Safety

This paper proposes a method of achieving fast detection of pedestrians, while simultaneously
maintaining good performance regardless of variation in illumination, and in both shape and
scale of pedestrians with a single camera. Regions of interest (ROIs) are acquired by optical
flow fields using the Lucas-Kanade algorithm, and classified by convolutional neural
networks (CNNs) whether they are pedestrians. Detected pedestrians are tracked by using a
particle filter based on adaptive fusion frameworks. The CNNs allow the proposed system to
be robust to variation in illumination and in both shape and scale of pedestrians; and proposed
methods of setting ROIs and tracking pedestrians allow this system to detect a dangerous
situation and warn it to a driver fast. A single camera is only used to conduct this method, thus,
the proposed system is also economically efficient.

Pohang University of Science and Technology (POSTECH)

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

The Minnesota Department of Transportation (MnDOT) has undertaken the multi-year Connected Vehicle Safety, Mobility, and User Fee (CVSMUF) Project to demonstrate the utility of Connected Vehicle applications for the agency and the traveling public in Minnesota. As the name implies, the program addresses multiple transportation operational needs. The safety applications provide in-vehicle signing to supplement the driver’s view of the roadway and potential intersection conflicts; the mobility applications both collect and provide traffic data; and the user fee system demonstrates collection of mileage-based revenues while protecting the driver’s privacy. As a Connected Vehicle program, information is shared between in-vehicle and infrastructural systems through wireless communications. Unlike the majority of Connected Vehicle demonstrations, the in-vehicle components are developed entirely from aftermarket devices, the communications include both commercially-available 3G cellular and local agency-installed Dedicated Short Range Communications (DSRC), and the infrastructure ranges from the roadside to the back office. The program has developed and is deploying a set of applications to 500 volunteer test participants in the Twin Cities area. Complete results of the program will become available in 2013. Keywords: Minnesota, Connected Vehicle, safety, signing, probe, user fee, aftermarket.

Minnesota Department of Transportation

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

This paper discusses the design and development process for an Overheight Vehicle
Detection System (OVDS) intended for deployment at a busy interchange in central Virginia.
The primary goal of the system was to reduce the likelihood of a collision of an overheight
vehicle with the supporting girders of the overpass. This particular structure had suffered a
number of collisions, resulting in so many repairs that structures and bridge staff had
concluded that the structure could not be repaired again and would have to be replaced if
severely damaged again. The subject overpass carries a high volume of traffic on I-95 and
replacement would result in severe disruption to traffic operations and costs estimated in the
tens of millions of dollars. Getting a system in place rapidly was a critical requirement for
the Virginia Department of Transportation (VDOT).

Iteris, Inc.

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

The Oregon Department of Transportation undertook a critical review of their incident management program in the Portland Metropolitan area and concluded that certain technical and operational changes were needed to improve daily operations in incident management and to reduce operations and maintenance costs. At the heart of the revamped incident management program was a planned upgrade to the Transport Advanced Transportation Management System. The ATMS is a valuable tool in the overall strategy of highway incident management in the Portland region and its upgrade would provide an excellent return on investment for both incident management effectiveness and maintaining a reasonable level of recurring costs to operate certain automated incident management tools. As part of the planned Release 2 of the ATMS, the AVL and alpha-numeric paging functions are being integrated into the ATMS software, and operator Graphical User Interfaces are being enhanced to provide more data and more efficient data entry capabilities. These upgrades in turn necessitated installing new AVL and Paging subsystems to improve reliability and provide an easier path to integration into the ATMS. The result will be an improved incident management program operated at a lower recurring cost.

National Engineering Technology Corporation

Oregon DOT Region 1

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

This  paper  proposes  a  real-time  adaptive  control  model  for  signalized  intersections  that
decides optimal control parameters commonly found in modern actuated controllers, aiming
to exploit the adaptive functionality of traffic-actuated control and to improve the performance
of  traffic-actuated  signal  system.  This  model  incorporates  a  flow  prediction  process  that
estimates the future arrival rates and turning proportions at target intersections based on the
available signal timing plan and detector information. Signal control parameters are optimized
dynamically cycle-by-cycle to satisfy these estimated demands. The proposed adaptive control
strategy is tested on a network consisting of thirty-eight actuated signals using microscopic
simulation. Simulation results show that the proposed adaptive model is able to improve the
performance of the study network, especially under off-peak traffic conditions.

University of California, Irvine

University of California, Berkeley

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