Traffic estimation and prediction (or dynamic traffic assignment) models are
expected to contribute to the reduction of travel time delays. In this book, an
on-line calibration approach that jointly estimates all model parameters is
presented. The methodology imposes no restrictions on the models, the
parameters or the data that can be handled, and emerging or future data can
be easily incorporated. The modeling approach is applicable to any simulation
model and is not restricted to the application domain covered in this book.
Several modified, non-linear Kalman Filter methodologies are presented, e.g.
Extended Kalman Filter (EKF), Iterated EKF, Limiting EKF, and Unscented
Kalman Filter. Extensive case studies on freeway networks in Europe and the
US are used to demonstrate the approach, to verify the importance of on-line
calibration, and to test the presented algorithms. The main target audience of
this book comprises Intelligent Transportation Systems researchers and
graduate students, as well as practitioners, including Metropolitan Planning
Organization engineers and Traffic Management Center operators, and any
reader with an interest in dynamic state and parameter estimation.