation method for spatio-temporal crowdsourcing based on intelligent vehicles provided in this embodiment include: being able to achieve efficient task allocation under different environmental conditions, improving the flexibility and response speed of allocation. Combining the adaptive algorithm of real-time dynamic adjustment and trajectory prediction technology, it realizes more reasonable and flexible task allocation, effectively improving the working efficiency of the vehicles and the service quality of the vehicle company. Introducing a multi-level task priority and intelligent optimization algorithm makes the task allocation more scientific and can meet the needs of different types of tasks. Considering the real-time state of the vehicles and environmental changes comprehensively, optimizing the task allocation strategy, and improving the overall efficiency and reliability of the crowdsourcing system.

[0076] See Figure 4 , which shows a schematic structural diagram of a spatio-temporal crowdsourcing multi-objective optimization task allocation system provided by an embodiment of the present application. The system includes: an acquisition module 401, a prediction module 402, an allocation module 403, and an optimization module 404, where: The acquisition module 401 is configured to respectively acquire the attribute sets of crowdsourcing vehicles and crowdsourcing tasks. The attribute set of crowdsourcing vehicles includes social attributes, and the attribute set of crowdsourcing tasks includes spatio-temporal attributes and location information; The prediction module 402 is configured to perform trajectory prediction on the crowdsourcing vehicles according to the social attributes to obtain the predicted trajectories of the crowdsourcing vehicles; the allocation module 403 is configured to formulate a task allocation strategy based on the predicted trajectories, spatio-temporal attributes, and location information; The optimization module 404 is configured to dynamically adjust the task allocation strategy by using a multi-objective optimization algorithm to obtain an optimal solution for task allocation.

[0077] An embodiment of the present application provides an electronic device, as Figure 5 shown Figure 5 The electronic device 500 shown includes: a processor 501 and a memory 503. Among them, the processor 501 and the memory 503 are connected, such as connected through a bus 502. Optionally, the electronic device 500 may further include a transceiver 504. It should be noted that in actual applications, the transceiver 504 is not limited to one, and the structure of the electronic device 500 does not constitute a limitation to the embodiments of the present application.

[0078] The processor 501 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present application. The processor 501 may also be a combination that implements a computing function, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0079] The bus 502 may include a path for transmitting information among the above components. The bus 502 can be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 502 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 5 it is only represented by a thick line in Figure 5 , but it does not mean that there is only one bus or one type of bus.

[0080] The memory 503 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or it can also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0081] The memory 503 is used to store the application program code for executing the solution of this application, and is controlled by the processor 501 to execute. The processor 501 is used to execute the application program code stored in the memory 503 to implement the content shown in the foregoing embodiments of the method for spatio-temporal crowdsourcing multi-objective optimization task allocation based on intelligent vehicles.

[0082] Figure 5 The illustrated electronic device is only an example and should not impose any limitation on the functions and usage scope of the embodiments of this application.

[0083] The embodiments of this application provide a computer-readable storage medium, on which a computer program is stored. When it runs on a computer, it enables the computer to execute the content shown in the foregoing embodiments of the method for spatio-temporal crowdsourcing multi-objective optimization task allocation based on intelligent vehicles.

[0084] It should be understood that although the steps in the flowchart of the accompanying drawings are shown sequentially in the direction of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless there is a clear indication in this document, there is no strict order restriction for the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0085] An embodiment of the present application provides a computer program product, including a computer program, which when executed by a processor, implements the content shown in the foregoing embodiment of the method for spatio-temporal crowdsourcing multi-objective optimization task allocation based on intelligent vehicles.

[0086] The above are only partial embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A spatiotemporal crowdsourcing multi-objective optimization task allocation method based on intelligent vehicles, characterized in that: include: Acquire attribute sets of crowdsourcing vehicles and crowdsourcing tasks respectively, wherein the attribute set of crowdsourcing vehicles includes social attributes, and the attribute set of crowdsourcing tasks includes spatiotemporal attributes and location information; Predicting the trajectory of the crowdsourced vehicle according to the social attribute to obtain a predicted trajectory of the crowdsourced vehicle; Formulate a task allocation strategy based on the predicted trajectory, the spatiotemporal attributes and the location information; The task allocation strategy is dynamically adjusted using a multi-objective optimization algorithm to obtain the optimal solution for task allocation.

2. The method for spatiotemporal crowdsourcing multi-objective optimization task allocation based on intelligent vehicles according to claim 1 is characterized in that: The method of dynamically adjusting the task allocation strategy using a multi-objective optimization algorithm to obtain an optimal solution for task allocation includes: Defining optimization objectives and constraints, wherein the optimization objectives include mission service quality and vehicle company cost; Based on the optimization goal and the constraint conditions, the task allocation strategy is dynamically adjusted using the MOEA multi-objective optimization algorithm based on PBI decomposition to obtain the optimal solution for task allocation.

3. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 2 is characterized in that: The task service quality is: Q=U w ×r ek ×f(t); Among them, U w Indicates the vehicle's willingness to perform the task; r ek represents the ability of the vehicle to complete the task; f(t) is a time function used to calculate the impact of the task completion time on the service quality of the task; Where j represents the vehicle number; d(l wj ,l tj ) represents the vehicle position l wj With the task location tj The distance between max The preset maximum distance threshold.

4. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 2 is characterized in that: The vehicle company costs are: Where COST is the cost of the vehicle company; j is the vehicle number; m is the total number of crowdsourcing vehicles that perform crowdsourcing tasks; a0 is the basic reward, which means the basic reward obtained by each vehicle after completing the task; d(l wj ,l tj ) represents the vehicle position l wj With the task location tj The distance between them; a is the scheduling reward coefficient per unit distance; c is the scheduling reward coefficient per unit distance j is the cumulative reward for completing the task; β is the task difficulty coefficient, which reflects the impact of task difficulty on vehicle reward; d j is the difficulty level of the task; cost p Fixed costs for the vehicle company.

5. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 2 is characterized in that: The defining of the constraint condition comprises: Defining crowdsourcing vehicle time constraints, crowdsourcing task time constraints, scope constraints, capability constraints and invariant constraints; the crowdsourcing vehicle includes multiple vehicles; the crowdsourcing task includes multiple tasks; The crowdsourcing vehicle time constraint is as follows: each of the crowdsourcing vehicles is assigned a task after entering the vehicle company; The crowdsourcing task time constraint is: each task in the crowdsourcing task is assigned after entering the vehicle company, and each task has a corresponding time range, and the assigned task is required to be completed within the corresponding time range; The range constraint is: each of the crowdsourcing vehicles performs the assigned task within the corresponding daily route range; The capability constraint is that the number of tasks assigned to each vehicle in the crowdsourcing vehicle does not exceed the vehicle capability range; The invariant constraint is that any vehicle in the crowdsourcing vehicle forms a matching pair with the assigned task, and the matching pair does not change.

6. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 1 is characterized in that: The performing trajectory prediction on the crowdsourced vehicles according to the social attributes to obtain the predicted trajectory of the crowdsourced vehicles includes: performing a trajectory prediction step on each of the crowdsourced vehicles according to the social attributes; The trajectory prediction step includes: performing spatiotemporal graph modeling based on the social attributes of the current vehicle; extracting vehicle trajectory features from the spatiotemporal graph modeling using a spatiotemporal graph neural network; Using a trajectory prediction convolutional neural network to predict the vehicle trajectory characteristics through a trajectory prediction formula to obtain a predicted trajectory of the current vehicle; The trajectory prediction formula is: The value range of t is [1, T], which represents each time point in the time series; p nt represents the actual position at time t; represents the predicted position at time t; T is the time length of the entire trajectory.

7. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 1 is characterized in that: The attribute set of the crowdsourced vehicle is: in, Indicates that after the vehicle enters the vehicle company The real-time location at this moment; Indicates the upper limit of the number of tasks each vehicle can undertake at the same time; is the reputation value, which indicates the reputation and work quality of the vehicle.

8. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 1 is characterized in that: The attribute set of the crowdsourcing task is: in, is the starting position of the task; The time required to complete the task; is the time for the task to enter the vehicle company; C is the task budget; ψ is the task difficulty coefficient.

9. The method for allocating multi-objective optimization tasks based on spatiotemporal crowdsourcing of intelligent vehicles according to claim 1 is characterized in that: The social attributes include: the traffic conditions of the road where the crowdsourcing vehicle is located when performing the task and the route planning conditions of the mutual influence between multiple vehicles.

10. A spatiotemporal crowdsourcing multi-objective optimization task allocation system based on intelligent vehicles, characterized in that: include: An acquisition module, used to respectively acquire attribute sets of crowdsourcing vehicles and crowdsourcing tasks, wherein the attribute set of the crowdsourcing vehicle includes social attributes, and the attribute set of the crowdsourcing task includes spatiotemporal attributes and location information; A prediction module, used for predicting the trajectory of the crowdsourced vehicle according to the social attribute to obtain a predicted trajectory of the crowdsourced vehicle; An allocation module, configured to formulate a task allocation strategy based on the predicted trajectory, the spatiotemporal attributes and the location information; The optimization module is used to dynamically adjust the task allocation strategy using a multi-objective optimization algorithm to obtain an optimal solution for task allocation.

Images