Database Reference
In-Depth Information
data, recent research has been investigating ways to reformulate and enrich
movement data to make them better correspond to application requirements
and scenarios. This is done by adding to the movement data contextual data
that describe where the object moved (e.g., the roads it followed, the places
where it stopped), when (e.g., during which time period, during which event),
how (e.g., using which transportation means), what for (e.g., which activity it
performed when it stopped). Enriched movement tracks are nowadays referred
to as
semantic trajectories
. Chapters
6
and
7
in this topic discuss how to build
and use semantic trajectories.
This initial chapter introduces the reader to a global understanding of the
trajectory domain. It spans from raw data to data transformation and enrichment,
to end up with the analysis tasks needed to fulfill application requirements. The
chapter covers both the static representation of the domain (what a trajectory is
and how it can be represented) and its behavioral aspects (how to understand
and characterize mobility in terms of why things move, what they do while
moving, which are meaningful movement sequences, etc.). Given the diversity of
application requirements, several representations of trajectories are considered.
Basic concepts and terminology are defined, explained, and documented via
examples.
1.2 Trajectory: Definition and Application Scenario
Mobility is a recent domain where people use diverse terminologies and con-
cepts, without much consensus on choices and definitions. To limit misunder-
standing and confusion, this section defines a set of concepts and vocabulary
that together form a consistent framework for discussing trajectories and their
analysis as understood in this topic.
At the source of our movement data processing concerns there is a moving
object, that is, an object that can over time change its position in space (its
spatial coordinates). In this topic, we don't address deformation issues raised
when considering moving objects, such as hurricanes and oil spills, that span
over a changing area or volume. We focus instead on moving objects represented
as points. Keeping movement data about a moving object consists in keeping
the history of its successive positions, that is, creating a record that holds, for
this object, all past, present, and sometimes future positions and the associated
instants. We will not discuss future positioning at this point, and call this record
the movement track of the object. The sequence can be unbounded. The time
intervals between successive positions may have the same duration or different
durations.
Definition 1.1. The
movement track
of a moving object is the temporally ordered
sequence of spatio-temporal position records captured by a positioning device
