Thomas Schnell

Improved Flight Technical Performance in Flight Decks Equipped With Synthetic Vision Information System Displays

Synthetic vision information systems (SVIS) are the next generation of cockpit display systems that will become an integral part of the commercial flight deck in the future. The goal of this study was to evaluate SVIS against conventional glass cockpit displays to assess whether or not SVIS can improve pilot performance, situation awareness (SA), and workload. The evaluation was conducted in a fixed-base laboratory flight simulator located in the Operator Performance Laboratory at the University of Iowa. A total of 12 pilot evaluators hand-flew fairly difficult approaches into the Eagle County, Colorado, airport. Specially designed questionnaires were used in various phases of the flights to estimate workload, SA, and user preference. Information relating to SA was collected using the Situation Awareness Global Assessment Technique and workload estimates were obtained using the NASA Task Loading Index technique. Eye movement data were collected using a modified IScan ETL 500 system, and analyzed to find a link analysis (an analysis of eye movement transitions between areas of interest on the displays), scan length, fixation frequencies, location of fixations, and duration of fixations. Flight technical error performance was derived. Analysis results indicate that the SVIS is superior to the conventional displays for the majority of the measures that were obtained in the experiments.

VISIBILITY OF ROAD MARKINGS AS A FUNCTION OF AGE, RETROREFLECTIVITY UNDER LOW-BEAM AND HIGH-BEAM ILLUMINATION AT NIGHT

A study was conducted to obtain visibility data for old and young subjects performing a pavement marking end detection task on a fully marked road under headlamp illumination conditions. A total of 20 healthy subjects (10 old and 10 young) were used. It was found that observer age had a highly significant effect on pavement marking visibility. The effect of headlamp illumination (high-beam versus lowbeam) was found to be rather weak, probably because of the low-beam headlamp pattern, which favors the right-edge line. Pavement marking retroreflectivity was found to be significant at the levels investigated in the experiment. The new yellow dashed centerlines (tape) with a gapto-stripe ratio of 9.1/3 had a retroreflectivity of RL = 222 mcd/m2/lx and RL = 399 mcd/m2/lx under the medium and high conditions, respectively. The new white edge lines (tape) had a retroreflectivity of RL = 268 mcd/m2/lx and RL = 706 mcd/m2/lx under the medium and high conditions, respectively. The average end detection distances ranging from 124.8 m (average preview time 5.1 s for 88.5 km/h) for the old group under low-beam illumination on the medium retroreflectivity road to 237.3 m (average preview time 9.7 s for 88.5 km/h) for the young group under high-beam illumination on the high retroreflectivity road may appear to be rather long. However, the reader should be aware that markings as bright as the ones used in this research are the exception instead of the rule in real-world driving situations. The aim of this study was to determine the end detection distances obtainable under fairly favorable pavement marking material conditions.

Minimum In-Service Retroreflectivity of Pavement Markings

Minimum in-service retroreflectivity values for pavement markings are presented based on visibility computations performed with the CARVE (Computer-Aided Road-Marking Visibility Evaluator) computer model. CARVE accurately computes all geometric and photometric relationships for each headlamp separately; applies the human visual luminance contrast threshold database from Blackwell (Part III, 1946) adjusted by a field factor function that has been obtained from a number of pavement marking visibility field experiments; and provides retroreflectivity values for the pavement markings for any selected single-point geometry (e.g., ASTM 30-m geometry, observation angle = 1.05°, entrance angle = 88.7°). Based on the CARVE computation results, a set of in-service pavement marking retroreflectivity values are derived for fully marked, dark, straight, and dry roads using paint-and-beads pavement markings. The derived minimum retroreflectivity values for fully marked roads without raised pavement markers (RPMs) are highly speed dependent, because the computations are based on a constant minimum preview time of 3.65 s (3.0 s true preview and 0.65 s for eye-fixation duration). A separate set of minimum retroreflectivity values, based on a constant preview time of 2.0 s, is provided for fully marked roads with RPMs in good working order. It was found that the minimum retroreflectivity requirements for pavement markings could be substantially relaxed if RPMs (in good photometric working condition) were used. The proposed minimum retroreflectivity values are based on a 62-year-old driver (about the 85th percentile of the licensed driver population, about the 95th per-centile of the nighttime driver population based on trip frequency data as a function of the time of day).

EOG artifact removal using a wavelet neural network

In this paper, we developed a wavelet neural network (WNN) algorithm for electroencephalogram (EEG) artifact. The algorithm combines the universal approximation characteristics of neural networks and the time/frequency property of wavelet transform, where the neural network was trained on a simulated dataset with known ground truths. The contribution of this paper is two-fold. First, many EEG artifact removal algorithms, including regression based methods, require reference EOG signals, which are not always available. The WNN algorithm tries to learn the characteristics of EOG from training data and once trained, the algorithm does not need EOG recordings for artifact removal. Second, the proposed method is computationally efficient, making it a reliable real time algorithm. We compared the proposed algorithm to the independent component analysis (ICA) technique and an adaptive wavelet thresholding method on both simulated and real EEG datasets. Experimental results show that the WNN algorithm can remove EEG artifacts effectively without diminishing useful EEG information even for very noisy datasets.

Evaluation of Traffic Flow Analysis Tools Applied to Work Zones Based on Flow Data Collected in the Field

The Ohio Department of Transportation (ODOT) has identified the maintenance of traffic (MOT) as a top priority to serve the motoring public as part of a department strategic initiative. A key component of this strategy is to ensure that traffic flows efficiently through work zones. Work is described that was performed for ODOT to determine whether commercially available traffic simulation models could be calibrated to yield accurate queue length and delay time predictions for planning purposes in freeway work zones. Four work zones on multilane freeways were selected by ODOT for collection of the calibration data. Traffic flow video records were obtained at the four selected work zones by two ODOT video recording vans equipped with 15-m masts. Traffic flow parameters were extracted from the video records with the Mobilizer-PC software package. The traffic simulation and prediction tools investigated included the Highway Capacity Software (HCS), Synchro, CORSIM (under ITRAF and TRAFVU), NetSim, and a macroscopic model called QueWZ92. Simulation models were constructed with all models for the selected work zones, and the simulated queue lengths and delay times were compared with the data extracted from the field data with Mobilizer-PC. The results of this study indicated that the microscopic simulation packages could not be calibrated to the oversaturated conditions that existed at the work zones. The calibrated microscopic simulation packages underestimated the length of the queues that formed in the real world. The macroscopic QueWZ92 produced more accurate estimates than did the microscopic packages.

Effect of Luminance and Text Size on Information Acquisition Time from Traffic Signs

Although visual performance was measured in terms of visual response time in many psychophysical studies, such an approach has not been used in evaluating the effect of luminance on traffic sign legibility performance. Traffic sign and retroreflective sign sheeting performance at night have been historically identified with the threshold (farthest) distance for legibility and in many cases from stationary vehicles with no restrictions on viewing time. Because traffic signs are not always read at threshold distances or threshold luminances and because the time available to read traffic signs is usually limited in the real world, a proper assessment of sign legibility performance requires determining information acquisition times above threshold conditions. This study investigated the effect of (legend) luminance and letter size on the information acquisition time and transfer accuracy from simulated traffic signs. Luminances ranged from 3.2 cd/m2 to 80 cd/m2 on positive-contrast textual traffic sign stimuli with contrast ratios of 6:1 and 10:1, positioned at 33 ft/in. and 40 ft/in. legibility indices, and viewed under conditions simulating a nighttime driving environment. The findings suggest that increasing the sign luminance significantly reduces the time to acquire information. Similarly, increasing the sign size (or reducing the legibility index) also reduces the information acquisition time. These findings suggest that larger and brighter signs are more efficient in transferring their message to the driver by reducing information acquisition time, or alternatively, by increasing the transfer accuracy. In return, reduced sign viewing durations and increased reading accuracy are likely to improve roadway safety.

Visibility of New Centerline and Edge Line Pavement Markings

The objective of the presented two-part study was to provide nighttime pavement-marking visibility data obtained under automobile low-beam illumination conditions in the field for further calibration of the Ohio University pavement-marking visibility model CARVE (Computer Aided Road Marking Visibility Evaluator). A total of 10 subjects participated in Part 1 of the study. The goal of Part 1 was to determine the end-detection distance of finite-length, new, medium-retroreflectivity, yellow pavement marking tape centerlines without edge lines as a function of the centerline configuration (single dashed, single solid, and double solid) and width. The results of Part 1 confirm that an increase in the average end-detection distance of about 55 percent represents the approximate upper limit of what can be achieved by adding more retro-reflective area to a yellow centerline. Part 2 consisted of four main experiments involving 10 subjects each and two follow-up experiments involving a total of 10 subjects. The goal of Part 2 was to provide pavement-marking end-detection distance data for fully marked roads consisting of left and right white edge lines with either a yellow double solid or single dashed centerline. The main effects of width, retroreflectvity, and centerline configuration were found to be statistically significant. A fully marked road consisting of both edge lines and a centerline generally appears to provide end-detection distances that are on average about twice as long as the end-detection distances that can be achieved with a centerline alone, without edge lines.

PERFORMANCE EVALUATION OF PAVEMENT MARKINGS UNDER DRY, WET, AND RAINY CONDITIONS IN THE FIELD

The nighttime visibility of three types of pavement markings—a largebeaded permanent pavement marking and two types of patterned pavement marking tapes, one with high-index beads and the other with mixed high-index beads—was evaluated under dry, wet (just after rainfall), and simulated rain conditions (ongoing 1-in./h rainfall). The dependent measures were pavement-marking end-detection distance and retroreflectance. The experiment was conducted at the 3M Transportation Research Center in Cottage Grove, Minnesota. This facility features a level and straight roadway section where 1-in./h rainfall can be simulated. Eighteen subjects between the ages of 55 and 75 participated in the study. The participants drove an experimental vehicle on dry and wet test sections with the pavement-marking treatments. Participants noted the earliest point at which they were able to see the end of the right edge line pavement markings. The retroreflectance of each pavement-marking material was also measured with three handheld retroreflectometers under the three weather conditions, according to ASTM E1710 (dry), ASTM E2177 (wet recovery), and ASTM E2176 (continuous wetting) standards. In terms of end-detection distances, the patterned tape with mixed high-index beads performed best under all three weather conditions. The paint and large-bead pavement markings were comparable to the tape with high-index beads under wet and rainy conditions. Under dry conditions, paint and large-bead pavement markings performed the worst. The detection distances seemed to strongly correlate with the standard retroreflectances only if the retroreflectance data from the corresponding test method are used.

Driver-Headlamp Dimensions, Driver Characteristics, and Vehicle and Environmental Factors in Retroreflective Target Visibility Calculations

The inputs that are necessary for comprehensive visibility models are discussed. Inputs required by visibility models may deal with the physical arrangement of the observer and headlamps in three-dimensional space, the headlamp characteristics in terms of candlepower output and efficiency, observer age, observation time, probability of detection, and minimum preview time. There are also a number of environmental and vehicular parameters such as horizon sky luminance, atmospheric transmissivity, and windshield transmission. One problem associated with using comprehensive visibility models is the lack of standards, guidelines, and recommendations for the required model inputs. Consequently, it is currently difficult or even impossible to duplicate visibility model runs reported by others. Standardized headlamp-observer-target scenarios, standardized observer characteristics, and standardized environmental parameters would greatly facilitate cross-validation of model outputs. Numerous model input dimensions and values that are based on the technical literature are provided here. However, because there are no comprehensive eye location percentile dimensions available for late-model automobiles and for recent anthropometric dimensions, an exploratory vehicle survey was conducted and the measured vehicular dimensions were combined with anthropometric dimensions extracted from the 1988 U.S. Army personnel survey. The result of this effort is a table that shows the driver eye locations as a function of the population percentile, observer gender, and vehicle class. Further results include data pertaining to the location of low-beam headlamps. The recommended model input values should be considered for adoption as a standard that would facilitate easier cross-validation of visibility model results.

LEGIBILITY OF TRAFFIC SIGN TEXT AND SYMBOLS

An exploratory daytime and nighttime sign recognition and legibility field driving experiment involving 11 new reflectorized right-shoulder-mounted traffic signs and 10 young, healthy subjects (three replications) is described here. It was hypothesized that daytime conditions would provide longer recognition and legibility distances than nighttime conditions and that signs with bold symbols would provide longer recognition and legibility distances than textual signs. Also of interest was the formulation of the instructions to the subjects, which is likely to influence the legibility and recognition distances. Whereas other researchers have encouraged subjects to guess the symbols or legends, it was emphasized in the instructions of this experiment that the subjects were to say aloud the information on the traffic signs at that point during the approach when they could clearly (with near 100 percent certainty) identify all visual details of the message or the symbol. By comparing these results with the results of other studies, it was found that legend and symbol identification under clearly seen conditions provides legibility and recognition distances that are considerably shorter than those obtained if guessing is allowed. The average daytime legibility and recognition distances were about 1.8 times longer than the average nighttime legibility and recognition distances. It may be tentatively concluded that when reporting on field legibility research, it is of utmost importance to provide the reader with detailed information about how the subjects were instructed to respond to the stimuli (e.g., guessing was allowed versus clearly seen message). From the data obtained in this study, it appears that the effect of the subject instruction formulation is stronger under nighttime low-beam conditions than during daytime.