Data-Driven Trajectory Smoothing

Frédéric Chazal, Daniel Chen, Leonidas J. Guibas, Xiaoye Jiang, and Christian Sommer
GIS 2011 - 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (pp. 251-260)

Motivated by the increasing availability of large collections of noisy GPS traces, we present a new data-driven framework for smoothing trajectory data. The framework, which can be viewed of as a generalization of the classical moving average technique, naturally leads to efficient algorithms for various smoothing objectives. We analyze an algorithm based on this framework and provide connections to previous smoothing techniques. We implement a variation of the algorithm to smooth an entire collection of trajectories and show that it performs well on both synthetic data and massive collections of GPS traces.

 author    = {Fr{\'e}d{\'e}ric Chazal 
              and Daniel Chen 
              and Leonidas J. Guibas
              and Xiaoye Jiang 
              and Christian Sommer},
 title     = {Data-driven trajectory smoothing},
 booktitle = {19th ACM SIGSPATIAL International Symposium on 
              Advances in Geographic Information Systems (GIS)},
 year      = {2011},
 pages     = {251--260},
 url       = {},
 doi       = {10.1145/2093973.2094007},

Official version
Local version (1.4 MB)

HomePublications → Data-Driven Trajectory Smoothing