期刊文献+

基于Cardinal样条的车道偏离预警测评关键参数估计

Estimation of Key Parameters for Evaluation of Lane Departure Warning Based on Cardinal Spline
下载PDF
导出
摘要 为了对车道偏离预警系统开展客观定量的测试评价,进行了相应的研究。目前国内外车道偏离预警系统发展迅速,但针对车道偏离预警系统的测试和评价依然十分缺乏,虽然我国已有标准《智能运输系统车道偏离报警系统性能要求与检测方法》(GB/T 26773—2011),但该标准仅给出了测评场景及流程,未给出相关测评指标的具体计算方法。针对上述问题,提出了一种基于Cardinal样条的车道偏离预警系统测评关键参数估计方法,该方法首先分析了不同的GNSS/INS组合算法的优劣,选择深组合算法作为高精度差分组合导航系统的融合算法;其次利用上述组合导航系统分别获取了车道线以及测试车辆的基础参数数据,包括车道线位置、车辆位置、速度、加速度等;然后对采集到的离散车道线坐标进行Cardinal样条拟合,从而获取连续且平滑的车道线曲线,同时与Hermite样条拟合算法进行对比试验,验证了Cardinal样条拟合可以得到更高的测量精度和更好的拟合效果;在得到高精度车道线的基础上,结合牛顿迭代算法,进行了车辆与车道线距离、偏离速度等车道偏离预警测评关键参数的解算,最后经过了试验验证。结果表明:所提出的方法能够对车道偏离预警系统中的关键参数进行准确的估计和测试,为车道偏离预警系统的定量化评价提供了详细的指标及计算方法。 In order to test and evaluate the lane departure warning system objectively and quantitatively, the corresponding research is carried out. At present, the lane departure warning systems at home and abroad are developing rapidly, but the test and evaluation of lane departure warning system are still very lacking. Although China has the standard of Intelligent Transport Systems-Lane Departure Warning Systems-Performance Requirements and Test Procedures(GB/T 26773—2011), the standard only provides the evaluation scenario and process, and does not give the specific calculation method of relevant evaluation indicators. To solve the above problem, a method for estimating the key parameters for evaluating lane departure warning system based on Cardinal spline is proposed. First, the advantages and disadvantages of different GNSS/INS combination algorithms are analyzed by this method, and the deep combination algorithm is selected as the fusion algorithm of high-precision differential integrated navigation system. Second, the basic data of lane lines and tested vehicles, including lane line position, vehicle position, speed, acceleration, etc., are obtained by using the above integrated navigation system. Then, the collected discrete lane line coordinates are fitted with Cardinal spline to obtain the continuous and smooth lane line curve, and a comparative experiment with Hermite spline fitting algorithm is carried out to verify that Cardinal spline fitting can obtain higher measurement accuracy and better fitting effect. On the basis of the obtained high-precision lane lines, combining with Newton iterative algorithm, the key parameters for evaluating lane departure warning, such as the distance between vehicle and lane line and departure speed, are calculated. Finally, the experimental verification is conducted. The result shows that the proposed method can accurately estimate and test the key parameters of the lane departure warning system, it provides detailed indicators and calculation method for the quantitative evaluation of lane departure warning system.
作者 李宏海 陆红伟 卢立阳 徐启敏 LI Hong-hai;LU Hong-wei;LU Li-yang;XU Qi-min(Research Institute of Highway,Ministry of Transport,Beijing 100088,China;School of Instrument Science and Engineering,Southeast University,Nanjing Jiangsu 210096,China)
出处 《公路交通科技》 CAS CSCD 北大核心 2022年第7期157-165,共9页 Journal of Highway and Transportation Research and Development
基金 国家重点研发计划项目(2018YFB1600804)。
关键词 智能交通 智能车路系统 Cardinal样条拟合 车道偏离预警系统 牛顿迭代算法 ITS intelligent vehicle-road system Cardinal spline fitting lane departure warning system Newton iterative algorithm
  • 引文网络
  • 相关文献

参考文献9

二级参考文献68

  • 1许斌,王勇,顾浩,冯义.战场情景模拟及其在 SGI ONYX/RE_2 实时多媒体环境下的系统实现[J].中国图象图形学报(A辑),1997,2(2):171-175. 被引量:3
  • 2孙振平.ALV体系结构与局部路径规划.国防科技大学硕士学位论文[M].,2000..
  • 3晏东.移动机器人局部路径规划方法研究.国防科技大学硕士学位论文[M].,1998..
  • 4吴晖.ALV路径跟踪控制方法研究.国防科技大学硕士学位论文[M].,1997..
  • 5孙振平 安向京.自主地面车辆系统中延迟问题的研究[M].中国控制与决策学术年会,2000..
  • 6张新晖.基于分布式多Agent的自主式智能机器人建模研究[M].杭州:浙江大学,1999..
  • 7王寻羽.分布式虚拟环境系统的研究[M].杭州:浙江大学,1999..
  • 8[1]Broggi A. Vision-based driving assistance[J]. IEEE Expert,Intelligent System & Their Application, 1998, 13(6): 22~23.
  • 9[2]Murphy R R. Sensor and information fusion for improved visionbased vehicle guidance[J]. IEEE Expert, Intelligent System & Their Application, 1998, 13(6): 49~56.
  • 10[3]Bertozzi M, Broggi A. GOLD: A parallel real-time stereo vision system for generic obstacle and lane detection [J]. IEEE Transactions on Image Processing, 1998,7(1):62~81.

共引文献158

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部