Research On Multi-objective Task Assignment Methods Of Mobile Crowdsourcing

In recent years,with the development of the Internet of Things(Io T),mobile crowdsourcing(spatio-temporal crowdsourcing)has become a new paradigm of group intelligence perception.It combines the ideas of crowdsourcing and the sensing capabilities of mobile devices to complete location-based sensing tasks.In the mobile crowdsourcing system,in order to motivate crowd workers to actively participate in crowd tasks and provide high-quality and reliable sensing data,it is necessary to design an effective incentive mechanism.The incentive mechanism of mobile crowdsourcing mainly involves issues such as task assignment,privacy protection,and quality control.This paper mainly studies the multi-objective task assignment methods in the mobile crowdsourcing system.Multi-objective optimization problems are very common and in a very important position in real life such as engineering applications.Similarly,in the mobile crowdsourcing system,there are multi-objective optimization problems both on the platform side and on the crowd worker side.However,in previous studies,only the multi-objective of the platform was optimized or even only a single object,which affected the maximization of social welfare and the development of mobile crowdsourcing systems.At the same time,with the continuous development of artificial intelligence technology and the continuous increase of data,the mobile crowdsourcing system based on the traditional cloud center can no longer meet the increasing demand for data,so it needs to be turned to edge computing to accelerate the speed of data analysis,so that the mobile crowdsourcing system can make faster and better decisions.The research contents of this thesis are shown as follows:1.In order to solve the problem of multi-objective task assignment that maximizes the utility of the platform and the utility of crowdsourcing workers at the same time,firstly,a incentive mechanism that considers the attributes of task difficulty and worker reputation,ability,and privacy sensitivity is proposed to encourage crowds workers to execute remote and non-popular areas crowd tasks.In addition,a Weighted and Multi-Objective Particle Swarm Combination(WAMOPSC)algorithm is designed to solve the multi-objective task assignment problem and provide more Pareto solutions for the platform.Finally,the effectiveness of the algorithm is verified on the real data set.2.In order to improve the service quality of the mobile crowdsourcing system,the vehicle-based three-objective optimization task assignment problem is studied.First,use the mobility of the vehicle and the perception ability of the mobile device to complete the query task and the automatic perception task.Then,three optimization goals are formalized by formulas: maximizing task coverage,maximizing task reliability,and minimizing crowds worker cost.Finally,an evolutionary many-objective optimization algorithm using reference point-based nondominated sorting approach(NSGA-III)is used to optimize the three objectives to obtain more Pareto solutions,and the effectiveness of the method is verified by experiments on real data sets.