首页> 外文会议>International Conference on Intelligent Transportation Systems >A Framework to Enhance the Transferability of the SHRP2 Naturalistic Driving Study by Considering Heterogeneity of Driver Behavior Using Spatial-Temporal Factors in a Trajectory Level
【24h】

A Framework to Enhance the Transferability of the SHRP2 Naturalistic Driving Study by Considering Heterogeneity of Driver Behavior Using Spatial-Temporal Factors in a Trajectory Level

机译:通过使用轨迹水平上的时空因素考虑驾驶员行为的异质性来增强SHRP2自然驾驶研究的可传递性的框架

获取原文

摘要

It is known that both spatial and temporal features of roadway environment can affect driver performance and behavior, which might also impact the risk of crashes. One of the most important research questions in the field of naturalistic driving study (NDS) is how the results obtained from the NDS for a specific temporal and geographical location can be transferred to other locations without introducing a bias related to the inter-driver and intra-driver heterogeneity. Hence, this study utilized a high-resolution data obtained from the SHRP2 NDS to develop a framework for capturing inherent effects of driver-behavioral factors on driver behavior and performance. The data collected from 144 drivers 16-84 years old in New York, equivalent to 200 hours and 19,200 km of driving, to examine the effectiveness of the developed method. In order to control for spatial and temporal effects, a methodology was developed in a way that similar observations, in case of spatial and temporal characteristics, were categorized and analyzed together. The results revealed that there was less intra-driver variability in speeding, lane keeping, and acceleration/deceleration behaviors when taking into account spatial and temporal characteristics, which shows the effectiveness of the developed methods to increase the transferability of the NDS results worldwide.
机译:众所周知,道路环境的空间和时间特征都会影响驾驶员的性能和行为,这也可能会影响交通事故的风险。自然驾驶研究(NDS)领域中最重要的研究问题之一是,如何在不引入与驾驶员之间和驾驶员内部相关的偏见的情况下,从NDS获得的特定时间和地理位置的结果可以转移到其他位置。 -驱动程序异质性。因此,本研究利用从SHRP2 NDS获得的高分辨率数据来开发一个框架,以捕获驾驶员行为因素对驾驶员行为和性能的内在影响。从纽约16-84岁的144位驾驶员收集的数据(相当于200小时和19,200公里的行驶时间)来检验所开发方法的有效性。为了控制空间和时间的影响,开发了一种方法,将在空间和时间特征的情况下相似的观察结果进行分类和分析。结果表明,考虑到时空特征时,驾驶员在驾驶室内,在保持车速和加速/减速行为方面的变异性较小,这表明开发的方法可有效提高NDS结果在世界范围内的可传递性。

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