A Study on Obstacle Detection with Automatic Pedestrian Tracking at Railway Level Crossings by using Laser Range Scanners

Abstract

In this paper, the concept of novel obstacle-detecting methods for railway level crossings, which are capable of detecting pedestrians and bicycles besides motorcars based on both the laser range scanner technologies and digital signal processing techniques, has been proposed. The mathematical models for pedestrian tracking by using Kalman filter have been investigated. The performances of both the obstacle detection and pedestrian- tracking algorithm have been verified through field tests at an operating level crossing. The obstacle-characterisation-based countermeasures against level crossing accidents have been discussed. The test result shows that this method is promising from the points of views of preventing accidents and reducing operational cost.