01023103

Component

This paper describes how car-following models are one of the main logical processes in all microscopic traffic simulation models such as FRESIM, NETSIM, INTRAS, CARSIM and INTELSIM, and also in modern traffic flow theory. They have also become essential in developing auto cruise control strategies recently. Modeling car-following behavior has been the focus of a big number of research projects, most of which was dealing with the statistical approach. However, previous experimental studies and models on car-following behavior have some important limitations, which make them inconsistent with real driving experiences. This paper presents an Artificial Neural Networks (ANN) alternative to the classical techniques. Two networks were designed to obtain two models for car-following behavior in Beijing expressway using experimental data measured by RTK Global Positioning System (GPS) receivers. The first is to model car-following behavior based on widely used back propagation (BP), and the second car-following model is based on the Radial Basis Function (RBF) neural network. Simulation of the two networks with data gathered from three different expressways in Beijing revealed that the proposed RBF network has a higher precision and requires shorter training compared with the BP network in the prediction of the car-following model and accurate models can be achieved. The network was trained with all the data mixed up. Despite we dealt with three car-following states (acceleration, deceleration and stable state), only one network was enough for each case to provide reasonable results.

01020180

06-1115

English

26p

2006

Transportation Research Board 85th Annual Meeting

CD-ROM

Figures (9) ; References (15) ; Tables (4)

Backpropagation; Behavior; Cruise control; Global Positioning System; Microscopic traffic flow; Neural networks

FRESIM (Traffic simulation model); Netsim (Computer program)

Car following models; Radial basis function; Statistical techniques

Beijing (China)

Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2006 Paper #06-1115

TRIS, TRB

Mar 3 2006 10:35AM