arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.08005eess.SYcs.SY
收藏

面向自动驾驶最优决策的人车交互与AI协作研究综述

Survey on Human-Vehicle Interactions and AI Collaboration for Optimal Decision-Making in Automated Driving

Abu Jafar Md Muzahid, Xiaopeng Zhao, Zhenbo Wang

AI总结:

该综述梳理了自动驾驶领域人车交互与AI协作的最新进展,分析发展历程与现存问题,介绍融合人类认知、模拟人类行为的相关技术,旨在助力开发更高效适配的自动驾驶系统。

摘要:

自动驾驶车辆的能力正快速提升,但实现完全自动驾驶仍是一项重大挑战,决策过程中仍需要人类认知的持续参与。将人类认知融入控制算法已变得愈发重要,研究人员正致力于制定策略,以尽量减少人类驾驶员与AI系统之间的冲突。尽管已取得显著进展,但诸多挑战依然存在,凸显出该领域进一步创新与完善的必要性。本综述涵盖了人车交互(HVI)与AI协作在车辆控制方面的最新进展。首先,我们梳理了HVI的发展历程,点明关键进展并指出仍存在的问题。其次,我们讨论了现有技术,包括将人类直觉与认知融入决策过程的方法,以及开发可模拟人类行为以实现最优驾驶策略、打造更安全高效交通系统的相关方法。本综述旨在通过强化人-AI协作,为开发更有效、更具适应性的自动驾驶系统提供助力。

英文摘要:

The capabilities of automated vehicles are advancing rapidly, yet achieving full autonomy remains a significant challenge, requiring ongoing human cognition in decision-making processes. Incorporating human cognition into control algorithms has become increasingly important, as researchers work to develop strategies that minimize conflicts between human drivers and AI systems. Despite notable progress, many challenges persist, underscoring the need for further innovation and refinement in this field. This review covers recent progress in human-vehicle interaction (HVI) and AI collaboration for vehicle control. First, we start by looking at how HVI has evolved, pointing out key developments and identifying persistent problems. Second, we discuss the existing techniques, including methods for integrating human intuition and cognition into decision-making processes and developing systems that can mimic human behavior to enable optimal driving strategies and achieve safer and more efficient transportation. This review aims to contribute to the development of more effective and adaptive automated driving systems by enhancing human-AI collaboration.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑