Forecasting fine-grained sensing coverage in opportunistic vehicular sensing. (March 2023)

• Record Type: Journal Article
• Authors: Hu, Wenyan; Winter, Stephan; Khoshelham, Kourosh
• Is Part Of: Computers, environment and urban systems. Volume 100(2023)
• Publication Date: 2023-03
• Subjects: Urban sensing -- Deep learning -- Coverage -- Spatio-temporal big data -- GIS; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2023.101939
• Languages: English
• ISSNs: 0198-9715

Abstract

Highlights: Solving the cold start problem for new opportunistic vehicular sensing platforms. Forecasting fine-grained sensing coverage of various vehicle fleets. Selecting the appropriate vehicle fleet to increase sensing quality. Evaluating the proposed systematic framework on a real taxi dataset in a case study. Abstract: Vehicles equipped with portable sensors can be used to collect data for large-scale urban sensing. However, the independent movement and non-uniform distribution of vehicles in opportunistic vehicular sensing raises questions about the quality of the sensing coverage. Appropriate vehicles are typically selected based on their historic trajectories, but this means that forecasting the sensor coverage for new vehicles is challenging. In this paper, we propose a method for tailored vehicle selection based on the forecast fine-grained sensing coverage without trajectory data. First, we propose a model, which is able to forecast fine-grained sensing coverage by coarse-grained information of candidate vehicles instead of trajectories. Then, by integrating the forecast sensing coverage with the genetic algorithm, a vehicle selection algorithm is proposed to select the appropriate vehicle fleet from the candidate vehicles to maximize the sensing quality. The method is assessed using taxi trajectory data in the evaluation experiments. The results demonstrate that the selected vehicles based on our method can achieve a higher sensing quality than two other baselines. This research provides fundamental guidelines for coverage estimation and vehicle selection in urban vehicular sensing applications. The demo is available at https://github.com/WenyanClaraHu/FiSC .