期刊
TRANSPORTMETRICA B-TRANSPORT DYNAMICS
卷 10, 期 1, 页码 831-848出版社
TAYLOR & FRANCIS LTD
DOI: 10.1080/21680566.2020.1782786
关键词
Lane change detection; sequence learning; ADAS; deep LSTM
Lane-changing detection is a challenging task in ADAS due to the complexity and uncertainty of driving behaviors. To address this issue, an innovative DLCD system is proposed using deep learning techniques, which can model driving context in spatial domain, extract innovative features, and capture the dependencies within lane change events simultaneously. Experimental results show that the DLCD system outperforms other advanced models in learning lane change behaviors.
Lane-changing detection is one of the most challenging tasks in advanced driver assistance system (ADAS). However, modeling driver's lane-changing process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, a novel sequential model, data-driven lane change detection (DLCD) system is proposed using deep learning techniques. Firstly, DLCD system explores to modeling driving context in spatial domain instead of traditional temporal domain. Secondly, DLCD has an ability of extracting innovative features, i.e. vehicle dynamics feature, lane boundary based distance feature and visual scene-centric feature from multi-modal input data efficiently. Finally, an improved focal loss-based deep long short-term memory (FL-LSTM) network is introduced to learn co-occurrence features and capture the dependencies within lane change events simultaneously. The experimental results on a real-world driving data set show that the DLCD system can learn the latent features of lane change behaviors and significantly outperform other advanced models.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据