Database Reference
In-Depth Information
including raster representations of movement, constrained movements, and rel-
ative movement.
In a complementary direction, it is important to push much further the study
of collective behavior, that is, coordinated movement of persons, of animals,
and of any moving object driven by humans (e.g., cars, planes, ships). While
animals are obviously important to ecologists, humans' collective movements
characterize a large number of applications, including the national security and
intelligence domain that has become so critical in our current society.
Collective movement illustrates a more general research question: How to
analyze relationships among trajectories. Current advances in this domain are
mainly in terms of clustering, classification, and similarity analyses. Yet, other
relationships could be defined and useful: inverse trajectories, useful for identi-
fying return trips, and concatenation of trajectories, to make a global sense out
of a sequence of trajectories, are just two examples of how the knowledge about
trajectory understanding could be expanded. Also somehow related to collective
movement is the study of interactions among trajectories. Indeed, a trajectory
of a moving object may influence the trajectories of nearby moving objects.
In a car traffic situation, for example, the behavior of a driver can influence
nearby drivers. Open research questions include detecting such interactions and
identifying the actors, their roles, and how influences propagate among moving
objects.
Another direction for future research is investigating the new concept of
user-centered mobility data, where all the footprints left by a moving user and
collected via different means (e.g., GPS, social networks and mobile phones)
are combined together to form a global vision on the user's movements. This
raises clear interoperability and integration issues that have not been addressed
in this topic.
Trajectory Reconstruction
Data acquisition depends on the sensing technologies that are available and
appropriate for the application at hand. Usual technological evolution will cer-
tainly introduce new features that will prompt innovation in trajectory recon-
struction approaches. Future work in this domain may include the exploration
of intelligent ways to automatically extract proper values of trajectory recon-
struction parameters according to a number of characteristics of data sets, as
well as the extension of this technique to be able to identify different movement
types (pedestrian, bicycle, motorbike, car, truck, etc.) so as to enable applica-
tion of customized reconstruction techniques, resulting in better identification
of trajectories.
Existing techniques have to be reconsidered, taking a more global approach.
For example, map matching can be significantly improved by taking into account
semantic aspects (e.g., the purpose of stops). More sophisticated analyses can
