...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Accident Analysis & Prevention >Expert Drivers' Prospective Thinking-Aloud to Enhance Automated Driving Technologies - Investigating Uncertainty and Anticipation in Traffic
【24h】

Expert Drivers' Prospective Thinking-Aloud to Enhance Automated Driving Technologies - Investigating Uncertainty and Anticipation in Traffic

机译：专家司机的预期思考 - 大声加强自动化驾驶技术 - 调查交通的不确定性和预期

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Current automated driving technology cannot cope in numerous conditions that are basic daily driving situations for human drivers. Previous studies show that profound understanding of human drivers' capability to interpret and anticipate traffic situations is required in order to provide similar capacities for automated driving technologies. There is currently not enough a priori understanding of these anticipatory capacities for safe driving applicable to any given driving situation. To enable the development of safer, more economical, and more comfortable automated driving experience, expert drivers' anticipations and related uncertainties were studied on public roads. First, driving instructors' expertise in anticipating traffic situations was validated with a hazard prediction test. Then, selected driving instructors drove in real traffic while thinking aloud anticipations of unfolding events. The results indicate sources of uncertainty and related adaptive and social behaviors in specific traffic situations and environments. In addition, the applicability of these anticipatory capabilities to current automated driving technology is discussed. The presented method and results can be utilized to enhance automated driving technologies by indicating their potential limitations and may enable improved situation awareness for automated vehicles. Furthermore, the produced data can be utilized for recognizing such upcoming situations, in which the human should take over the vehicle, to enable timely take-over requests.
机译：目前的自动化驾驶技术无法应对人类驱动程序的基本日常驾驶情况的众多条件。以前的研究表明，需要对人类驱动程序解释和预测交通情况的深刻理解，以便为自动化驾驶技术提供类似的能力。目前还没有足够的优先考虑这些预期能力，以适用于适用于任何给定的驾驶情况。为了实现更安全，更经济，更舒适的自动化驾驶经验，在公共道路上研究了专家司机的预期和相关的不确定性。首先，通过危险预测测试验证了预期交通情况的推动教练的专业知识。然后，选择的驾驶教练在真正的交通中开车，同时大声思考展开事件的想法。结果表明在特定交通情况和环境中的不确定性和相关自适应和社会行为的来源。此外，讨论了这些预期能力对当前自动化驾驶技术的适用性。本发明的方法和结果可用于通过指示其潜在限制来增强自动化驾驶技术，并且可以实现自动车辆的改善的情况意识。此外，所产生的数据可以用于识别这种即将到来的情况，其中人应该接管车辆，以便及时接管请求。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号