CN115629885A - Multi-source task allocation method and system based on local budget sharing in crowd sensing - Google Patents

Multi-source task allocation method and system based on local budget sharing in crowd sensing Download PDF

Info

Publication number
CN115629885A
CN115629885A CN202211644417.2A CN202211644417A CN115629885A CN 115629885 A CN115629885 A CN 115629885A CN 202211644417 A CN202211644417 A CN 202211644417A CN 115629885 A CN115629885 A CN 115629885A
Authority
CN
China
Prior art keywords
task
perception
tasks
sensing
source
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202211644417.2A
Other languages
Chinese (zh)
Other versions
CN115629885B (en
Inventor
魏凯敏
蔺晓川
李哲涛
漆国姿
赵诗婷
康政
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinan University
Original Assignee
Jinan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jinan University filed Critical Jinan University
Priority to CN202211644417.2A priority Critical patent/CN115629885B/en
Publication of CN115629885A publication Critical patent/CN115629885A/en
Application granted granted Critical
Publication of CN115629885B publication Critical patent/CN115629885B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5066—Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a multi-source task allocation method and system based on local budget sharing in crowd sensing, wherein the method comprises the following steps: determining a role type; the role types include: the system comprises a task requester, a mobile crowd sensing platform and mobile crowd sensing participants; collecting perception tasks through a mobile crowd perception platform; establishing a multi-source perception task distribution model according to the perception tasks; and acquiring an optimal allocation scheme based on the multi-source perception task allocation model. The present application adopts the idea of aggregation by aggregating closely located and similar perceptual tasks and making them share a budget. And a path planning method is designed to help the participants complete the task at a lower cost before the task deadline. Finally, each perception task is iteratively allocated with proper participants to achieve the task allocation goal: maximizing task completion quality and minimizing total travel distance. The problem of multi-task allocation of dual heterogeneity of sensing tasks and participants in mobile crowd sensing is solved.

Description

Multi-source task allocation method and system based on local budget sharing in crowd sensing
Technical Field
The application relates to the technical field of computer networks, in particular to a multi-source task allocation method and system based on local budget sharing in crowd sensing.
Background
With the rise of smart cities and digital cities, more and more environmental data need to be monitored so as to implement corresponding policies and be corrected. And some companies or research institutes also need to perform related perception tasks for research. However, it is a huge expense to establish large-scale, special data monitoring stations, and the data monitoring stations are fixed in position and difficult to adapt to complex and flexible data perception requirements. The occurrence of the mobile crowd sensing mode solves this problem. On the one hand, more and more mobile terminals realize strong data perception capability, and the abundant built-in sensors make it possible to complete professional data collection tasks. On the other hand, the mobile crowd sensing platform fully utilizes the mobility of the crowd. Because the mobility of people enables perception to be limited in one area, the perception device can adapt to more complex and flexible perception requirements.
Task allocation is an important design point of a mobile crowd-sourcing aware platform. A good task allocation strategy can help the task to be completed more efficiently. In the mobile crowd sensing scene, along with the construction of a general mobile crowd sensing platform, more and more sensing tasks with different sensing requirements are published on the platform. And as the platform scales, more and more participants register on the mobile aware platform. These participants differ in their perception abilities due to the different devices they purchase. Current research is primarily concerned with task or worker unilateral task assignment studies. Complex situations of bilateral isomerism of perception tasks and participants in mobile crowd sensing are ignored. And in privacy concerns, workers are often reluctant to disclose all of their sensors unless they get paid to meet their expectations. In summary, a task allocation method capable of effectively coping with the situation is needed to ensure the smooth operation of the mobile crowd sensing platform.
Disclosure of Invention
According to the method, the multi-source task allocation system is modeled, and the thought and the solving method based on local budget sharing are designed for the model, so that the task completion quality is improved under the condition that the participants and the perception task are dual and heterogeneous, and the movement distance of the participants is further reduced.
In order to achieve the above object, the present application provides a multi-source task allocation method based on local budget sharing in crowd sensing, comprising the steps of:
determining a role type; the role types include: the system comprises a task requester, a mobile crowd sensing platform and mobile crowd sensing participants;
collecting perception tasks through the mobile crowd sensing platform;
establishing a multi-source perception task distribution model according to the perception tasks;
and acquiring an optimal allocation scheme of the perception task based on the multi-source perception task allocation model.
Preferably, the method for establishing the multi-source perception task allocation model comprises the following steps: after the perception tasks are issued on the mobile crowd sensing platform, the mobile crowd sensing platform simultaneously considers the maximization of task completion quality and the minimization of total moving distance to establish the multi-source perception task distribution model under the requirements of the perception tasks and the attribute constraints of participants.
Preferably, the objectives of the multi-source perceptual task allocation model include: maximizing task completion quality and minimizing total travel distance.
Preferably, the constraint conditions of the multi-source perception task allocation model include:
the task must recruit participants under limited budget constraints;
each participant must perform a perception task under the workload of its sensors;
the sensors owned by the participants performing the perception task must cover the perception task requirements;
participants performing the perception task must arrive at the location of the task before the task deadline.
Preferably, the method for obtaining the optimal allocation scheme includes:
aggregating the perception tasks into a plurality of task sets to obtain set tasks;
planning a task path for a participant executing the set task, and calculating the reward of the participant for completing the set task according to the path;
iteratively recruiting participants for the set of missions to obtain the optimal allocation plan according to the consideration.
Preferably, the method for obtaining the set task includes: the mobile crowd sensing platform firstly aggregates the sensing tasks into a plurality of task sets according to the positions of the sensing tasks; and when the size of the task set exceeds the set maximum task number, dividing the task set for multiple times until all the task sets meet the set size limit to obtain the set tasks.
Preferably, the method for dividing comprises: and selecting a random task from the task set, and selecting the task with the closest similarity to the random task and the random task to form a task set.
Preferably, the method for planning the task path comprises:
s301, marking the position of a participant as a current position, marking the time of the participant at the current position as current time, and setting an execution sequence as null;
s302, calculating the urgency degree of the current position to each subtask in the set task, sorting according to the urgency degree, and marking the most urgent task as a candidate task;
s303, calculating the distance between the current position and other positions, selecting the perception task with the closest distance, marking the perception task as a second candidate task, and executing the second candidate task better than the candidate task;
s304, verifying whether the candidate task and the second candidate task can be completed before the task deadline to obtain a verification condition;
s305, updating the current position, the candidate task and the second candidate task information according to the verification condition;
s306, repeating S301-S305 until each task in the set of tasks enters an execution sequence.
Preferably, the method for obtaining the optimal allocation scheme includes:
calculating the reward of each participant for executing the set task to obtain a temporary distribution result;
calculating the implementation utility of the set task according to the temporary distribution result;
obtaining the optimal allocation scheme based on the realized utility.
The application also provides a multi-source task allocation system based on local budget sharing in crowd sensing, which comprises the following steps: the role dividing module, the collecting module, the constructing module and the optimizing module;
the role division module is used for determining role types; the role types include: the system comprises a task requester, a mobile crowd sensing platform and mobile crowd sensing participants;
the collection module is used for collecting perception tasks through the mobile crowd sensing platform;
the construction module is used for establishing a multi-source perception task distribution model according to the perception tasks;
the optimization module is used for obtaining the optimal distribution scheme of the perception tasks based on the multi-source perception task distribution model.
Compared with the prior art, the beneficial effects of this application are as follows:
the problem of multi-task allocation of double heterogeneous sensing tasks and participants in mobile crowd sensing is solved. The present application adopts the idea of aggregation by aggregating closely located and similar perceptual tasks and making them share a budget. And a path planning method is designed to help the participants complete the task at a lower cost before the task deadline. Finally, each perception task is iteratively allocated with proper participants to achieve the task allocation goal: maximizing task completion quality and minimizing total travel distance.
Drawings
In order to more clearly illustrate the embodiments of the present application or technical solutions in the prior art, the drawings required to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of interaction among parties of a mobile crowd sensing platform according to the present application;
FIG. 2 is a schematic diagram of a mobile crowd sensing platform multi-source task allocation mechanism according to the present application;
fig. 3 is a schematic diagram of the system structure of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, the present application is described in further detail with reference to the accompanying drawings and the detailed description.
Example one
In this embodiment, first, the role type is determined: task requesters, mobile crowd sensing platforms and mobile crowd sensing participants. Task requesters include different companies, government agencies, and individuals, and each task requester has varying professional abilities. They have different budgets and perceptual requirements for perceptual tasks. The attribute requirements of the perceptual task include: (1) Participants performing perceptual tasks need to arrive at a task location by an expiration date; (2) The perception task requires a plurality of participants to execute and provide perception data; (3) The budget of a aware task varies from task requester to task requester. Participants are benefit-driven to register on the mobile crowd sensing platform, carry different mobile devices and have different privacy concerns for different sensing data, and the privacy concerns can change due to rewards. The attribute requirements of the participants are: (1) each participant has a different set of sensors; (2) Each participant has a different attitude (i.e., open or not open) to the sensor; (3) When a participant receives a certain bonus, the participant changes the attitude for a particular sensor; (4) Each participant has a set of latitude and longitude coordinates representing the current location of the participant. The mobile crowd sensing platform is responsible for receiving tasks and distributing the sensing tasks to participants according to a task distribution strategy. The interaction process of the mobile crowd sensing platform with task requesters and participants is shown in figure 1.
As shown in fig. 2, a process of a multi-source task allocation mechanism in a mobile crowd-sourcing aware platform is presented.
First, the sensing tasks are collected by the mobile crowd sensing platform.
The mobile crowd-sourcing aware platform collects awareness tasks from different organizations, companies, or individuals. Meanwhile, participants with different perception abilities are registered to the mobile crowd sensing platform. Task requesters from different companies and organizations specify the requirements of the sensing tasks according to actual requirements, then submit the sensing tasks to the mobile crowd sensing platform, and pay the budget of the tasks to the platform. Participants carrying different mobile devices are registered in the mobile sensing platform under the drive of interests, and have different privacy concerns for different sensing data.
And then, establishing a multi-source perception task distribution model according to the perception tasks.
After the perception tasks are issued on the mobile perception platform, the mobile crowd sensing platform simultaneously considers the maximized task completion quality and the minimized total moving distance to establish a multi-source perception task distribution model under the requirements of the perception tasks and the attribute constraints of participants.
In the multi-source task allocation model, the first objective is to maximize the completion quality of the task. The completion quality of the task is determined by the number of participants who perform the task, and the formula comprises the following components:
Figure 436535DEST_PATH_IMAGE001
wherein Q is i Which is indicative of the quality of the completion of the task,
Figure 556938DEST_PATH_IMAGE002
indicating the number of participants performing task i,
Figure 680883DEST_PATH_IMAGE003
represents the minimum number of participants required to perceive task i, and
Figure 748196DEST_PATH_IMAGE004
indicating the number of participants that perceived task i needs at most. Based on equation (1), the first objective of task assignment for the mobile crowd-sourcing aware platform can be expressed as:
Figure 500251DEST_PATH_IMAGE005
wherein T represents a task set in the mobile crowd sensing platform.
And the second goal of the multi-source task allocation model is: minimizing the total travel distance.
The total movement distance is defined as the total movement distance of the participants performing the task.
Figure 853872DEST_PATH_IMAGE006
Wherein, the first and the second end of the pipe are connected with each other,Prepresenting a set of participants performing the task,
Figure 58589DEST_PATH_IMAGE007
indicating whether to task
Figure 460751DEST_PATH_IMAGE008
The assignment to participant j performs, in particular,
Figure 67313DEST_PATH_IMAGE009
representing a task
Figure 201622DEST_PATH_IMAGE008
Assigned to participant j. And then
Figure 159214DEST_PATH_IMAGE010
Representing the distance traveled by participant j to perform perception task i.
It should be noted that the constraints of multi-source task allocation include:
(1) The task must recruit participants under limited budget constraints; namely that
Figure 161805DEST_PATH_IMAGE011
Wherein, the first and the second end of the pipe are connected with each other,
Figure 888452DEST_PATH_IMAGE012
representing participant j to perform a perceptual task
Figure 724821DEST_PATH_IMAGE008
The platform needs to pay the reward to participant j,
Figure 232026DEST_PATH_IMAGE013
representing perceptual tasks
Figure 913674DEST_PATH_IMAGE008
The budget of (2).
(2) Each participant must perform a perceptual task under its workload; namely that
Figure 760408DEST_PATH_IMAGE014
Wherein, the first and the second end of the pipe are connected with each other,
Figure 361153DEST_PATH_IMAGE015
representing participant j to perform a perceptual task
Figure 496599DEST_PATH_IMAGE008
At the same time use the sensor
Figure 778676DEST_PATH_IMAGE016
。
Figure 542233DEST_PATH_IMAGE017
Sensor representing participant j
Figure 720405DEST_PATH_IMAGE016
The workload of (2).
(3) The sensors owned by the participants performing the perception tasks must cover the perception task requirements. Namely:
Figure 874305DEST_PATH_IMAGE018
wherein
Figure 287969DEST_PATH_IMAGE019
A set of sensors representing the need to perceive task i, and
Figure 617336DEST_PATH_IMAGE020
representing participantspHaving a set of sensors.
(4) Participants performing the perception task must arrive at the location of the task before the task deadline. Namely:
Figure 763147DEST_PATH_IMAGE021
wherein the content of the first and second substances,
Figure 466660DEST_PATH_IMAGE022
representing the participant j current time and v representing the participant's moving speed.
Figure 293802DEST_PATH_IMAGE023
Representing perceptual tasks
Figure 969634DEST_PATH_IMAGE008
The cutoff time of (d).
And finally, acquiring an optimal allocation scheme based on the multi-source perception task allocation model.
The step of obtaining the optimal allocation scheme comprises the following steps: gathering the perception tasks together to obtain a set task; planning a task path for a participant executing the set task; and acquiring an optimal allocation scheme according to the task path.
Firstly, the perception tasks are aggregated into a plurality of task sets to obtain a set task.
In order to obtain the optimal task allocation scheme, the mobile crowd sensing platform firstly gathers the close tasks according to the positions of the sensing tasks (namely, sensors required by the sensing tasks are completed), and the gathered tasks are expressed as collective tasks. After the task requester submits the sensing tasks to the mobile sensing platform, the mobile sensing platform classifies the sensing tasks into a plurality of task sets based on the geographic positions of the sensing tasks, and the used clustering method includes but is not limited to a DBSCAN clustering algorithm.
And further decomposing the task set obtained in the step according to the sensing requirement similarity so as to ensure that the size of the task set meets a specified range (such as maxPTs tasks). For each task set, randomly selecting a sensing task from the task sets, and then sequentially selecting maxPTs-1 sensing tasks most similar to the sensing task from the sensing task sets to form a new task set, which is called as a set task. This process is repeated until all tasks have been repartitioned.
Thereafter, a task path is planned for the participants performing the collective task.
S301, when a set task is distributed to participants, the platform needs to plan a task completion path for the participants, so that the movement distance of the participants is minimized while each task in the set task can be completed before the task is ended.
S302, marking the position of the participant as the current position, and marking the time at the moment as the current time of the participant. And sets the execution sequence to empty.
S303, calculatingUrgency of the current location for each subtask in the set task. After the calculation is completed, all subtasks in the set of tasks are sorted according to the urgency degree, and the more urgent tasks are ranked more ahead. And recording the most urgent task as a candidate task. Participant j pair task
Figure 614242DEST_PATH_IMAGE008
The urgency of (d) is defined as follows:
Figure 680418DEST_PATH_IMAGE024
s304, calculating the distance between the current position and other positions. And selecting the perception task with the closest distance, and marking the perception task as a second candidate task. And attempt to execute the second candidate task better than the candidate task.
S305, verifying whether the candidate task and the second candidate task can be completed before the task deadline. If all can be finished, the following operations are executed: (1) adding the second candidate task into the execution sequence; (2) updating the current position to a second candidate task, and removing the second candidate task from the remaining tasks; (3) jumping to step S304 and continuing to execute until all the sensing tasks are added to the execution sequence. If after the adjustment, there is a task that can not be completed, the following operations are executed: I. adding the candidate tasks into an execution sequence; updating the current position to a candidate position, and removing the candidate task from the rest tasks; jump to step S303 to continue execution until all the aware tasks are added to the execution sequence.
The mobile crowd-sourcing aware platform plans a task path for a participant performing a collective task to complete the collective task, and calculates a reward to be paid by the platform for hiring the participant to complete the collective task.
Finally, participants are recruited iteratively for each mission ensemble to obtain an optimal allocation scheme based on their return to completion.
The platform distributes the collective tasks as well as the individual tasks to the appropriate participants for execution. And the participants executing the perception tasks return the perception data to the mobile crowd sensing platform after finishing the perception when executing the independent tasks. When executing the set task, the participants execute the path perception data according to the task specified by the platform, and after the set task is completely executed, the perception data is returned to the mobile crowd sensing platform.
a. As the optimization directions of the original problem models are opposite, the quality of the distribution result is difficult to judge. Maximizing task completion quality translates into minimizing the loss of quality of completed tasks. The loss in finished task quality is represented by the difference between the maximum task finished quality and the actual task finished quality, i.e.:
Figure 107988DEST_PATH_IMAGE025
wherein, the first and the second end of the pipe are connected with each other,
Figure 435064DEST_PATH_IMAGE026
indicating a loss of quality in the performance of the task,
Figure 125940DEST_PATH_IMAGE027
indicating the quality of completion of the task.
The goal of the task allocation model at this time is:
Figure 741729DEST_PATH_IMAGE028
b. for each perceptual task (collective task), the reward required for each participant to perform this task is calculated. All rewards are ranked in order of small to large and participants are greedy selected under the budget constraints of the task. It should be noted that the budget of the set task is the sum of the budgets of all tasks included in the set task.
c. And c, calculating the effectiveness which can be realized after each perception task (set task) receives the corresponding task allocation scheme according to the temporary allocation result obtained in the step b. The utility defines a linear weighted sum of the impairment rate of the perceived quality and the distance traveled. Namely:
Figure 300886DEST_PATH_IMAGE029
wherein
Figure 623414DEST_PATH_IMAGE030
Is a trade-off factor for the number of,
Figure 344246DEST_PATH_IMAGE031
indicating participant completion of task
Figure 447331DEST_PATH_IMAGE008
The total distance moved.
d. The mobile perception platform adds the distribution scheme of the perception task (the set task) with the minimum utility into the platform task distribution scheme. And updating the states of the perception tasks and the participants, and jumping to the step b to continue to execute until all the tasks are distributed or the effectiveness of all the perception tasks (set tasks) is 0.
Taking the integrated mobile crowd-sourcing aware platform as an example, task requesters from different organizations submit their task requirements and budgets to the platform. If a certain city environmental protection agency submits a noise monitoring task, the task needs to use an audio sensor and a GPS and needs 3 participants to complete, and the budget is 30 yuan; the atmospheric environment research institute in some city submits the air quality monitoring task, and the task needs to use image sensor and GPS and needs 6 participants to accomplish, and its budget is 45 yuan. A number of participants with different device conditions and privacy concerns are registered on the mobile crowd-sourcing platform, such as participant a has an image sensor and GPS, but is reluctant to open the image sensor unless he gets a 5-dollar premium and his image sensor can perform two tasks while GPS can perform three tasks; participant B, having an image sensor, an audio sensor and GPS, would like to have all sensors open, each of which can perform four tasks. After receiving the request of a task requester, the mobile crowd sensing platform re-divides the sensing task based on the proposed task packing algorithm, redefines the attribute of the set task and enables the tasks in the set task to share budget. The platform then computes a task execution sequence for each participant execution set task according to the proposed path planning algorithm. Finally, participants are iteratively selected for each task based on a greedy idea. In each iteration, the least remunerated participants are preferably recruited for each task and the utility that can be achieved by the task is calculated. The least effective task is assigned to the participant first. The idea of sharing budget by the set task enables the set task to have more chances to select more participants, and further the completion quality of the task is improved. After a participant has requested or requested a task to be performed, the participant moves to the location-aware data where the task is located. And if the task to be executed is an independent task, the participant returns the perception data to the mobile crowd sensing platform after the perception is finished. And if the task to be executed is a set task, the participants execute the path perception data according to the task specified by the platform, and after the set task is completely executed, the perception data is returned to the mobile crowd sensing platform.
Example two
Fig. 3 is a schematic diagram of the system structure of the present embodiment. Firstly, determining the role type by using a role division module: the system comprises a task requester, a mobile crowd sensing platform and a mobile crowd sensing participant. Task requesters include different companies, government agencies, and individuals, and each task requester has varying professional abilities. They have different budgets and perceptual requirements for perceptual tasks. The attribute requirements of the perceptual task include: (1) Participants performing the perception task need to arrive at the task location by the expiration date; (2) The perception task requires multiple participants to execute and provide perception data; (3) The budget of a aware task varies from task requester to task requester. Participants are registered on a mobile crowd sensing platform under the drive of benefits, carry different mobile devices and have different privacy concerns for different sensing data, and the privacy concerns can be changed due to rewards. The attribute requirements of the participants are: (1) each participant has a different set of sensors; (2) Each participant has a different attitude (i.e., open or not open) to the sensor; (3) When a participant receives a certain bonus, the participant changes the attitude for a particular sensor; (4) Each participant has a set of latitude and longitude coordinates representing the current location of the participant. The mobile crowd sensing platform is responsible for receiving tasks and distributing the sensing tasks to participants according to a task distribution strategy. The interaction process of the mobile crowd sensing platform with task requesters and participants is shown in figure 1.
As shown in fig. 2, a process of a multi-source task allocation mechanism in a mobile crowd-sourcing aware platform is illustrated.
First, a collection module collects sensing tasks through a mobile crowd sensing platform.
The mobile crowd-sourcing aware platform collects awareness tasks from different organizations, companies, or individuals. Meanwhile, participants with different sensing capabilities register to the mobile crowd sensing platform. Task requesters from different companies and organizations specify the requirements of the perception tasks according to actual requirements, then submit the perception tasks to the mobile crowd sensing platform, and pay the budget of the tasks to the platform. Participants carrying different mobile devices are registered in the mobile sensing platform under the drive of interests, and have different privacy concerns for different sensing data.
And then, the construction module establishes a multi-source perception task distribution model according to the perception tasks.
After the perception tasks are issued on the mobile perception platform, the mobile crowd sensing platform simultaneously considers the maximized task completion quality and the minimized total moving distance to establish a multi-source perception task distribution model under the requirements of the perception tasks and the attribute constraints of participants.
In the multi-source task allocation model, the first objective is to maximize the completion quality of the task. The completion quality of the task is determined by the number of participants who perform the task, and the formula comprises the following components:
Figure 154387DEST_PATH_IMAGE032
wherein the content of the first and second substances,
Figure 659318DEST_PATH_IMAGE033
which indicates the quality of the completion of the task,
Figure 285471DEST_PATH_IMAGE034
indicating the number of participants performing task i,
Figure 79115DEST_PATH_IMAGE003
indicates the number of participants who need the least amount of the perception task i, and
Figure 652179DEST_PATH_IMAGE004
indicating the number of participants most needed to perceive task i. Based on equation (12), the first goal of task allocation for the mobile crowd-sourcing aware platform can be expressed as:
Figure 73933DEST_PATH_IMAGE035
wherein T represents a task set in the mobile crowd sensing platform.
And the second goal of the multi-source task allocation model is: minimizing the total travel distance.
The total movement distance is defined as the total movement distance of the participants performing the task.
Figure 11933DEST_PATH_IMAGE036
Wherein the content of the first and second substances,Prepresenting a set of participants performing the task,
Figure 824031DEST_PATH_IMAGE007
indicating whether to task
Figure 528682DEST_PATH_IMAGE008
The assignment to participant j performs a process of, in particular,
Figure 680309DEST_PATH_IMAGE009
representing a task
Figure 851527DEST_PATH_IMAGE008
Assigned to participant j. While
Figure 478817DEST_PATH_IMAGE010
Representing the distance traveled by participant j to perform perceptual task i.
It should be noted that the constraints of multi-source task allocation include:
(1) The task must recruit participants under limited budget constraints; namely, it is
Figure 331367DEST_PATH_IMAGE037
Wherein, the first and the second end of the pipe are connected with each other,
Figure 727713DEST_PATH_IMAGE012
representing participant j to perform a perceptual task
Figure 69833DEST_PATH_IMAGE008
The platform needs to pay participant j a reward,
Figure 794206DEST_PATH_IMAGE013
representing perceptual tasks
Figure 575080DEST_PATH_IMAGE008
The budget of (2).
(2) Each participant must perform the perception task under its workload; namely, it is
Figure 763616DEST_PATH_IMAGE038
Wherein the content of the first and second substances,
Figure 214320DEST_PATH_IMAGE015
representing participant j to perform a perceptual task
Figure 816203DEST_PATH_IMAGE008
At the same time use the sensor
Figure 72872DEST_PATH_IMAGE016
。
Figure 319177DEST_PATH_IMAGE017
Sensor representing participant j
Figure 330995DEST_PATH_IMAGE016
The workload of (2).
(3) The sensors owned by the participants performing the perception task must cover the perception task requirements. Namely:
Figure 764381DEST_PATH_IMAGE039
wherein
Figure 152637DEST_PATH_IMAGE019
A set of sensors representing the need to perceive task i, and
Figure 50186DEST_PATH_IMAGE020
representing participantspHaving a set of sensors.
(4) Participants performing the perception task must arrive at the location of the task before the task deadline. Namely:
Figure 842693DEST_PATH_IMAGE040
wherein the content of the first and second substances,
Figure 419168DEST_PATH_IMAGE022
representing the participant j current time and v representing the participant's moving speed.
Figure 17639DEST_PATH_IMAGE023
Representing perceptual tasks
Figure 707378DEST_PATH_IMAGE008
The cutoff time of (c).
And finally, the optimization module obtains an optimal distribution scheme based on the multi-source perception task distribution model.
The work flow of the optimization module comprises the following steps: gathering the perception tasks together to obtain a set task; planning a task path for a participant executing the set task; and acquiring an optimal allocation scheme according to the task path.
Firstly, the perception tasks are aggregated into a plurality of task sets to obtain a set task.
In order to obtain the optimal task allocation scheme, the mobile crowd sensing platform firstly gathers the close tasks according to the positions of the sensing tasks (namely, sensors required by the sensing tasks are completed), and the gathered tasks are expressed as collective tasks. After the task requester submits the sensing tasks to the mobile sensing platform, the mobile sensing platform classifies the sensing tasks into a plurality of task sets based on the geographic positions of the sensing tasks, and the used clustering method includes but is not limited to a DBSCAN clustering algorithm.
And further decomposing the task set obtained by the process according to the sensing requirement similarity so as to ensure that the size of the task set meets a specified range (such as maxPTs tasks). For each task set, randomly selecting a perception task from the task sets, and then sequentially selecting maxPTs-1 perception tasks which are most similar to the task from the perception tasks to form a new task set, which is called as a set task. This process is repeated until all tasks have been repartitioned.
Thereafter, a task path is planned for the participants performing the collective task.
S301, when a set task is distributed to participants, the platform needs to plan a task completion path for the participants so as to ensure that each task in the set task can be completed before the task is ended and simultaneously minimize the movement distance of the participants.
S302, the position of the participant is marked as the current position, and the time at the moment is marked as the current time of the participant. And sets the execution sequence to empty.
And S303, calculating the urgency of the current position for each subtask in the set task. After the calculation is completed, all subtasks in the set of tasks are sorted according to the urgency degree, and the more urgent tasks are ranked more ahead. And recording the most urgent task as a candidate task. Participant j pair task
Figure 326578DEST_PATH_IMAGE008
The urgency of (d) is defined as follows:
Figure 796874DEST_PATH_IMAGE041
s304, calculating the distance between the current position and other positions. And selecting the perception task with the closest distance, and marking the perception task as a second candidate task. And attempt to execute the second candidate task better than the candidate task.
S305, verifying whether the candidate task and the second candidate task can be completed before the task deadline. If all can be finished, the following operations are executed: (1) adding the second candidate task into the execution sequence; (2) updating the current position to a second candidate task, and removing the second candidate task from the remaining tasks; (3) jumping to step S304 continues execution until all the sensing tasks are added to the execution sequence. If after the adjustment, there is a task that can not be completed, the following operations are executed: I. adding the candidate tasks into an execution sequence; updating the current position to a candidate position, and removing the candidate task from the rest tasks; jump to step S303 to continue execution until all the aware tasks are added to the execution sequence.
The mobile crowd-sourcing aware platform plans a task path for a participant who executes the collective task to complete the collective task, and calculates the reward that the platform needs to pay to hire the participant to complete the collective task.
Finally, the participants are iteratively recruited for each tasking set to obtain an optimal allocation plan based on rewards for participants to complete the tasking set.
The platform distributes the collective tasks as well as the individual tasks to the appropriate participants for execution. And the participants executing the perception tasks return perception data to the mobile crowd sensing platform after finishing the perception when executing the independent tasks. When executing the set task, the participants execute the path perception data according to the task specified by the platform, and after the set task is completely executed, the perception data is returned to the mobile crowd sensing platform.
a. As the optimization directions of the original problem models are opposite, the quality of the distribution result is difficult to judge. Maximizing task completion quality translates into minimizing the loss of quality of completed tasks. The loss in finished task quality is represented by the difference between the maximum task finished quality and the actual task finished quality, i.e.:
Figure 136719DEST_PATH_IMAGE042
wherein the content of the first and second substances,
Figure 743281DEST_PATH_IMAGE026
indicating a loss of quality in the performance of the task,
Figure 267803DEST_PATH_IMAGE027
indicating the quality of completion of the task.
The goal of the task allocation model at this time is:
Figure 428657DEST_PATH_IMAGE043
b. for each perceptual task (collective task), the reward required for each participant to perform this task is calculated. All rewards are ranked in order of small to large and participants are greedy selected under the budget constraints of the task. It should be noted that the budget of the set task is the sum of the budgets of all tasks included in the set task.
c. And c, calculating the effect which can be realized after each perception task (set task) receives the corresponding task allocation scheme according to the temporary allocation result obtained in the step b. The utility defines a linear weighted sum of the impairment rate of perceived quality and the distance traveled. Namely:
Figure 368932DEST_PATH_IMAGE044
wherein
Figure 423475DEST_PATH_IMAGE030
Is a trade-off factor for the number of,
Figure 994265DEST_PATH_IMAGE031
indicating participant completion of task
Figure 501470DEST_PATH_IMAGE008
The total distance moved.
d. The mobile perception platform adds the distribution scheme of the perception task (set task) with the minimum utility into the platform task distribution scheme. And updating the states of the perception tasks and the participants, and jumping to the step b to continue to execute until all the tasks are distributed or the effectiveness of all the perception tasks (set tasks) is 0.
The above-described embodiments are merely illustrative of the preferred embodiments of the present application, and do not limit the scope of the present application, and various modifications and improvements made to the technical solutions of the present application by those skilled in the art without departing from the design spirit of the present application should fall within the protection scope defined by the claims of the present application.

Claims (10)

1. The multi-source task allocation method based on local budget sharing in crowd sensing is characterized by comprising the following steps:
determining the role type; the role types include: the system comprises a task requester, a mobile crowd sensing platform and a mobile crowd sensing participant;
collecting perception tasks through the mobile crowd sensing platform;
establishing a multi-source perception task distribution model according to the perception tasks;
and acquiring an optimal allocation scheme of the perception task based on the multi-source perception task allocation model.
2. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 1, wherein the method for establishing the multi-source sensing task allocation model comprises: after the perception tasks are issued on the mobile crowd sensing platform, the mobile crowd sensing platform simultaneously considers the maximized task completion quality and the minimized total moving distance to establish the multi-source perception task distribution model under the requirements of the perception tasks and the attribute constraints of participants.
3. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 2, wherein the objectives of the multi-source sensing task allocation model include: maximizing task completion quality and minimizing total travel distance.
4. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 3, wherein the constraint conditions of the multi-source sensing task allocation model include:
the task must recruit participants under limited budget constraints;
each participant must perform a perception task under the workload of its sensors;
the sensors owned by the participants performing the sensing task must cover the sensing task requirements;
participants performing the perception task must arrive at the location of the task before the task deadline.
5. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 2, wherein the method for obtaining the optimal allocation scheme comprises:
aggregating the perception tasks into a plurality of task sets to obtain set tasks;
planning a task path for a participant executing the task set, and calculating the reward for the participant to complete the task set according to the path;
iteratively recruiting participants for the set of missions to obtain the optimal allocation plan according to the consideration.
6. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 5, wherein the method for obtaining the task set comprises: the mobile crowd sensing platform firstly aggregates the sensing tasks into a plurality of task sets according to the positions of the sensing tasks; and when the size of the task set exceeds the set maximum task number, dividing the task set for multiple times until all the task sets meet the set size limit to obtain the set tasks.
7. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 6, wherein the dividing method comprises: and selecting a random task from the task set, and selecting the task with the closest similarity to the random task and the random task to form a task set.
8. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 5, wherein the method for planning the task path comprises:
s301, marking the position of a participant as a current position, marking the time of the participant at the current position as current time, and setting an execution sequence as null;
s302, calculating the urgency degree of the current position to each subtask in the set task, sorting according to the urgency degree, and marking the most urgent task as a candidate task;
s303, calculating the distance between the current position and other positions, selecting the perception task with the closest distance, marking the perception task as a second candidate task, and executing the second candidate task better than the candidate task;
s304, verifying whether the candidate task and the second candidate task can be completed before the task deadline date to obtain a verification condition;
s305, updating the current position, the candidate task and the second candidate task information according to the verification condition;
s306, repeating S301-S305 until each task in the set of tasks enters an execution sequence.
9. The multi-source task allocation method based on local budget sharing in crowd sensing according to claim 5, wherein the method for obtaining the optimal allocation scheme comprises:
calculating the reward of each participant for executing the set task to obtain a temporary distribution result;
calculating the implementation utility of the set task according to the temporary distribution result;
obtaining the optimal allocation scheme based on the realized utility.
10. The multisource task allocation system based on local budget sharing in crowd sensing is characterized by comprising the following components: the system comprises a role dividing module, a collecting module, a constructing module and an optimizing module;
the role division module is used for determining role types; the role types include: the system comprises a task requester, a mobile crowd sensing platform and mobile crowd sensing participants;
the collection module is used for collecting perception tasks through the mobile crowd sensing platform;
the construction module is used for establishing a multi-source perception task distribution model according to the perception tasks;
the optimization module is used for obtaining the optimal distribution scheme of the perception tasks based on the multi-source perception task distribution model.
CN202211644417.2A 2022-12-21 2022-12-21 Multi-source task allocation method and system based on local budget sharing in crowd sensing Active CN115629885B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211644417.2A CN115629885B (en) 2022-12-21 2022-12-21 Multi-source task allocation method and system based on local budget sharing in crowd sensing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211644417.2A CN115629885B (en) 2022-12-21 2022-12-21 Multi-source task allocation method and system based on local budget sharing in crowd sensing

Publications (2)

Publication Number Publication Date
CN115629885A true CN115629885A (en) 2023-01-20
CN115629885B CN115629885B (en) 2023-04-07

Family

ID=84910868

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211644417.2A Active CN115629885B (en) 2022-12-21 2022-12-21 Multi-source task allocation method and system based on local budget sharing in crowd sensing

Country Status (1)

Country Link
CN (1) CN115629885B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116663855A (en) * 2023-07-25 2023-08-29 暨南大学 Bilateral satisfaction online task matching method and system in crowd sensing
CN116976600A (en) * 2023-07-14 2023-10-31 陕西师范大学 Crowd sensing environment monitoring task recommendation method based on concurrent task bundling

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100312729A1 (en) * 2009-06-05 2010-12-09 Chen Thomas C H System and method of real-time group intelligence creation with mass volume of web-based thinking grids
CN109347905A (en) * 2018-08-30 2019-02-15 天津工业大学 A kind of space tasks distribution mechanism in mobile intelligent perception
US20190122162A1 (en) * 2017-10-20 2019-04-25 Accenture Global Solutions Limited Intelligent crowdsourced resource assistant
CN111277646A (en) * 2020-01-19 2020-06-12 徐州工业职业技术学院 Task distribution algorithm based on incidence relation and geographic position prediction in crowd sensing
CN111639856A (en) * 2020-05-29 2020-09-08 中国人民解放军国防科技大学 City environment object calibration method, system and medium based on crowd sensing technology
CN114925941A (en) * 2022-07-21 2022-08-19 深圳市信润富联数字科技有限公司 Crowd sensing task allocation method, device, equipment and storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100312729A1 (en) * 2009-06-05 2010-12-09 Chen Thomas C H System and method of real-time group intelligence creation with mass volume of web-based thinking grids
US20190122162A1 (en) * 2017-10-20 2019-04-25 Accenture Global Solutions Limited Intelligent crowdsourced resource assistant
CN109347905A (en) * 2018-08-30 2019-02-15 天津工业大学 A kind of space tasks distribution mechanism in mobile intelligent perception
CN111277646A (en) * 2020-01-19 2020-06-12 徐州工业职业技术学院 Task distribution algorithm based on incidence relation and geographic position prediction in crowd sensing
CN111639856A (en) * 2020-05-29 2020-09-08 中国人民解放军国防科技大学 City environment object calibration method, system and medium based on crowd sensing technology
CN114925941A (en) * 2022-07-21 2022-08-19 深圳市信润富联数字科技有限公司 Crowd sensing task allocation method, device, equipment and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JIANGTAO WANG等: "Fine-Grained Multitask Allocation for Participatory Sensing With a Shared Budget" *
王鑫: "面向任务需求的群智感知任务分配模型" *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116976600A (en) * 2023-07-14 2023-10-31 陕西师范大学 Crowd sensing environment monitoring task recommendation method based on concurrent task bundling
CN116976600B (en) * 2023-07-14 2024-03-22 陕西师范大学 Crowd sensing environment monitoring task recommendation method based on concurrent task bundling
CN116663855A (en) * 2023-07-25 2023-08-29 暨南大学 Bilateral satisfaction online task matching method and system in crowd sensing
CN116663855B (en) * 2023-07-25 2024-02-06 暨南大学 Bilateral satisfaction online task matching method and system in crowd sensing

Also Published As

Publication number Publication date
CN115629885B (en) 2023-04-07

Similar Documents

Publication Publication Date Title
CN115629885B (en) Multi-source task allocation method and system based on local budget sharing in crowd sensing
Alexander et al. Assessing the impact of real-time ridesharing on urban traffic using mobile phone data
Weng et al. Homefinder revisited: Finding ideal homes with reachability-centric multi-criteria decision making
Perugia et al. Designing a home-to-work bus service in a metropolitan area
CN111402616B (en) Intelligent parking control method, system, storage medium and terminal
WO2013171780A1 (en) Transportation route distribution ratio control system and transportation route distribution control method
CN113077090A (en) Passenger flow prediction method, system and computer readable storage medium
Tang et al. Online operations of automated electric taxi fleets: An advisor-student reinforcement learning framework
CN110309952B (en) City employment spatial layout optimization auxiliary system based on commuting model
US20210272039A1 (en) Systems and methods for evaluating a microtransit service
Chen et al. The impact of rainfall on the temporal and spatial distribution of taxi passengers
CN113672846A (en) Network appointment scheduling method and device, electronic equipment and storage medium
CN109041218A (en) A kind of method and Intelligent hardware for predicting user location
Guo et al. A force-directed approach to seeking route recommendation in ride-on-demand service using multi-source urban data
CN111801701A (en) System and method for scheduling service providers
Yan et al. An incentive mechanism for private parking-sharing programs in an imperfect information setting
CN110288125A (en) It is a kind of based on the commuting method for establishing model of mobile phone signaling data and application
Monnot et al. Routing games in the wild: Efficiency, equilibration, regret, and a price of anarchy bound via long division
Guo et al. Fine-grained dynamic price prediction in ride-on-demand services: Models and evaluations
CN115759660A (en) Scheduling method, device, equipment and medium for unmanned vehicle
de Almeida Buosi et al. A crowdsensing campaign and data analytics for assisting urban mobility pattern determination
CN116108970A (en) Smart city sharing bicycle putting and operation area planning method and Internet of things system
CN115713213A (en) Crowd-sourcing task allocation method based on man-machine cooperation
Mastroianni et al. Local optimization strategies in urban vehicular mobility
Hobi The impact of real-time information sources on crowd-sourced parking availability prediction

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant