计算机科学
质量(理念)
控制(管理)
事件(粒子物理)
过程(计算)
形势意识
驾驶模拟器
碰撞
车辆动力学
模拟
汽车工程
计算机安全
人工智能
工程类
操作系统
物理
哲学
认识论
航空航天工程
量子力学
作者
Timotej Gruden,Jaka Sodnik,Grega Jakus
出处
期刊:IEEE Access
[Institute of Electrical and Electronics Engineers]
日期:2023-01-01
卷期号:11: 107940-107957
被引量:1
标识
DOI:10.1109/access.2023.3320286
摘要
This paper discusses the concept of take-over (TO) in conditionally automated vehicles. Most of the current studies consider TO as a discrete event that is completed when the driver takes full control of the vehicle. Two problems with this approach are that the driver 1) needs time to gain sufficient situational awareness and 2) sometimes takes over only the lateral or only the longitudinal coordination of the vehicle, neglecting the other. To overcome these two problems and increase the quality (effectiveness, efficiency, and satisfaction) of TO, we propose two new approaches to the take-over process: partial take-overs (PTO) and assisted take-overs (ATO). The proposed PTO approach allows the driver to take over only the lateral or longitudinal coordination of the vehicle separately, instead of assuming a full TO. With ATO, the driver is monitored even after taking control of the vehicle and is assisted with automatic soft braking as well as additional warning and emergency braking if the time to collision falls below the appropriate critical levels. The approaches were evaluated in a user study with 44 participants in a driving simulator. We were able to confirm that the proposed ATO approach significantly improves the TO quality in terms of both effectiveness and efficiency without compromising driver satisfaction. Contrary to our expectations, the PTO approach did not have a significant effect on the effectiveness of TO, but only provided significantly lower reaction time to first braking and longer time to lane crossing. When combined, ATO and PTO were at least as useful as either approach individually and should be considered in future TOR user interfaces.
科研通智能强力驱动
Strongly Powered by AbleSci AI