Photogrammetric reconstruction methodology & engineering validation f. video-capt. pedestrian collis

More information

Main author

Daniel J. Melcher; Thomas N. Rush; Jay J. Przybyla; Rachel E. Keller; Paul E. Montalbano

Co-Authors

-

Type of media

PDF

Publication type

Lecture

Publication year

2015

Publisher

24. EVU Conference, Edinburgh

Citation

-

edinburg2015 10aMost empirical methodologies for correlating vehicle speed with pedestrian throw-distance have historically been developed and validated primarily with cadaver and dummy testing. In order to rigorously validate empirical and statistical modelling, it is desirable to expand the body of knowledge in the collision reconstruction community by including the results of actual vehicle-pedestrian collision events into these data sets. This paper discusses a methodology which can be used to expand the existing knowledge base about pedestrian collisions, and can also be used to reconstruct individual collision events and develop effective passive pedestrian avoidance technology.
Video-captured vehicle-pedestrian collision events provide a unique opportunity to gather data related to point of impact, point of final rest, vehicle speed, pedestrian speed, impact configuration, pedestrian response to an impending collision, and driver response to an impending collision. By applying a rigorous methodology for extracting critical information from video footage of vehicle-pedestrian collision events and combining it with traditional accident reconstruction diagrammatic and physics approaches, the reconstructionist can complete a full evaluation of the sequence, paths, speeds, distances, times, lines-of-sight, and avoidance capabilities for involved vehicles and pedestrians.
This paper will present the raw data for 6 video-captured vehicle-pedestrian collision events for incorporation into existing analysis frameworks, and will demonstrate that they further validate the empirical reconstruction methodologies that have been presented previously in the literature.

edinburg2015 10b

(EVU-members can read the full article as PDF)