M-Atlas: creating the atlas of urban mobility by mining massive GPS trajectory data.

 

FOSCA GIANNOTTI

ISTI-CNR, PISA, ITALY AND CCNR, NORTHEASTERN UNIVE

 

The technologies of mobile communications and ubiquitous computing pervade our society, and wireless networks sense the movement of people and vehicles, generating large volumes of mobility data, such as mobile phone call records and GPS tracks. In this work, we illustrate the analytical power of massive collections of trajectory data, sensed by vehicular GPS devices at a fine spatio-temporal resolution, in unveiling the complexity of urban mobility. We present the results of a large scale experiment, based on a real life GPS dataset, obtained from 17,000 private cars with on-board GPS receivers, tracked during one week of ordinary mobile activity in the metropolitan area of the city of Milan, Italy. The observed population consists of anonymous and heterogeneous car drivers participating in a specific car insurance program. On the basis of this experiment, we show how a comprehensive atlas of urban mobility, the M-Atlas, can be created, which reveals the relevant mobility behaviors: commuting trips, frequently followed itineraries, convergent patterns, slow-down patterns, etc. The M-Atlas concept goes beyond the O-D matrix: not only the flux among locations is analyzed, but also the movement patterns obtained by learning from trajectory micro-data. The M-Atlas can be browsed by a mobility analyst (by the hours of the day, the days of the week, the geographic area, …), in order to explore the typical mobility of a city in varying circumstances, also to observe emerging deviations from normal. In order to turn raw GPS tracks into such forms of mobility knowledge, a thorough infrastructure has been created, designed around a core of models and algorithms for trajectory data mining and analysis, including trajectory pattern mining, trajectory clustering according to various similarity notions, trajectory classification and prediction.