Making it Safer to Ease on Down the Road

By Kathleen Kocks

Bethany Koetje, civil engineering master’s student in NCAC, digitizes a crash test dummy.

In 2005 motor vehicle crashes were the leading cause of U.S. deaths for every age 3 through 6 and age 8 through 34. Released in April by the National Highway Traffic Safety Administration, the data are derived from the most current U.S. statistics on overall causes of death.

These facts underscore the importance of two transportation research centers at GW’s Virginia Campus—the Center for Intelligent Systems Research and the National Crash Analysis Center. Operating within the School of Engineering and Applied Science, the centers research two prongs of traffic safety: preventing motor vehicle crashes and surviving when they do occur.

GW’s transportation safety research began in 1992 with the founding of NCAC, which focused its research on improving vehicle crashworthiness. Today NCAC performs crash investigation and crash-test analysis that address vehicle crashworthiness, occupant protection, and safer roadway equipment, says professor Azim Eskandarian, NCAC’s cofounder and CISR’s founder and current director.

CISR followed in 1996 and was chartered to focus on emerging intelligent transportation technologies that could prevent crashes, Dr. Eskandarian says. “Today CISR researches intelligent in-vehicle devices as well as integrated traffic-management systems.”

CISR conducts research within four laboratories. The Driving Simulation Laboratory and the Truck Driving Simulation Laboratory perform driver behavior studies using human subjects to develop intelligent in-vehicle systems. One simulator uses a real car and wide-angle visual system to generate scenarios. The truck simulator does not use a real truck, but its cabin, driver control functions, and visual system faithfully simulate driving aspects unique to commercial trucks.

The other CISR laboratories are the Traffic and Networks Research Laboratory and the Virtual Reality Laboratory. The former uses equipment found in Advanced Traffic Management Centers to study how sensor technology can benefit traffic-flow management. The Virtual Reality Laboratory creates an interactive 3D environment to simulate multifaceted traffic-system infrastructures.

Hyundai-Kia scholarship students (from left) Seong-Woo Hong, Yong Soek Park, and Chung-Kyu Park (seated) with NCAC senior researcher Pradeep Mohan, who recently completed his doctorate at GW.

Asked to depict what CISR research might bring to the future, Dr. Eskandarian describes an integrated traffic system in which every vehicle has intelligent sensors and wireless communication equipment. Every vehicle can transmit traffic conditions to a traffic-management center, which analyzes the data and transmits traffic information back to vehicles, where it is displayed to drivers. The transmission could report traffic congestion or crashes ahead, then display alternate routes to avoid the problem.

“Additionally, intelligent devices on vehicles alert drivers to developing conditions, such as detecting an impending crash. One such device already available is adaptive cruise control. It operates like conventional cruise control but senses the car ahead. If that car slows, the adaptive cruise control automatically slows its vehicle.

“Intelligent in-vehicle systems could have different levels of operation. Some could provide autonomous driving, where the vehicle autonomously goes from point A to point B. The basic scientific and engineering tools are available; however, we are very far from having a safe autonomous system,” Dr. Eskandarian says. “Another level is semiautonomous control, where devices sense a current or developing problem and assist the driver to avoid it, for example, assisting the driver in an evasive maneuver before a collision occurs.”

A recent CISR driver-assistance project using the car simulator tested an intelligent speed-adaptation system that helps drivers maintain speed limits.

“Earlier research by others looked at two approaches. One examined purely informative systems—the car warned drivers if they exceeded a speed limit. The other examined mandatory systems that prevented the car from exceeding a speed limit. Research revealed the former system wasn’t effective, and the latter system was disliked by drivers, who would turn it off,” Dr. Eskandarian says.

“We looked at a happy medium between the two, with an advanced vehicle speed-adaptation system. We tested drivers using no system to gather their slowdown and acceleration profiles and tendency to exceed the speed limit. Based upon their profiles, we developed a system that gave drivers two warnings—yellow as they approached the speed limit and red when they exceeded their own thresholds. We also made the gas pedal stiffer as the speed limit was passed, making it harder to accelerate further. Our post-experiment survey showed drivers preferred this system to the others.”

Another CISR driver-assistance project addressed the problem of drowsy automobile and truck drivers.

“We’ve worked extensively on unobtrusive detection of drowsy drivers. Using both driving simulators, we tested individual car and truck drivers in alert and drowsy conditions and observed their performance to detect abnormalities as they tired,” Dr. Eskandarian says.

“The most dominant trait is a difference in steering patterns when drivers are drowsy. Their lane-keeping ability degrades, and they steer to the left and right more often. The performance degradation occurs three to four minutes before they get into trouble, so this gives plenty of time to generate a warning and prevent accidents.”

Other CISR research with drivers studied intelligent cruise control systems, vehicle rollover mitigation, active suspension controls, and collision avoidance. As part of the latter research, CISR is studying intelligent systems that activate when braking will not prevent an accident. The research involves driver-assistance in steering control and evasive maneuvers.

“Opportunities also exist for evaluations of distracted drivers,” Dr. Eskandarian remarks. “This is a large area, with such distractions as cell phones, conversations with others, children in the car, and more. Intelligent in-vehicle devices could also distract drivers. How do we design these to avoid that?”

Regarding traffic-management research, CISR will soon study the communications side of the integrated vehicle/traffic-management network Dr. Eskandarian envisions for the future.

“With vehicles having wireless communication, GPS tracking, and other intelligent sensors, we can obtain a rich set of traffic information from them. If traffic management centers have this information, they would continuously know traffic conditions,” Dr. Eskandarian says.

Such a network is akin to today’s air traffic control system. Aircraft carry transponders that pinpoint their positions to ATC controllers and to other aircraft in close proximity to them. This information enables ATC to safely and efficiently manage air-traffic flow. Having each aircraft know the location of nearby aircraft also facilitates safety.

“We will simulate the entire communications environment, which no one has yet done. We are studying how to establish a traffic-management system that can monitor real-time traffic conditions and predict changes down the road. This system could enable optimization of lanes on our highways; for example, switching the direction of traffic in a lane to accommodate rush-hour traffic,” Dr. Eskandarian explains.

“We’re also researching what backups would be necessary if the system has a breakdown. The first stage will be to develop tools that simulate the details of the communications systems and can be integrated with traffic-modeling methods. This enables total simulation of intervehicular and vehicle-to-infrastructure communications, along with various traffic congestion and safety mitigation methods.”

Improving Survivability

While CISR studies the potential for intelligent systems to prevent crashes and improve traffic infrastructures, NCAC takes a broader approach to improve crash survivability.

CISR researchers monitor characteristics of drowsy driving exhibited by truck drivers in the driving simulation lab.

An international authority in vehicle safety, NCAC is a cooperative effort among GW, the Federal Highway Administration (FHWA), NHTSA, and industry experts—including top automakers. NCAC has the world’s most comprehensive library of crash-test data, vehicle reports, and computerized finite element (FE) models of vehicles, crash dummies, and roadside equipment created by NCAC—all available to the public.

NCAC houses the Vehicle Modeling Laboratory, which uses reverse engineering of real vehicles to create FE vehicle models for simulation-based research. Another lab is the Hyundai-Kia Automotive Safety Research Laboratory; it is funded by a multiyear grant from the carmakers to improve crashworthiness of vehicles and child-restraint systems. The center also includes the High-Performance Computing Laboratory to support research projects. Additionally, NCAC operates a full-scale, outdoor crash test laboratory at FHWA’s Turner-Fairbank Highway Research Center in McLean, Va.

Research covers three areas: vehicle safety and biomechanics, highway safety and infrastructure, and simulation and advanced computing.

Vehicle safety research concentrates on vehicle crashworthiness. NCAC uses real vehicles at the crash-test facility and FE vehicle models in simulated scenarios to analyze crashworthiness from pre-crash, crash, and post-crash perspectives. The research includes crashes of cars, light trucks, tractor-trailer trucks, and even motorcycles.

To accomplish this, NCAC has created 15 FE models of vehicles and more than 20 FE models of roadside equipment that accurately reveal damages during simulated crashes. More than 20 NCAC models are industry standards for automotive testing.

A relatively new vehicle project at NCAC addresses crashes between traditional sedans and the SUVs and crossover models that are more prevalent today.

“The federal government asked us to develop new test methodology to evaluate vehicle compatibility. This involves how vehicles interface with each other or with roadside equipment during impacts,” explains professor Cing-Dao “Steve” Kan, NCAC’s director. “We are working with automotive manufacturers to analyze different vehicle structures, simulate impacts, and evaluate how energy is distributed during impacts.”

NCAC’s biomechanics research brings in the human factor, studying different types of crashes to prevent injuries. NCAC uses sophisticated human, child, and infant dummies in its actual crash tests, and it has developed equally sophisticated FE dummy models for its crash simulations. This research has resulted in the addition of side-impact airbags and improved restraint systems in today’s cars.

The study of restraint systems is a large part of NCAC research. Helping fund this work is a multimillion dollar grant from Ford Motor Co. that mandates special attention to child safety research.

“As a part of that grant, we are developing a test methodology for child safety seats,” Dr. Kan reports. “This is a difficult issue because you have to consider the vehicle structure, the interface between it and the child seat, and the child.

“We’re working with automotive manufacturers, child seat manufacturers, and test dummy manufacturers on this project. One important recommendation we made speaks to the need to redesign pelvic areas of child dummies to better represent a child. We are also working with dummy manufacturers to develop adult and child dummies that mimic human movement during impacts from different impacting directions, improving upon today’s dummies that are limited in their movement in just frontal and side directions.”

Civil engineering master’s student Joseph Cuadrado develops a computer model of an anthropomorphic testing device in the NCAC Vehicle Modeling Lab.

NCAC also is researching the safety of elderly vehicle occupants, Dr. Kan reports. “Our research shows that elderly people are subject to different kinds of injury risks. To explore the topic fully, we are recommending that the government conduct research at vehicle speeds lower than the current 35-mph test standard.”

For its occupant research, NCAC is developing additional FE dummy models. These models simulate a 50th percentile male, a 50th percentile female, a 3-year-old child, and a 6-year-old child. They are being used in simulations of frontal and side-impact crashes.

NCAC’s research also benefits the medical community. Using its vast vault of crash data, NCAC can predict injuries caused by different types of accidents. For example, side-impact crashes often cause aortic injuries. If aware of this, emergency responders can immediately address the possibility of aortic injury and save lives.

Such findings fueled NCAC’s participation in the Crash Injury Research and Engineering Network (CIREN). This NHTSA program involves eight Level 1 trauma centers working with academia, industry, and government safety experts to create a database of severe motor vehicle crashes, crash reconstruction, medical injury profiles, and treatment. CIREN’s goal is to improve emergency processes and victim outcomes.

“We also coordinated with GW’s Medical Center, emergency responders, and public health groups regarding emergency triage,” Dr. Kan adds. “It is very important for doctors to understand injury risks following certain accidents. We’ve used computer simulation to educate rescue people to injuries crash victims could have. When the victim arrives at the emergency room, the medical staff already is aware of possible injuries.”

NCAC’s injury research may play a role in today’s automatic crash notification systems. One such system, General Motor’s OnStar, uses sensors to detect a crash and contacts the system’s network to send help. Sensor data about the impact could be combined with NCAC crash-injury data to inform responders about possible injuries.

Other NCAC vehicle-research projects address rollover risk, airbag system performance, and automotive fire safety. However, one of NCAC’s newest projects reflects the future direction of automotive design.

“We are doing research for Miles Electric Vehicles. They make small electric-powered vehicles that are used on large campuses or industrial sites, but they want to expand their line,” Dr. Kan explains. “They developed a battery-operated vehicle prototype for roadway use and asked us to study the vehicle’s safety performance and help them comply with safety requirements to make it street legal.”

Sometimes it’s not the car impact that causes injuries, but what the car impacted. NCAC’s highway safety and infrastructure research addresses this issue by studying roadside hardware to reduce injury.

Roadside hardware is any equipment routinely found along highways. NCAC has researched designs of Jersey wall concrete barriers, heights for guardrails, impact effects of curb and medians, and designs of breakaway sign supports.

“We recently received a grant from the Transportation Science Academy to evaluate cable-medium barriers used nationwide. The designs vary greatly, with different cable tensions and configurations. We are investigating all available systems and contacting all state governments to survey which barriers they use. We will gather crash data from the states and perform computer simulations to evaluate the performance of different barriers. Working with the states, federal government, and barrier manufacturers, we will develop new design, installation, and maintenance guidelines for these barriers,” Dr. Kan says.

NCAC’s roadside hardware research attracted a new customer a few years ago—the U.S. Department of State. Following terrorist attacks on some federal government facilities, the State Department asked NCAC to test anti-ram protective barriers and other security equipment at federal sites. NCAC found the chinks in the armor, resulting in improved security to protect U.S. personnel and property.

The enabling technology for NCAC research is its advanced computing equipment at the High-Performance Computing Laboratory. The lab has a parallel-processing Silicon Graphics Origin 3800 server and more than 60 associated processors, a gigabit Ethernet network of SGI workstations, and a Force10 E300 switch/router. Compatible simulation and design programs include LS-DYNA, MSC software, MADYMO, and Hypermesh.

This equipment enables researchers to use 3D computer-aided graphics to develop FE models, create animated crash simulations, analyze damages, and determine design changes to reduce damage. The lab also provides digital media streaming for NCAC’s massive library and operates computer, network, and email services for GW’s entire Virginia Campus.

The expertise within NCAC attracts projects outside of its traditional research. One such project involves aircraft engines.

“The FAA wants to use simulation to certify new derivatives of existing engines. We are helping the FAA work with the analysis community and engine manufacturers to ensure standardized modeling techniques and consistent methodology are used for the simulation,” Kan explains.

“For this multiyear research project, we are initially developing computer simulation of engine containment. This involves developing models of materials used inside the engines and using simulation to determine whether the engine’s design will ‘contain’ these materials inside the engine’s casing in the event of an engine fan breakout.

“The materials being researched include composites, aluminum, titanium, and Kevlar fabric. This work will eventually benefit our vehicle research activities,” Dr. Kan adds.

Research at CISR and NCAC has already contributed much to roadway safety and efficiency. Given the capabilities and expertise at the two centers, the potential to do more is obvious.

back to top

© 2008 The George Washington University
The George Washington University is an equal opportunity/affirmative action institution.