Research on train anti-collision method based on deep learning

DOI

10.2991/iceesd-18.2018.198How to use a DOI?

Keywords

Ranging, Collision Avoidance, Safe Driving, Driving Efficiency

Abstract

Real-time detection of train distance from the front of the obstacle to ensure that the train in the case of safe braking speed driving is to improve the efficiency of train driving the main way. In this paper, a collision avoidance method based on deep learning is proposed. The system acquires the obstacle, the orbit environment and the signal status in front of the train through the long and short focus cameras, and then uses the depth learning image processing method to give the distance between the vehicle and the obstacle and the signal color. On the basis of ensuring safe braking, a recommended speed and driving route are given to the user, and when the signal light is in a red state and the current train is far away from an obstacle, voice prompts and alarms with different frequencies are respectively provided to ensure the safe and effective running of the train, Multiple test data show that this method can accurately determine the obstacle distance and orbit environment and signal status, and can give different voice prompts according to demand, to ensure the safe driving of the train and improve the driving efficiency and train defense Hit the coefficient, with good practicality.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jianming Zhang
AU  - Qijin Lu
AU  - Shuting Zao
PY  - 2018/05
DA  - 2018/05
TI  - Research on train anti-collision method based on deep learning
BT  - Proceedings of the 2018 7th International Conference on Energy, Environment and Sustainable Development (ICEESD 2018)
PB  - Atlantis Press
SP  - 1086
EP  - 1091
SN  - 2352-5401
UR  - https://doi.org/10.2991/iceesd-18.2018.198
DO  - 10.2991/iceesd-18.2018.198
ID  - Zhang2018/05
ER  -