Safety

With continued progress in the development and implementation of driver assistance systems and vehicle warning systems, more and more sensors and components are slated to debut on the vehicle to support more and more active safety functions, making the trend to integrate multiple safety functions in a single system increasingly appealing. In this paper, we present a highly integrated stereo vision based system that successfully encompasses a variety of active safety functions in one single module. Using a megapixel stereo image sensor, we are able to generate high resolution, high dynamic range, wide field of view intensity images as well as range images for all the supported functions. Efficient computational design and hardware implementation of vision algorithms enables the system to accomplish multiple safety functions, including forward collision warning, lane departure warning, pedestrian protection, collision mitigation by braking, adaptive cruise control, traffic sign recognition, advanced headlamp control and driver alertness warning. This integrated hardware module design paves the way for more coherent and seamless cooperation and coordination of multiple feature functions, making the presented system a powerful and cost effective onboard system for active safety applications.

Takata Holdings, Inc.

Presented at the 18th World Congress on ITS, October 2011, Orlando, Florida

This paper extensively reviews a wide range of state-of-the-art techniques for incident detection algorithms. The principles of those algorithms are introduced and the advantage and disadvantage are compared. This paper examines the current development in this research area, identifies the problems pending to be solved on this subject, and recognizes some positive experiences in AID algorithm development. This paper gives a comprehensive literature review for the researcher attempting to obtain a better understanding of AID algorithm. The conclusions of the paper provide useful insight in AID algorithm development and improvement.

AECOM USA, Inc.

Florida International University

Paper submitted for publication and presentation at the ITS America’s 2009 Annual Meeting and Exposition

The magnitude of societal and economic impacts associated with recent natural and manmade
disasters has generated heightened awareness of the importance of infrastructure
resiliency. Transportation systems are key to response and recovery at the regional level.
These systems must hold up under stress maintaining baseline service levels and must be
robust enough in physical design and operational concept to provide a degree of selfrestoration
to prevent a destructive event from becoming the catalyst for a degenerative
epoch. Implications of non-resilient transportation systems on quality of life and economic
efficacy of a locality or region are tremendous. Engineering disciplines have made great
advances in design and assessment, increasing resiliency for the built environment. However,
those involved in infrastructure investment decisions face a significant challenge when
seeking guidance for measuring resiliency for complex and adaptive systems such as
transportation. Decision makers need metrics, integrating frameworks, and decision support
tools to test investment strategies against a range of potential event sequences. The objective of this paper is to present a conceptual framework and methodology to aid in the
quantification of the concept of transportation resiliency.

Utah State University

University of Massachusetts Amherst

Paper submitted for publication and presentation at the ITS America’s 2009 Annual Meeting and Exposition

Tunnels are sensitive road segments in the road network. The singularity of the tunnels
presents characteristics such as closed environment, artificial luminosity, ventilation, a limited
number of exits, etc. These characteristics imply not only the installation of specific
equipment (CCTV, CO, fire detectors, etc.) but also the traffic management and control in the
surrounding area has to be managed in a distinct way of the rest of the elements of the road
network. When an incident nearby the tunnel, it is necessary to develop a set of measures of
management and coordination between organizations that control the surrounding area of the
tunnel because the consequences can affect the traffic in the tunnel, so new problems can
appear. The aim of the present article is to show the developed methodology by the
‘Dirección General de Tráfico’ (DGT) to create the actuation protocols face to incidents in
tunnels.

Dirección General de Tráfico

Universitat Jaume I de Castelló

Instituto de Robótica – Universidad de Valencia

Presented at the 15th World Congress on Intelligent Transport Systems, November 16-20, 2008, New York, New York

With the objective of reducing injury and accident severity, as well as improving the safety and efficiency of human driving, vision-based driver assistance system becomes an active research area among au-tomotive manufacturers, suppliers and Universities. This paper presents
the real-time monocular vehicle detection system in Ford’s Ford Telemat-ics to Safety (T2S) Concept Vehicle. The system acquires grey level images through a forward facing low light camera and achieves an average detec-tion rate of 10Hz. The monocular system was tested under different traffic scenarios (e.g., simply structured highway, complex urban street, varying
weather conditions), illustrating good performance.

Ford Motor Company

University of Nevada, Reno

Presented at the ITS America Annual Conference and Exposition, April 26 - 28, 2004 San Antonio, Texas