Citation

Hampel F, Leibner P, Manjunatheswaran H, Schotten A, Schindler C. Transp. Res. Proc. 2023; 72: 884-891.

Copyright

(Copyright © 2023, Elsevier Publications)

DOI

10.1016/j.trpro.2023.11.498

PMID

unavailable

Abstract

With the advance in technologies that enable assisted and driverless operations, developing detection systems that can detect any obstacle within a train's clearance gauge is critical. As a first step, it is vital to understand the hazardous track situations resulting in railway accidents. Currently, to the best of our knowledge, no public railway databases that document accident situations focus on obstacles on or near the tracks. In this study, we collect railway accident reports from different sources and establish a database2 of collisions and near-misses, including descriptions of the obstacle involved. Furthermore, our analysis provides a comprehensive list of obstacle classes and critical scenarios to consider during the development and testing of obstacle detection frameworks for a railway system.

Language: en

Keywords

Accidents; Automatic Train Operation; Obstacle detection; Perception System; Safety; Situations