Achilleas Kourtellis

Naturalistic Driving Study: Field Data Collection

This report describes the six Strategic Highway Research Program 2 (SHRP 2) naturalistic driving study (NDS) data collection centers and documents their data collection activities and strategies. The study centers were located in Bloomington, Indiana; State College, Pennsylvania; Buffalo, New York; Tampa, Florida; Durham, North Carolina; and Seattle, Washington. They collected data from more than 3,000 volunteer participants and their vehicles over a 3-year period. Information is provided on the recruitment and assessment of test participants, installation of NDS data acquisition systems into participant vehicles, management of the enrolled participants and their vehicles, retrieval of data from the vehicle fleet, and lessons learned. The report will be of interest to analysts wanting background on where and how the NDS data were collected and to researchers planning future large-scale NDS projects.

Rearview Video System as Countermeasure for Trucks’ Backing Crashes: Evaluating the System’s Effectiveness by Controlled Test

In general, the operation of large trucks involves many different types of maneuvers. The backing maneuver, in particular, requires a higher level of driver attention because of the limited rear view. A growing number of trucks in the United States are equipped with a rearview video system (RVS) that can help the driver see much of the area behind the vehicle. An RVS consists of one or more cameras and one monitor. It is expected that an RVS can help drivers reduce potential backing crashes. To evaluate the effectiveness of the system, this study performed a controlled driver test with 45 truck drivers. The test used three backing maneuvers and a pedestrian dummy for observation of potential crashes. The results showed that the use of an RVS increased the stop rate of the drivers during the straight-line backing maneuver by 46.7%, which could be interpreted as an increase in the odds of avoiding potential backing crashes during the backing maneuver. The stop rate increased 4.4% and 17.8% for the offset right backing and alley dock backing maneuvers, respectively. Driver age, commercial driving experience, and experience with an RVS showed no statistical association with the increased stop rate, which means an RVS can be adopted by drivers quickly. In general, drivers showed positive attitudes toward using an RVS, and more than 90% of respondents agreed that an RVS could reduce the rear blind spot for large trucks.

A Pilot Study on Interactions between Drivers and Pedestrian Features at Signalized Intersections-Using the SHRP2 Naturalistic Driving Study Data

This paper aims to assess driver behaviors at signalized intersections with four identified pedestrian features (“Stop Here on Red,” “No Turn on Red,” “Turning Vehicles Yield to Pedestrians,” and “Right on Red Arrow after Stop”) via a pilot study by using an innovative safety data: Strategic Highway Research Program 2 (SHRP2) Naturalistic Driving Study (NDS) data. The major findings from the pilot study include: (1) “No Turn on Red” has the highest rate of compliance (70%), followed by “Right on Red Arrow after Stop” (67%), “Turning Vehicles Yield to Pedestrians” (67%), and “Stop Here on Red” (55%); (2) compared to control group (without pedestrian features), three features (“Stop Here on Red,” “No Turn on Red,” and “Right on Red Arrow after Stop”) increased the likelihood of compliant behaviors; and (3) mid-aged drivers (25-59) showed the highest percentage of compliance at feature sits (83%), followed by older drivers (60+, 69%) and young drivers (16-24, 61%).

Do as I Say, Not as I Do: Observed Compliance Versus Stated Understanding of Pedestrian Crossing Laws

Florida has consistently had one of the highest pedestrian fatality rates in the country. NHTSA data indicate that Florida has had one of the four highest pedestrian fatality rates since 1994. In response, the Florida Department of Transportation launched the Bicycle–Pedestrian Focused Initiative to increase awareness of and decrease fatalities among bicyclists and pedestrians in Florida. A study conducted by the Center for Urban Transportation Research at the University of South Florida tracked the crossing behavior of individuals at high crash locations and then asked them about their understanding of pedestrian traffic laws. The results showed that most of the people surveyed about the traffic laws knew the correct answer and followed the law. The study found that 82% of the people interviewed crossed at the crosswalk and correctly answered the question about whether it was illegal to cross midblock; 57% said that they pressed the pedestrian push button when it was available and also were observed to press it; 64% of people were observed to start to cross on green pedestrian time and also said they did so when asked. Further, the study compared the behavior of people exposed to the safety campaign before the survey and also how their behavior changed when they were first asked the questions and then were observed during their crossing. The data showed that people exhibited safer behavior when they were asked the questions first and observed to cross afterward. The study showed results for several attributes, including crossing location, pedestrian signal indication while crossing, pedestrian push button use, and distraction.

Evaluation of video camera system to reduce side collisions of transit buses

An aftermarket sideview video system has great potential to reduce side crashes of transit buses. This paper presents a research project that investigated the use of a sideview video system as a countermeasure against most problems with mirrors that caused side crashes in transit operations. Measurements showed that the video camera system with a regular-angle lens could reduce about 64% of the blind zone of a flat mirror system. The video system could reduce about 40% of the blind zone of a common system that combined flat and convex mirrors. With a wide-angle lens, the blind zones on both sides of transit buses could be completely eliminated. The controlled driving test, designed to determine whether drivers could use the system to perform everyday maneuvers, showed no statistically significant difference between distances perceived with the mirrors and with the camera system in static and dynamic conditions. During testing, several advantages of using a camera-based system over one with mirrors were observed. The testing confirmed that the system could be implemented and that further enhancement of the performance of the system could improve its potential over mirrors used on transit vehicles, with a resulting increase in safety and reduction in crashes.

Mixed Effects Logistic Model to Address Demographics and Neighborhood Environment on Pedestrian Injury Severity

This paper examines the effects of demographics and neighborhood environment on pedestrian injury severity to inform proactive countermeasures for improving pedestrian safety. A mixed effects logistic model addressing unobserved heterogeneity was developed from 3,948 pedestrian-involved crashes that occurred in Florida from 2011 to 2014. Six normally distributed random parameters were identified to reflect random effects on the pedestrian injury severity. The heterogeneity of two demographic factors (older and male pedestrians) suggested the need for more customized education programs to improve pedestrian safety awareness and knowledge, especially for older pedestrians. Relative to low-income areas, 67.7% of pedestrians involved in crashes in higher-income areas were less likely to sustain severe injury. Analysis of sample data also indicated that low-income areas tended to have had more unsafe behaviors by pedestrians related to higher injury severity (e.g., crossing at dark in unlighted areas). Higher-income areas tended to have had more unsafe behaviors by drivers related to higher injury severity (e.g., distracted driving). Other significant factors included lighting conditions (daylight, darkness without lighting), speed limit, alcohol or drug impairment, dart or dash behavior, crossing indicator, and traffic control device indicator. Regarding neighborhood land use types, two indicators about the presence of bus stops and department stores or supermarkets nearby were significant, and their effects were also random. Further investigations are needed to identify systematically the need for effective countermeasures in severe injury crash clusters in the future.

Planning and Design for Prevention of Incorrect Turns onto Highway-Rail Grade Crossings

In recent years, a number of injuries and fatal collisions have occurred at certain highway-rail grade crossings located adjacent to highway intersections, driveways, or freeway on-ramps. Some guide signage, pavement markings, and other traffic control devices currently present at the crossings may have confused drivers and caused them to turn onto the railroad tracks rather than at the nearby intersections, driveways, or freeway on-ramps. Due to limited research in this new area of prevention of incorrect turns at highway-rail grade crossings, a research project was sponsored by the National Center for Transit Research and Florida Department of Transportation to seek cost-effective solutions. This paper first provides an intensive literature review to identify available sources of information and best practices throughout the world for innovative and cost-effective infrastructure treatments. It then presents analysis of the contributing causes of crashes that occurred in the past and offers a set of countermeasures for planning and design aimed at improving incorrect turns at highway-rail grade crossings.

Prototyping of a portable data logging embedded system for naturalistic motorcycle study

The primary objective of this work is to design a highly portable data logging embedded system for naturalistic motorcycle study with capability of collecting many types of data such as images, speed, acceleration, time, location, distance approximation, etc. The proposed embedded system design is based on an Arduino microcontroller and is capable of storing up to 220 hours of text/image data during a one month study period. We have successfully designed and implemented the system. The data acquired has been validated and found to be accurate.

Evaluation on the Effectiveness of Sideview Video Systems to Reduce Transit Bus Side Crashes

A camera-based system has great potential to reduce transit bus side crashes. This paper evaluated the effectiveness of sideview video systems to reduce transit bus side crashes through measuring the reduction of blind zones, and analyzing the results of controlled driving tests and driver surveys using sideview video systems. The results from measurements of blind zone reduction showed that the sideview video systems with a regular angle lens can reduce about 64 percent of the blind zones of a flat mirror system. It can reduce about 43 percent of blind zones of a common combined flat and convex mirror system. The results of the controlled driving test from 28 bus drivers was positive on distance/depth perception and lane change maneuvers using sideview video systems. The driver survey results also confirmed that bus drivers participating in the study valued the benefits of sideview video systems.