Click for next page ( 27

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 26
26 Lights against foliage in downhill view Lights against sky in uphill view. Figure 5. Views of the experimental truck. asked to rate the discomfort glare and urgency for each warn- ment. Trends and the correlation of the photometry and the ing light at 600 ft and 300 ft. When the last set of ratings was dynamic testing were compared to provide further insight into complete, the participants were instructed to drive back to the the lighting system performance. building where they were paid for their participation. Summary of Results Fourth Driving Session: Daytime Fog The results of the dynamic conditions--lane-change, vehi- During the fourth driving session, participants viewed each cle identification, and pedestrian detection distances--will be warning light at 600 ft and 300 ft. At each distance, subjects discussed first. These results are then followed by the analy- were first asked if they could see each warning light. If they sis of the rankings of attention-getting, confidence, urgency, answered "yes," they were also asked how confident they were and glare. that they saw the light using the confidence rating scale, to rate the attention-getting nature of the light using the attention- Dynamic Conditions getting rating scale, and to rate the urgency of the light. These questions were asked for each light and catch trial (no light pre- Lane Change Distance sented). Once both sets of ratings were done, the participants The LED warning lights caused significantly longer lane- were instructed to drive back to the building where they were change distances than either the beacons (low- and high- paid for their participation. Subjects who participated in all four mounted) or the strobes. This result may be related to driving sessions also received a $30 bonus. the light's effective intensity that caused participants to change lanes to avoid glare from the lights. Results from all other warning lights were not significantly Data Analysis different from each other. A statistical analysis was undertaken that compared the re- Vehicle Identification Distance sults using analysis of variance (ANOVA) testing for each of Vehicle identification distances were significantly shorter the independent and dependent variables. Factors and inter- (worse) in rain and fog conditions than in dry conditions actions were considered to be significant at an = 95% level. (Figure 6). When possible, a Student-Newman-Keuls (SNK) pairwise post High-mounted beacons provided significantly longer hoc test was used to further determine significant factors. (better) identification distances than the other warning The final step was the photometric analysis of the light lights (Figure 7). The light's separation from the vehicle's sources based on the measurements from the earlier experi- tail lights may have been a contributing factor because

OCR for page 26
27 Vehicle Identification Distance (ft) 2000 400 1800 Pedestrian Detection Distance (ft) 350 1600 1400 300 1200 250 1000 No Glare 200 800 Glare 600 150 400 100 200 50 0 Dry Rain Fog 0 Weather Dry Rain Fog Weather Figure 6. Mean vehicle identification distance for each glare and weather condition. Figure 8. Mean pedestrian detection distance for each weather condition. a majority of participants suggested that the tail lights were the main influence in determining that the light high effective intensity, which washed out the pedestrian's source was located on a vehicle. reflective vest. LEDs provided significantly shorter identification dis- All other warning lights produced results that were not tances than the other warning lights (Figure 7). This phe- significantly different from having no warning lights at all. nomenon may be attributed to the lights washing out the vehicle's tail lights because of their proximity and high Analysis of Ratings intensity. Pedestrian Detection Distance Attention-Getting Rating Pairwise post hoc SNK analysis found that detection dis- On average, the LEDs were rated as the most attention- tances were significantly shorter in rain and fog conditions getting, and the high-mounted beacons were rated the than in dry conditions (Figure 8). lowest. This result may be due to the relatively high A pairwise SNK analysis also found that the LEDs had a effective intensity of the LEDs and the low effective inten- significantly shorter pedestrian detection distance than sity of the high-mounted beacons. all other warning lights (Figure 9). This phenomenon Attention-getting ratings were significantly worse for all is likely attributed to a bloom effect caused by the light's lights except the low-mounted beacons during the uphill 350 1400 Pedestrian Detection Distance (ft) 300 Vehicle Identification Distance (ft) 1200 250 1000 200 800 150 600 100 400 50 200 0 0 High- Low- LED Strobe No Lights High-Mounted Low-Mounted LED Strobe Mounted Mounted Beacon Beacon Beacon Beacon Warning Light Warning Light Figure 7. Mean vehicle identification distance for each Figure 9. Mean pedestrian detection distance for each warning light. warning light.

OCR for page 26
28 7 100 Attention-Getting Rating 6 80 5 High-Mounted Beacon Confidence Rating Low-Mounted Beacon 4 LED 60 Strobe 3 40 2 1 20 Downhill Uphill Direction 0 Figure 10. Mean attention-getting rating for each High-Mounted Low-Mounted LED Strobe warning light by direction (17 scale). Beacon Beacon Warning Light condition (Figure 10). The high-mounted beacons were Figure 12. Mean confidence rating for each warning light (0100 scale). most affected by the "daylighting" effect of viewing the lights against the bright sky. Because the high- and low-mounted beacons are iden- Increasing distance caused significant drops in confidence tical lights, their different attention-getting results can be rating for the high-mounted beacons, worsening the attributed to their placement on the experimental truck. effect of viewing the light against the sky. The high-mounted beacons become much less attention- Discomfort Glare getting during the uphill conditions because the lights For the surprise trial, average discomfort glare ratings for blend in with the sky behind them. Because the back- all warning lights were better than "satisfactory." Possible ground for the low-mounted beacons remains the same explanations are (1) the participants did not look directly (i.e., the tailgate of the truck), there is no significant loss at the lights because they were unaware the lights were the in attention-getting rating (Figure 11). focus of the study, or (2) the lit roadway provided light- Confidence Rating ing on or near the roadway that made the warning lights Post hoc SNK analysis shows confidence rating was sig- seem less glaring by comparison. nificantly lower for the high-mounted beacons than all On average, the LEDs had the highest discomfort glare other lights (Figure 12). ratings, and the high-mounted beacons had the lowest LEDs and low-mounted beacons provided the highest (Figure 14). confidence ratings. The LEDs were the only warning lights to get a lower Direction had a significant impact only for the high- discomfort glare rating in fog conditions, which may mounted beacons and strobes (Figure 13). be because of increased light scatter. 7 100 Attention-Getting Rating 6 80 5 Confidence Rating 4 60 Downhill Uphill Downhill 3 Uphill 40 2 20 1 1200 2400 3600 4800 1200 2400 3600 4800 High-Mounted Beacon Low-Mounted Beacon 0 Warning Light and Distance (ft) High-Mounted Beacon Strobe Warning Light Figure 11. Mean attention-getting rating for each beacon warning light by distance and direction Figure 13. Mean confidence rating by direction (17 scale). (0100 scale).

OCR for page 26
29 9 Post hoc analysis shows the LEDs were rated signifi- 8 cantly lower for urgency than the other warning lights (Figure 16), possibly because of the relatively long flash Discomfort Glare Rating 7 duration of LEDs compared to that of the beacons and 6 strobes. Dry A significant difference in urgency rating for the LEDs was 5 Rain due to age. This result may indicate that younger partic- 4 Fog ipants judge urgency based on flash duration, while older 3 participants judge based on intensity. Distance was a significant factor for urgency, with ratings 2 decreasing as distance increased. 1 High-Mounted Low-Mounted LED Strobe Beacon Beacon Photometric Comparison Warning Light For the photometric comparison, results of the photo- Figure 14. Mean discomfort glare rating for each metric measurement for each of the lighting systems were com- warning light by weather (19 scale). pared to the performance of each system in terms of the de- pendent variables. As before, the photometric measurements All warning lights received lower discomfort glare ratings used for the comparison were derived using the Form Factor when a glare vehicle was present. This outcome may be method. because participants attributed more of the glare to the headlights, or because the addition of the headlights Vehicle Identification Distance made the warning lights seem less glaring in comparison. For vehicle identification distance, the high- and low- Distance was a significant factor in all weather conditions, mounted beacons have the same effective intensity but with discomfort glare ratings decreasing with increased different performance levels. The distance of the beacons distance. from the vehicle's tail lights, which participants used to Urgency identify the vehicle, provided the higher performance For the surprise trial, the high-mounted beacons were for the high-mounted beacons. rated significantly lower for urgency than the other warn- The LEDs had the shortest vehicle identification distance ing lights. The similar low-mounted beacons had the despite having the highest effective intensity. The light's highest average rating, possibly because the low-mounted long flash duration gave the appearance of a slow flash, beacons were closer to the participant's eye height and the which participants confused the most with other road- light was reflected off the truck's tailgate making the lights way markings such as flashing signs and construction seem more intense (Figure 15). markers. 100 100 90 90 80 80 70 70 Urgency Rating Urgency Rating 60 60 50 50 40 40 30 30 20 20 10 10 0 0 High-Mounted Low-Mounted LED Strobe High-Mounted Low-Mounted LED Strobe Beacon Beacon Beacon Beacon Warning Light Warning Light Figure 15. Mean urgency rating for each warning Figure 16. Mean urgency rating for each warning light following the surprise trial (0100 scale). light (0100 scale).