Exploring shared structure among vehicle trajectories

Placeholder Show Content


Trajectory data, which record the evolution of objects in space and time, are everywhere. Examples are GPS trajectories recording the moving traces of a group of people, video sequences of moving cars, etc. Such trajectory collections contain rich and diverse information about their surrounding environment, as well as the collective behavior of their human agents. A common characteristic of trajectory data collected from mobile platforms is data uncertainty. The uncertainty comes from sensor noise, low sampling rate due to power, storage, or data transmission constraints. However, large amounts of trajectories traversing the same spatial-temporal space imply that the trajectories share a lot of common structures. A good understanding of this shared structure can benefit many applications, including route planning, autonomous driving, and ride sharing. In this work, we explore shared structure among trajectory collections and show their usefulness in various applications. We begin with the physical shared structure, i.e., the road network, and show that trajectory collections, such as GPS trajectories, or driving video sequences, have great value in creating and maintaining up-to-date, information-rich maps of the underlying road network. We then explore shared mobility patterns among trajectories from a unique and knowledgeable group of people, specifically taxi drivers. We describe common mobility patterns in such groups using pathlets, which represent popular sub-trajectories shared by many trajectories. We present an unsupervised pathlet learning framework that extracts a compact pathlet dictionary from a collection of GPS traces and demonstrate various applications.


Type of resource text
Form electronic; electronic resource; remote
Extent 1 online resource.
Publication date 2016
Issuance monographic
Language English


Associated with Chen, Chen
Associated with Stanford University, Department of Electrical Engineering.
Primary advisor Guibas, Leonidas J
Thesis advisor Guibas, Leonidas J
Thesis advisor Girod, Bernd
Thesis advisor Ye, Yinyu
Advisor Girod, Bernd
Advisor Ye, Yinyu


Genre Theses

Bibliographic information

Statement of responsibility Chen Chen.
Note Submitted to the Department of Electrical Engineering.
Thesis Thesis (Ph.D.)--Stanford University, 2016.
Location electronic resource

Access conditions

© 2016 by Chen Chen
This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).

Also listed in

Loading usage metrics...