Proceedings of the Japan Joint Automatic Control Conference
THE 53RD JAPAN JOINT AUTOMATIC CONTROL CONFERENCE
Session ID : 124
Conference information

Virtual Reality, Human Interface, Kansei Engineering
Modeling and Recognition of Driving Behavior and Individual Characteristics based on Hybrid systems
*Shunsuke OtaniYasumasa KaitaniMasayuki OkamotoKenko Uchida
Author information
CONFERENCE PROCEEDINGS FREE ACCESS

Details
Abstract
Modeling the human driving behavior is expected to play an important role in desiring driving assistance systems. The modeling method based on the hybrid dynamical systems (HDS) combined continuous and discrete event system is noticeable, but how to determine the mode numbers of the HDS is still an unsolved issue. In this paper, we propose a new method of estimating the number of the modes with Kullback Leibler Divergence and also a method of modeling of driving behavior with Stochastic Switched ARX model, and discuss the effectiveness of the method by using an actual data example.
Content from these authors
© 2010 JSME
Previous article Next article
feedback
Top