Object Detection As A Safety Check For Human Factors In Operating Remotely Controlled Bridges
E.A. de Groot (TU Delft - Civil Engineering & Geosciences)
PHAJM Gelder – Graduation committee member (TU Delft - Safety and Security Science)
Aaron Yi Ding – Graduation committee member (TU Delft - Information and Communication Technology)
D.F.J. Schraven – Graduation committee member (TU Delft - Integral Design & Management)
C.L. Kraaijenbosch – Mentor (Municipality of Zaanstad)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
According to a report published by the Dutch Safety Investigation Board in early September 2019, the safety of remotely controlled bridges is not su_cient (Onderzoeksraad voor de Veiligheid, 2019, pg.58). This report was published after the occurrence of two severe accidents in Zaandam, on the Den Uylbrug and the Prins Bernhardbrug. On both occasions, the victims were standing on the movable part of the bridge deck during the opening of the bridge, and despite being visible for over a minute on the camera screens, were not observed by the operators, making the accidents human factor-based. The Dutch Safety Investigation Board concluded that part of the problem was safety mainly being considered a technical problem, instead of an integral one. In this research, the goal was to analyse how object detection could provide decision support for mitigating human factors for operating remotely controlled bridges. This was done by identifying the problems through literature studies, interviews and observations, and by building a proof of concept to mitigate these problems. Finally, this model was evaluated to gain experimental insights into the possibilities of object detection as a decision support tool.