Loading [a11y]/accessibility-menu.js
Effect of Visual Instruction on Drivers’ Reaction Pattern in Lateral Conflicts | IEEE Conference Publication | IEEE Xplore

Effect of Visual Instruction on Drivers’ Reaction Pattern in Lateral Conflicts


Abstract:

Side-impact accident is the most common form of traffic accident in China and contributes the most fatalities. Driver factor is widely believed to be the main causation o...Show More

Abstract:

Side-impact accident is the most common form of traffic accident in China and contributes the most fatalities. Driver factor is widely believed to be the main causation of those accidents. During lateral conflicts at intersections, braking with same-direction-swerving are used too often by drivers under many inappropriate conditions, resulting in traffic accidents which could have been avoided. In this driving simulation study, arrow-shaped visual instructions are displayed in front of drivers, in order to guide them and reduce their dangerous avoidance reaction pattern described above. Firstly, in pretest, subjects' natural emergency avoidance behavior during intersection conflicts was tested without any instructions; Later, in posttest, arrow-shaped visual instructions were shown in front of subjects when conflicts were triggered; Finally, one month later, drivers were asked to attend a tracking test with the same visual instructions displayed. Result shows that without instruction, subjects prefer to use braking and same-direction-swerving irrationally in lateral conflicts. Although such pattern is not helpful for avoiding crashes, they didn't manage to correct themselves during trials. In posttest, arrow-guidance is found significantly helpful for drivers to make better longitudinal and lateral avoidance choices. However, positive impact of visual instruction drops significantly in tracking test one month later. Learning effect is observed among trials in both posttest and tracking test. This study provides a data basis for driver's decision-making mechanism research under conflict conditions, and could be used for future intelligent connected vehicles' human-machine interaction design as well as collision-avoidance system development.
Date of Conference: 27-30 October 2019
Date Added to IEEE Xplore: 28 November 2019
ISBN Information:
Conference Location: Auckland, New Zealand

Contact IEEE to Subscribe

References

References is not available for this document.