Modeling and prediction of moving region trajectories

ABSTRACT

Data about moving objects is being collected in many different application domains with the help of sensor networks, GPS-enabled devices, and in particular airborne sensors and satellites. Such moving objects often represent not just point-based objects, but rather moving regions like hurricanes, oil-spills, or animal herds. One key application feature users are often interested in is the exploration and prediction of moving object trajectories. While there exist models and techniques that help to predict the movement of moving point objects, no such method for moving regions has been proposed yet.

In this paper, we present an approach to model and predict the development of moving regions. Our method not only predicts the trajectory of regions, but also the evolution of a region's spatial extent and orientation. For this, moving regions are modelled using minimum enclosing boxes, and evolution patterns of regions are determined using linear regression and a recursive motion function. We demonstrate the functionality and effectiveness of the proposed technique using real-world sensor data from different application domains.

References

1. J. Chen, X. Meng, Y. Guo, S. Grumbach, and H. Sun. Modeling and predicting future trajectories of moving objects in a constrained network. In Mobile Data Management, 2006.
2. M. Duckham, S. Nittel, and M. F. Worboys. Monitoring dynamic spatial fields using responsive geosensor networks. In ACM GIS, 2005.
3. C. Franke and M. Gertz. Detection and exploration of outlier regions in sensor data streams. In International Workshop on Spatial and Spatiotemporal Data Mining, with ICDM 2008, 2008.
4. H. Freeman and R. Shapira. Determining the minimum-area encasing rectangle for an arbitrary closed curve. Commun. ACM, 18(7), 1975.
5. P. Kalnis, N. Mamoulis, and S. Bakiras. On discovering moving clusters in spatio-temporal data. In SSTD, 2005.
6. R. Nowak and U. Mitra. Boundary estimation in sensor networks: Theory and methods. LNCS: Information Processing in Sensor Networks, 2003.
7. M. Pelanis, S. Saltenis, and C. S. Jensen. Indexing the past, present, and anticipated future positions of moving objects. ACM Trans. Database Syst., 31(1), 2006.
8. R. Praing and M. Schneider. Modeling historical and future movements of spatio-temporal objects in moving objects databases. In CIKM. ACM, 2007.
9. C. Robertson, T. Nelson, B. Boots, and M. Wulder. STAMP: spatial-temporal analysis of moving polygons. Journal of Geographical Systems, 9(3), 2007.
10. S. Subramaniam, V. Kalogeraki, and T. Palpanas. Distributed real-time detection and tracking of homogeneous regions in sensor networks. In Real-Time Systems Symposium, 2006.
11. Y. Tao, C. Faloutsos, D. Papadias, and B. Liu. Prediction and indexing of moving objects with unknown motion patterns. In SIGMOD, 2004.
12. Y. Tao, D. Papadias, and J. Sun. The TPR*-tree: an optimized spatio-temporal access method for predictive queries. In VLDB, 2003.
13. W.-G. Teng, M.-S. Chen, and P. S. Yu. A regression-based temporal pattern mining scheme for data streams. In VLDB, pages 93--104, 2003.
14. E. Tøssebro and R. H. Güting. Creating representations for continuously moving regions from observations. In SSTD, 2001.
15. V. J. Tsotras. Recent advances on querying and managing trajectories. In Tutorial at SSTD, 2007.
16. M. F. Worboys and M. Duckham. Monitoring qualitative spatiotemporal change for geosensor networks. International Journal of Geographical Information Science, 20(10), 2006.
17. W. Xue, Q. Luo, L. C. 0002, and Y. Liu. Contour map matching for event detection in sensor networks. In SIGMOD, 2006.
18. J. Zhao, C.-T. Lu, and Y. Kou. Detecting region outliers in meteorological data. In ACM GIS, 2003.

Index Terms

• Human-centered computing

  • Visualization

    • Visualization application domains

      • Geographic visualization

• Information systems

  • Information systems applications

    • Data mining

    • Spatial-temporal systems