ACKNOWLEDGEMENT We wish to thank - PDF document

ACKNOWLEDGEMENT We wish to thank Ichikoh Industries, Ltd. for their this research. The increasing availability electrochromics liquid crystals rearview mirror reflectivity promises dramatically improved control mirrors. These technologies allow mirror reflectivity to allow automatic adjustment response to changing lighting determine optimal mirror reflectivity given set have framed determining optimal reflectivity in rearview mirror functions: providing rearward vision, glare from the driver glare. All The directions that the basic, function of providing rearward vision off with better rearward vision, forward vision optimizing reflectivity level simply have the three ways, and valid relative could do those things, optimal reflectivity level given set however, the measurement (Flannagan, Sivak, & Gellatly, 1991) rearward vision discomfort glare found that, a standard & laboratory (Sivak, Flannagan, Ensing, & Simmons, 1991), visibility. Specifically, the benefit that resulted reduced mirror from the Schmidt-Clausen the eye from the rear. the most optimization problem or she see it. which must correct, the extremely four- reflectivity level that the antiglare setting According to this view, the only important things drivers rearview mirrors are the dark conditions glare reduction issue those vehicles of rearview mirrors) may be implicitly undervalued in the way reflectivity is controlled. This may be especially important in situations in which glare hm rear and the attempting a lane change a multilane has been use mathematical modeling to Although there stimuli drivers them, how they need to them, we have made some assumptions about then used this type the effects changing reflectivity. There exist mathematical models that can be used rearview mirrors discussed above. They been used extensively in the domain headlighting (e.g., Bhise et al., 1977), their extension rearview minors is straightforward. However, a standard discomfort glare seems systematically violated in the rearview mirrors of changing reflectivity (Flannagan al., 1991). have concentrated purtion the overall tradeoff and rearward seeing. visibility modeling a model published Internationale de 1'Eclairage (CIE) formulation follows Farber (1988) for use in modeling headlamp the model the observer, visual performance, (in terms angle subtended), background against appears, the illumination at the observer's eye a glare and the angle subtended the observer the stimulus. We extended minors images seen through mirror the reflectivity minor. this report tradeoff rearward seeing in specific scenario that we designed actual driving. implemented this as a laboratory experiment mathematical modeling laboratory setup. the scenario, a driver using low headlamps on straight, level �fnm an interior rearview while attempting stimulus (representing a standing beside (representing the a vehicle in lane). Mirror reflectivity is effects on luminance contrast necessary to see each compare the to the Twenty paid younger group 25 years and ten All subjects had natural corrected visual 20/30 test using Landolt schematic diagram laboratory setup 1, The subject was seated in a mockup 1985 Chrysler electrochromic rearview The mirror connected to voltage source that allowed any three preset voltages thereby setting spaced approximately 0.73,0.21, There was no window mockup. from the subject's eye point, there white screen. second white screen behind the subject. The from the the mirror from the to the Three slide projectors Two were access projectors, mockup. They were the two behind the rear screen, hole in to provide glare that the subject through the rearview mirror. All three projectors were equipped electronically controlled shutters. rear screens evenly, illuminated lamps shielded baffles. Each cd/m2. value has visual adaptation with no fmed lighting (Olson, Aoki, Battle, & Flannagan, through the mirror (i.e., the stimulus it been seen and 26.4 deg2. the same height as 20.3 cm (3.17 deg) fhm the glare source. control stimulus luminance, the front rear (rectangle) variable voltage for the bulb and 21 neutral density standard slide The voltage to the bulb coarse adjustment During the the random among the neutral density ranged from density values slide holder) to 2.0 in 0.1 steps. That corresponds transmission from constant factor During actual stimulus the images from the increments over the existing measure the luminance measured them other lights subjects the luminance increments (using the cam2 for the hnt stimulus and 0.80 cd/m2 for the rear adjustment permitted the neutral encompass the thresholds the younger subjects. older subjects their detection thresholds, projector voltages individually before the determined subject's front thresholds approximately and then that the neutral density roughly centered luminance increment density slides) cam2 front stimulus cam2 stimulus. After these individual voltage settings, the neutral density were used glare projector. Instead, light directed at mockup and centered on the rearview mimrr. neutral density that remained experiment. The glare source the direction thus constant That intensity would have at the observer's eyes had been This level is probably often world (Olson & Sivak, the shutter thus subtended 0.40 were produced the combination the three rearview single glare values at the eye point values at eye point corresponding (lx) Visual detection thresholds using random double staircase procedures procedure involves each trial this case subject said that not). The stimulus level presented trial of the stimulus and to looking directly were run blocks, one for mirror reflectivity stimulus direction. informed the beginning each block, about the the mirror. involved running double staircase stopping condition the number The number per block averaged about seconds. The the six blocks each subject. Before each trial, experimenter selected proper neutral density Nter verbal reminder and then that operated the glare the stimulus trial. The simultaneously, staying After each trial subject indicated verbally whether or she the stimulus. The effect of age was significant, F(1,17) required higher contrast the younger group This effect in Figure with the the predicted the younger reasonably accurate, (i.e., older subjects predicted). The on the within each that the confidence older subjects, reflecting performance varied more among the older subjects that the confidence interval nevertheless excludes the predicted value, indicating that better vision volunteer for studies stimulus direction, F(1,17) = 140.92, p = .0001, and mirror reflectivity, F(2,34) = 12.86, p = .0001, as was their interaction, F(2,34) panel shows younger group, the bottom panel shows predictions for contrast is required detect rear probably partly general dimming minor, even at the rearward the glare source was much stimuli shows that the significant strong interaction stimulus direction. The is almost entirely caused threshold for (meaning that rearward vision improves Threshold for the front stimuli also changes opposite direction; stimulud threshold increases reflectivity (meaning that forward vision worse as minor is so data groups (and statistically significant, parameters). Although small to prevent the effect for rear stimuli from appears to be roughly the reduced are and the rearward seeing are than predicted. the observed disparity in cost the predicted disparity. the observed results, costs rearward seeing are large relative to corresponding benefits forward seeing, evaluate the practical significance the selection it is necessary to know more about the and rearward stimuli, and the distances other circumstances which they need to It is weighting scheme seems clear informally that valid set strongly emphasize see forward, limitations, the and benefits thought provoking. small relative rearward seeing, combined with this extreme disparity in benefits and rearward costs, might lead observed values slope parameters log contrast threshold a linear function of mirror fit for and each subject. Predicted values are derived from REFERENCES Bhise, J. B., with headlights #770238). Warrendale, Automotive Engineers. 1'Eclairage (1981). An analytic model for describing the influence of lighting parameters upon visual performance: Vol. 2. Summary and application guidelines (Publication CIE 1912.2). Journal of Farber, E. (1988). Revising the DETECT seeing distance model (SAE Technical Paper Series #880716). Warrendale, Automotive Engineers. rearview mirrors: Driver Technical Paper #900567). driver visibility & discomfort glare electrochromic rearview minor (SAE Technical Paper Series #910822). for optimal Automotive Engineers. D. J. Design parameters automotive interior #870635), Automotive Engineers. Aoki, T., Battle, D. S., & Development of a UMTRI-90-41). Ann Arbor? The University of Michigan Transportation Research Institute. Olson, P. L., & (1984). Glare from automobile rear-vision mirrors.