Safety

In order to realize  vehicular following control without longitudinal acceleration information, the paper presents a  TSM control algorithm with non-singular, time continuous and fast convergent characteristics. Aiming at TSM control theory itself,  a  non-singular fast terminal sliding function is proposed to improve convergent speed in larger range.  And control law is synthesized with a terminal attractor with state’s exponential factor to obtain time continuous control input. By analyzing the feature of minimum sensor car-following system, a two-state integrating model including vehicular and inter-vehicular longitudinal dynamics is built and subsequently  its  TSM controller is designed. Theoretical analysis and simulation experiments
indicate that this controller outputs a time-continuous throttle angle and  tracks the preceding vehicle smoothly. A driver desirable tracking precision can still be guaranteed when such uncertainties as parametric error of vehicle mass and preceding vehicle’s acceleration exist.

State Key Laboratory of Automotive Safety and Energy, Tsinghua University

Presented at the ITS America Annual Conference and Exposition,November 16-20, 2008, New York, New York

In this paper, floating car data (FCD) are utilized to detect traffic congestion which happened
around the intersections in city road network. The actual real-time data of the city of Ningbo,
China, are presented to describe the characteristic of traffic  flow when congestion appears.
Then the basic traffic jam detection algorithm is developed. According to the detection results,
the traffic management center (TMC) can position the traffic congestion accurately in short
time. The validity of this traffic congestion detection method is proved by video surveillance
system of central Ningbo city  and the detecting performances are evaluated under different
traffic congested conditions.

Ningbo YiChuang Information Technology Co.

The Research Center for Software Engineering Technology of Anhui Province

College of Computer Science and Technology, Zhejiang University

The Road and Traffic Management Bureau of Anhui Province

Presented at the ITS America Annual Conference and Exposition,November 16-20, 2008, New York, New York

Geomagnetic vehicle detector (GVD) is a kind of traffic monitoring equipment
based on the effect that passing vehicles disturb the geomagnetic field. Having the advantages
of small size, easy installation and slight damage on the road surface, GVD is to be an
attractive alternative to inductive loops. Currently studying on GVD is usually based on the
vertical component of geomagnetic field and state machine algorithm, and is able to count
traffic flow, sort vehicle type and measure vehicle speed. But the precision of these algorithms
is usually affected by baseline drift and vehicle speed. A novel GVD method is presented here
based on the sophisticated and widely applied magnetic declination measurement technology.
With a proof test in the lab and data collected from field test, this paper proposed a vehicle
detecting technology on the state machine algorithm. The results of experiment and analysis
indicate that the GVD based on magnetic declination measurement is a proving traffic
monitoring technology.

College of Information Engineering, China Jiliang University, Hangzhou 310018,China

Presented at the ITS America Annual Conference and Exposition,November 16-20, 2008, New York, New York

The U.S Government and the automotive industry have developed and tested prototype systems to evaluate the effectiveness of using Dedicated Short Range Communications (DSRC) for vehicle safety and non-safety applications. While initial test results have indicated promise for these systems, benefits for cooperative vehicle safety applications are proportional to the number of vehicles equipped. Systems tested to date have used expensive, high-performance, “fully embedded” architectures to prove these systems are technically feasible. To expedite reaching installed equipment penetration levels necessary to provide visible collision avoidance benefits to drivers on a regular basis, viable deployment scenarios include options for retrofit or aftermarket devices for equipping the existing vehicle fleet. This paper provides test results comparing currently used and more cost-effective methods of obtaining accurate and timely dynamic vehicle information (i.e., yaw rate and longitudinal acceleration).

DENSO INTERNATIONAL AMERICA, INC.

Presented at the ITS America Annual Conference and Exposition, May 3-5, 2010, Houston, Texas

Image-based lane departure warning systems (LDWS) use software that analyze the images collected by a small camera to track the pavement markings and predict when a vehicle unintentionally drifts out of the travel lane.  The performance of LDWS is affected significantly by moderate to heavy rain during night-time for typical pavement marking installations.  It is expected that the performance of LDWS during these conditions can be enhanced by improving pavement marking quality.  Such improvements could include frequent replacements of conventional pavement markings or the installation of wet-reflective markings that are visible during night rainy conditions.

This paper investigates the cost-effectiveness of pavement marking improvement alternatives, taking into consideration their effects on LDWS performance.   The results indicate that some of the investigated pavement marking improvement alternatives can be justified from a benefit-cost analysis point of view depending on traffic volume, truck proportion, road segment horizontal alignment, number of related crashes at night rainy conditions, shoulder width, and LDWS market penetration.

Prasoon Sinha, PE, PTOE; Mohammed A. Hadi, Ph.D., PE; and  John R. Easterling IV, P.E., PTOE

Presented at the ITS America Annual Conference and Exposition, May 3-5, 2010, Houston, Texas