C. Schwarz, T. Brown, R. Sherony

Pages: 47-62

Abstract
Pedestrian fatalities are on the rise, having reached their highest number since 1990, many of them occurring at night. Meanwhile, new headlamp technologies are making possible lighting concepts that are more adaptive and intelligent than ever before. Designing systems to make interactions between drivers and vulnerable road users (VRU) more salient has the potential to reduce collisions by alerting both parties to the presence of the other. Investigation into pedestrian and bicyclist fatalities reported in the FARS and GES databases from 2010-2014 revealed the top crash types. The final set of scenarios selected for testing included overtaking a bicycle along the side of the road, bicycle crossing at a junction, bicycle crossing mid-block, pedestrian crossing mid-block, left turn across bicycle path, and left turn across pedestrian path. Six driving simulator experiments were designed to test the top crash scenarios using the National Advanced Driving Simulator (NADS) at the University of Iowa. We created three system functions and combined them to form a conceptual adaptive headlamp system (AHS). The first function was a spotlight that shined towards the base of the VRU. The second was a brightening of the headlamps. The third was an icon painted on the road next to the VRU. The most meaningful effects were observed with the use of the spotlight. The use of a brightness factor and icon had more mixed effects. The results reported do not address whether these functions would aid the VRU’s awareness of the vehicle – that was evaluated in a pedestrian and bicyclist simulator. Through this research, we hope to illuminate the most important features of such systems and alert designers to their potential benefits and any limitations.
Keywords: adaptive headlamps; driving simulation; road safety; vulnerable road users; nighttime