高级驾驶员辅助系统
车头时距
车道偏离警告系统
计算机科学
预警系统
主动安全
速度限制
事件(粒子物理)
避碰
碰撞
模拟
汽车工程
运输工程
计算机安全
工程类
人工智能
电信
物理
量子力学
作者
Wenjing Zhao,Siyuan Gong,Dezong Zhao,Fenglin Liu,N.N. Sze,Mohammed Quddus,Helai Huang
标识
DOI:10.1016/j.eswa.2023.121733
摘要
Advanced driver assistance systems (ADASs) can effectively enhance driving and safety performance. Due to the inherent limitations of in-vehicle technologies concerning information sharing, existing studies mainly focus on demonstrating the effectiveness of onboard sensor-based individual ADAS functions rather than their collaborative effectiveness. Thanks to the emerging connected vehicle (CV) technologies, it is viable to physically realize the collaboration and coordination between different ADAS functions. This study aims to synthesize seven ADAS functions into an integrated advanced driving assistance system (iADAS) in a CV environment. The seven ADAS functions include omni-directional collision warning, lane-change warning, curve speed warning, emergency event notification, car-following guidance, identification of variable speed limits, and information services. The activation indicators and activation conditions of these ADAS functions are derived based on a single local coordinate system. This derivation has considered the real-time motion states of the ego and surrounding vehicles. Different ADAS functions are classified based on their roles in accident reduction, traffic efficiency enhancement, and driver convenience improvement. Afterwards, their priority of releasing information is set by considering the classification and primary functionality. Finally, the effectiveness of the iADAS is validated in field tests. Testing results reveal that the iADAS helps reduce rear-end collisions and prevent rollovers or sideslips on curved roads. Furthermore, younger drivers respond faster with higher driving stability regarding lateral collision warnings. Young, well-educated, and low-risk taking drivers maintain a short but safe time headway to the leading vehicle.
科研通智能强力驱动
Strongly Powered by AbleSci AI