Decentralized Data Fusion and Active Sensing with Mobile Sensors for Modeling and Predicting Spatiotemporal Traffic Phenomena
Name
Jaillet_Decentralized data.pdf
Size
429.69 KB
Format
Adobe PDF
Checksum (MD5)
e6bb3f4b6edc791a539324c7c2e5426b
Author(s)
Chen, Jie
Low, Kian Hsiang
Tan, Colin Keng-Yan
Oran, Ali
Jaillet, Patrick
Dolan, John
Sukhatme, Gaurav
Date Issued
August 2012
Journal
Proceedings of the 2012 Conference on Uncertainty in Artificial Intelligence
Publisher
Association for Uncertainty in Artificial Intelligence (AUAI)
Citation
Chen, Jie et al. "Decentralized Data Fusion and Active Sensing with Mobile Sensors for Modeling and Predicting Spatiotemporal Traffic Phenomena." Proceedings of the 2012 Conference on Uncertainty in Artificial Intelligence, August 15-17, 2012.
Version
Author's final manuscript
Abstract
The problem of modeling and predicting spatiotemporal traffic phenomena over an urban road network is important to many traffic applications such as detecting and forecasting congestion hotspots. This paper presents a decentralized data fusion and active sensing (D[superscript 2]FAS) algorithm for mobile sensors to actively explore the road network to gather and assimilate the most informative data for predicting the traffic phenomenon. We analyze the time and communication complexity of D[superscript 2]FAS and demonstrate that it can scale well with a large number of observations and sensors. We provide a theoretical guarantee on its predictive performance to be equivalent to that of a sophisticated centralized sparse approximation for the Gaussian process (GP) model: The computation of such a sparse approximate GP model can thus be parallelized and distributed among the mobile sensors (in a Google-like MapReduce paradigm), thereby achieving efficient and scalable prediction. We also theoretically guarantee its active sensing performance that improves under various practical environmental conditions. Empirical evaluation on real-world urban road network data shows that our D[superscript 2]FAS algorithm is significantly more time-efficient and scalable than state-of-the-art centralized algorithms while achieving comparable predictive performance.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Persistent DSpace Link
