首页> 外文学位 >Learning in the freespace driving simulator: An exploratory study with active safety systems.
【24h】

Learning in the freespace driving simulator: An exploratory study with active safety systems.

机译:在自由空间驾驶模拟器中学习:具有主动安全系统的探索性研究。

获取原文
获取原文并翻译 | 示例

摘要

Driving simulators have been used in transportation research for decades, but human subject learning is mostly unknown. Most researchers understand the need for a familiarization phase, but the point at which learning is complete and "normal" driver behavior begins is not entirely understood. This study attempts to address this issue while also assessing driver trade-off behavior in response to active safety system implementation as it relates to risk compensation theories. In other words, possible changes in driver behavior, specifically speed and lane keeping, will be evaluated around the addition of active safety systems.;Four subjects drove a 27-mile four-lane road in the Freespace driving simulator repeatedly over a five week timeframe. Data related to lane keeping, speed, and encroachment incidents was collected and analyzed subject-by-subject, by scenario, and using statistical modeling. Subject 1 proved to be an aggressive driver who took many risks, but actually did not crash. Subject 2 appeared to be a slower learner, never taking a large number of risks but consistently improving. Subject 3 began the study with two user-caused crashes and afterwards appeared to be attempting to find the right level of confidence to match the skill level. Subject 4 began the experiment having already used the simulator multiple times during the calibration and testing phase. This experience resulted in interesting results where this subject rarely tested the limits of the simulator and ended up with the least number of total encroachment incidents.;Although each subject learned differently, there were definitive personal trends in lane keeping, speed, and encroachment incidents. Long-term trends were also different between subjects, essentially blending the line between learning and potential natural fluctuations in attention, risk perception, confidence, and driving. The data fluctuation may be explained by risk allostasis theory, where drivers adapt to an ever-changing perceived task difficulty through feelings of risk. Also, there were no obvious effects of active safety systems on driver behavior and performance as applied in this study. Future research involving more subjects and simulation runs would undoubtedly give a better picture of learning in a simulator and the potential effect of active safety systems.
机译:驾驶模拟器已经在交通运输研究中使用了数十年,但人类学科学习却鲜为人知。大多数研究人员都知道需要熟悉一个阶段,但是对学习完成和“正常”驾驶员行为开始的一点还不完全了解。这项研究试图解决这个问题,同时评估驾驶员权衡行为,以响应与风险补偿理论相关的主动安全系统的实施。换句话说,将在添加主动安全系统的同时评估驾驶员行为的可能变化,特别是速度和车道保持;;四名受试者在五周的时间内反复在Freespace驾驶模拟器中行驶了27英里四车道的道路。收集与车道保持,速度和侵犯事件有关的数据,并按情景,使用统计模型对每个对象进行分析。事实证明,对象1是一个勇于进取的驾驶员,承担了许多风险,但实际上并未坠毁。主题2似乎是学习速度较慢的人,从未冒过大量风险,但一直在进步。主题3的研究是由两次用户引起的崩溃开始的,之后似乎是试图找到与技能水平相匹配的正确置信度。受试者4在校准和测试阶段已多次使用模拟器,开始了实验。这项经验产生了有趣的结果,该受试者很少测试模拟器的极限并最终导致最少的侵占事件总数。;尽管每个受试者的学习方法都不尽相同,但在车道保持,速度和侵占事件方面存在明确的个人趋势。受试者之间的长期趋势也有所不同,本质上将学习与注意力,风险感知,信心和驾驶的潜在自然波动之间的界限融合在一起。数据波动可以通过风险同化理论来解释,其中驾驶员通过风险感觉适应不断变化的感知任务难度。此外,如本研究中所述,主动安全系统对驾驶员的行为和性能没有明显影响。未来涉及更多主题和模拟运行的研究无疑将更好地了解模拟器的学习情况以及主动安全系统的潜在影响。

著录项

  • 作者

    Sandstrom, Scott Matthew.;

  • 作者单位

    Purdue University.;

  • 授予单位 Purdue University.;
  • 学科 Psychology Behavioral.;Engineering Civil.
  • 学位 M.S.C.E.
  • 年度 2010
  • 页码 168 p.
  • 总页数 168
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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