Yingzi Lin, Ph.D.
Department of Mechanical and Industrial Engineering, Northeastern University
Wednesday, February 28, 2007, 3:30 PM to 5:00 PM in 164 WVH
Despite its great technical and social significance, the modeling of
human behaviours remains one of the greatest challenges in science
and technology development. A driver behaviour model is an essential
component in a driving simulator, an autonomous vehicle, and an
intelligent transportation system. It is known that driver behaviours
are highly nonlinear, uncertain, and random, which challenges many
scientific disciplines such as psychology for exploring first
principles that govern behaviours. In addition, driver mental states,
yet another mysterious entity, contribute to the driver behaviour.
|
 |
In this seminar, the speaker will present a study toward a new model
of the driver behaviour with consideration of driver-vehicle and
driver-environment interactions. It is proposed to apply the
artificial neural network technique for modeling the driver handing
behaviour and to apply fuzzy logic together with physiological
parameters for modeling the driver mental state. The model for the
driver mental state is incorporated into the framework of artificial
neural network to result in a driver behaviour model. The experiment
validated this model. Further, the driver behaviour model is
integrated with the vehicle dynamic model, which results in a model
for the driver-vehicle-environment closed-loop system, and this model
was experimentally validated.
The speaker will also give an overview of her previous and on-going
research projects in general human factors area, in particular: (1)
non-intrusive human sensing and measurement technology; (2) multi-
modality information fusion; and (3) interface design and measurement
toward an adaptive interface.
|