Abstract
Mobile crowd sensing is an innovative and promising paradigm in the construction and perception of smart cities. However, multi-task allocation in real-world scenarios is a huge challenge. There are many unexpected factors in the execution of mobile crowd sensing tasks, such as traffic jams or accidents, that make participants unable to reach the target area. In addition, participants may quit halfway due to equipment failure, network paralysis, dishonest behavior, etc. Previous task allocation approaches mainly ignored some of the heterogeneity of participants and tasks in the real-world scenarios. This paper proposes a real-world-oriented multi-task allocation approach based on multi-agent reinforcement learning. Firstly, under the premise of fully considering the heterogeneity of participants and tasks, the approach enables participants as agents to learn multiple solutions independently, based on modified soft Q-learning. Secondly, two cooperation mechanisms are proposed for obtaining the stable joint action, which can minimize the total sensing time while meeting the sensing quality constraint, which optimizes the sensing quality of mobile crowd sensing (MCS) tasks. Experiments verify that the approach can effectively reduce the impact of emergencies on the efficiency of large-scale MCS platform and outperform baselines based on a real-world dataset under different experiment settings.
Highlights
Mobile crowd sensing (MCS) is an innovative paradigm of sensing based on crowd sourcing
Task allocation is a key issue in mobile crowd sensing, which has an impact on the efficiency and
The key contributions of this paper include three parts: (1) we investigate the impact of different factors on multi-task allocation in real-world scenarios, and mainly take the heterogeneity of the participants and tasks into consideration to formulate an optimization problem which aims to minimize the total sensing time under meeting the sensing quality constraint; (2) we turn the quality optimization problem into a multi-pack problem whose optimal solution will be set as the sensing quality constraint; (3) we propose a novel multi-agent reinforcement learning framework
Summary
Mobile crowd sensing (MCS) is an innovative paradigm of sensing based on crowd sourcing. MCS leverages smart mobile devices (smart phone, wearable device, etc.) that ordinary people carry to form a large-scale perception system, which can complete large-scale sensing tasks that traditional static sensing devices cannot solve It has stimulated a lot of attractive applications, such as air quality monitoring, traffic information mapping, and infrastructure inspection. By leveraging the mobility of the mobile terminal users,bethe deployment cost of specialized has been widely used in many applications, including environmental monitoring [3], transportation [4], sensing infrastructure for large-scale data collection applications would be largely reduced. Workflow mobile crowdincludes sensing is as follows: the typicalofMCS system three types of objects: (1) data requester; (2) MCS platform; (3). Assigning sensing tasks to mobile terminals is issue to deal with the following steps.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
