Search Content

This report summarizes the findings of a human factors analysis to determine the effects of advanced warning flashers (AWFs) on simulated driving performance. The Minnesota Department of Transportation sponsored the project. Researchers used the flat-screen simulator at the University of Minnesota Human Factors Research Laboratory to conduct experiments. They measured vehicle speed, braking, and acceleration/deceleration during simulated driving and visually observed stopping behavior. In addition, they analyzed responses to a post-test questionnaire. They created a 11.3-mile simulated driving environment with 10 signalized intersections and configured four experimental models: low speed limit (SL) of 50 miles per hour with no AWFs, low SL with AWF at each intersection, high SL of 65 miles per hour with no AWFs, and high SL with AWF at each intersection. Researchers set different vehicle-signal proximity intervals, with all green/ no yellow as the control, and zero seconds with the vehicle adjacent to the signal, two seconds, three-and-a-half seconds, or five seconds. With each model, they assigned two intersections each proximity interval, with the sequence of intersection proximity intervals ordered differently for each model. Each of 24 subjects completed duplicate driving trials with each model. The study revealed that, relative to intersections with no AWFs, drivers who encountered yellow signals at AWFs intersections: stopped more frequently at low SLs but not at high SLs, drove more slowly while approaching intersections with two and three-and-a-half second proximity intervals, and displayed less inconsistent behavior at intersections with short proximity intervals. Researchers concluded that AWFs assist drivers with decision-making behavior and promote safer driving behavior. They recommended field research to study and actual environment.

This study analyzed the relationship between the size of the forward looking blindspot (FLB) produced by vehicles A-post (windshield frame), the speeds of two vehicles approaching an intersection at right angles, and driver behavior relative to a likely accident event. Researchers observed 28 volunteer participants directly and by four channels of on-board video cameras while they drove in a simulator at the Human Factors Research Laboratory. They noted the way that participants scanned the virtual environment and scored at four levels of scanning activity. They also tracked visual acquisition of the target vehicle and incidence of collision. Only 6.3% of the total fell into type one scanning (eyes fixed). Type II (eyes only) accounted for the highest incident rate at almost 44%. The study considered both as "inactive" forms of scanning. Target vehicle acquisition rate increased with the activity level of the scanning type. The target acquisition rate increased significantly from scanning level one to level two and from scanning level two to level three. There was not a significant increase in the acquisition rate from scanning level three to level four. Not surprisingly, collision rates decreased with increases in scanning level. Collision rates significantly dropped between scanning levels two and three and scanning levels three and four. Yield signs at intersections produced no significant correlation with acquisition rate, collision rate, or scanning level.

This study explores three techniques of signage in an attempt to reduce the incidence of vehicle/deer collisions on highways in Minnesota. A simulated environment was created along a stretch of U.S. Highway 23 near Marshall, Minnesota with participants chosen from the University of Minnesota and the surrounding community. The simulation consisted of a standard warning sign as well as a prototype of the experimental signage. The prototype was comprised of a beacon light attached to the top of the warning sign designed to flash when deer were present. During the simulation, participants were exposed to the standard signage as well as the new signage with and without the beacon flashing. The main objective was to determine whether the prototype signs would modify driver behavior such that they decreased their speed. The study found that the prototype signage was effective in decreasing the speed of the participants when the beacon light was flashing. These results were consistent across the variations of age and gender. The results for the prototype signage with the beacon light turned off were essentially no different from the standard signage.

Traffic data from 1996 and continuing through the first half of 2001 were analyzed for roadways with low to very low traffic volumes in ten southwestern Minnesota counties resulting in information about the causes of crashes. Three sets of analysis were carried out on the database. First was a descriptive analysis of the data to determine the general frequency rates of accidents. A second identified dangerous roadways. Counting the number of crashes on specific roadways and dividing this number by the average daily traffic (ADT) on a roadway generated crash rates for those roadways, including county state aid highways (CSAHs), county highways, and township roads. Roadways with the highest 5% were considered significantly dangerous. Crash rates were generated for specific locations. This method identified 15 dangerous locations, nine on CSAHs, three on county highways, and two on township roads. There were only 235 cases where no improper driving was indicated. The remaining 1,554 cases suggested that driver error was the major cause. The most likely factor in causing an accident on a highway with an ADT of less than 400 is a crash involving an animal. Road design factors such as number of lanes and the speed limit seem to be the factors related to these accidents.

This study evaluated the efficacy of active versus passive warnings at uncontrolled pedestrian (ped) crosswalks (Xwalks), by comparing how these two warnings types influenced behavior of drivers approaching such Xwalks. Vehicle-Xwalk interactions were observed at 18 sites with passive, continuously flashing, or ped-activated warnings, yielding 7,305 no ped present and 596 ped present interactions. Vehicle velocities and accelerations were averaged for each interaction. Findings show no significant effect of warning type on overall velocities for either interaction type. With peds present only, for average velocities at successive 5m distances from the Xwalk, a downward trend in velocities from 25 to 5m is observed for passive and active warning sites, but not for pedactivated warning sites. Various lines of evidence point to a number of sources of ambiguity regarding the salience of uncontrolled Xwalk warnings, resulting in behavioral uncertainty by drivers interacting with such warnings. Mixed findings on effects of warning type in this study point to the need for further analysis of this problem area.

Simulation offers a cost-effective way to conduct research on collision avoidance and accident prevention. To be effective, simulated performance must be a valid measure of real world performance. This project sought to validate real world driving performance based on the performance of individuals driving in simulation. The study presents performance data on 14 male and 12 female volunteer subjects who drove a route adjacent to the University of Minnesota campus and then performed in a similar computer-generated driving route. Generally, subjects reported the simulated driving test comfortable and realistic; performance and characteristics of driving in the simulator closely paralleled the real world; the qualitative pattern of driving was similar; and errors and the control parameters of driving performance suggested acceptable reliability between both driving worlds. Researchers concluded that the simulator performed reliably and provided a valid set of performance data that could be used to better understand driving behavior, especially as it related to accident prevention and collision avoidance.

This project evaluated how driver interaction with an in-vehicle navigation system (IVNS) affects driving performance and safety. Researchers collected measures of simulated driving performance during interaction by 13 different subjects with an IVNS digital map display, using a Honda Acura placed within a fixed-base wrap-around driving simulator. Subjects (Ss) navigated along a maze-like route laid out within a simulated road grid. Dummy Global Positioning System (GPS) coordinates, corresponding to the position of the vehicle in the grid, were transmitted to the IVNS and updated continuously as vehicle position in the simulation environment changed. A digital map of the grid, with an icon representing vehicle representing vehicle position superimposed, was displayed on a laptop computer placed in the Acura. Under the control condition, Ss were not given turn instructions. Results indicate that for the test relative to the control condition: * Visual interaction with the IVNS display was greater and task completion times longer. * More variability in vehicle control was observed for measures of average vehicle speed, peak speed, percent braking time, peak braking pressure, and vehicle heading. Subjective responses from simulated driving and a separate group of on-road Ss identify both navigation benefits and possible safety problems with the system. It is a reasonable assumption that increased variability in driving performance elevates driving accident risk. Both the simulated driving and subjective response results, therefore, point to possible safety implications in IVNS use for the driving public. The findings suggest that as IVNS use becomes more widespread, both navigation benefits and possible adverse driving safety effects of such systems need to be considered.