Citation

Wee, Jun Jie and Tee, Connie and Goh, Michael Kah Ong (2022) Forward Collision Warning for Autonomous Driving. Journal of Logistics, Informatics and Service Science, 9 (3). pp. 208-225. ISSN 2409-2665

Full text not available from this repository.

Abstract

Reckless driving poses great danger to users and vehicles on the road. Studies have shown that reckless driving accounts for 60% of traffic accidents every year. Reckless driving can be caused by various factors including wine racks, drag racing, sleep deprivation and inexperienced driving. Due to these reasons, autonomous driving has received immersive attention in in the recent years. Forward collision warning is one of the core safety components in the development of autonomous vehicle. A forward collision warning system issues an early warning when a potential collision is detected in front of the ego vehicle. This paper presents a pipeline approach for visual-based forward collision warning. Deep learningbased object detection and lane detection modules are integrated to sense the environment around the ego vehicle. If an object is sensed ahead of the vicinity of the ego vehicle, a warning will be triggered. A mean average precision 0.5 (mAP 0.5) of 37.2 has been achieved with the proposed method. Empirical tests show that the proposed approach can work well with different road conditions including straight and curved roads, junctions, as well as different times of the days (e.g. days and nights).

Item Type:	Article
Uncontrolled Keywords:	Autonomous driving, object detection, lane detection, computer vision
Subjects:	R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	31 Oct 2022 03:19
Last Modified:	31 Oct 2022 03:19
URII:	http://shdl.mmu.edu.my/id/eprint/10562

Downloads

Edit (login required)