| 基于高频高精度定位信息的车辆轮廓冲突瞬时预测方法 |
| |
| 引用本文： | 吴明先,许甜,刘建蓓,赵超杰,高晋生,李志锋.基于高频高精度定位信息的车辆轮廓冲突瞬时预测方法[J].中国公路学报,2019,32(6):105-113. |
| |
| 作者姓名： | 吴明先 许甜 刘建蓓 赵超杰 高晋生 李志锋 |
| |
| 作者单位： | 1. 中交第一公路勘察设计研究院有限公司, 陕西 西安 710065;2. 交通安全应急保障技术交通运输行业研发中心, 陕西 西安 710065;3. 广东省路桥建设发展有限公司, 广东 广州 510623 |
| |
| 基金项目： | 国家重点研发计划项目（2017YFC0803900）；广东省交通运输厅2017年度重大科技专项项目（科技-2017-01-002-007） |
| |
| 摘 要： | 车辆避撞预警系统是高级驾驶辅助系统（ADAS）研究的关键内容，也是降低道路事故率的有效途径。目前，车辆避撞预警的一般实施途径是通过在车辆上布设多元传感设备进行相对间距检测，并通过智能算法对碰撞条件进行判断。但由于该方法存在传感设备成本高昂、受环境噪声影响大等缺点，应用条件仅局限于单车智能。基于此，在卫星导航、车路协同技术快速发展的背景下，提出一种依托北斗高精度定位技术，利用车载终端获取高频（5 Hz）、高精度（厘米级）车辆定位数据进行车辆避撞预警的方法，该方法立足车路协同角度，构建包括路侧北斗连续运行参考站系统、车路通信系统、车载定位预警终端的车辆避撞预警体系，并建立基于实时位置信息的车辆轮廓冲突瞬时预测模型。为验证模型可靠性，设计动、静态试验对定位精度进行验证，并在西安绕城高速约7 km试验路段开展3次实车试验，共采集约6 000个有效样本数据对轨迹预测精度进行评估。研究结果表明：静态条件下，用于评价定位精度的圆概率误差CEP50，CEP95分别为1.51，3.24 cm；行车速度为80~100 km·h-1条件下，通过2 317组数据对比分析，采用车载定位设备与成熟产品天宝接收机（亚米级精度）获取的定位数据的误差均值为1 cm，标准差为1.38 cm；行车速度为80~100 km·h-1条件下，定位数据的真实值与预测值的横向误差标准差可达厘米级，纵向误差标准差可达分米级，该级别精度可满足车辆避撞短临预警要求。
|
| 关 键 词： | 交通工程 高级驾驶辅助系统 试验研究 高精度车辆定位 轮廓冲突瞬时预测 主动安全 |
| 收稿时间： | 2018-12-09 |
Instantaneous Prediction of Vehicle Outline Conflict Using High-frequency and High-precision Position Information |
| |
| WU Ming-xian,XU Tian,LIU Jian-bei,ZHAO Chao-jie,GAO Jin-sheng,LI Zhi-feng.Instantaneous Prediction of Vehicle Outline Conflict Using High-frequency and High-precision Position Information[J].China Journal of Highway and Transport,2019,32(6):105-113. |
| |
| Authors: | WU Ming-xian XU Tian LIU Jian-bei ZHAO Chao-jie GAO Jin-sheng LI Zhi-feng |
| |
| Affiliation: | 1. CCCC First Highway Consultants Co., Ltd., Xi'an 710065, Shaanxi, China;2. Research and Development Center on Emergency Support Technologies for Transport Safety Ministry of Transport, Xi'an 710065, Shaanxi, China;3. Guangdong Road and Bridge Construction Development Co., Ltd., Guangzhou 510623, Guangdong, China |
| |
| Abstract: | The vehicle collision warning system is a key component of advanced driver-assistance systems (ADAS), which can effectively reduce the traffic accident rate. The core technology of the vehicle collision warning system involves the detection of distance to a vehicle in front by multiple on-board sensors and determination of safety indicators. However, owing to its high cost and environmentally sensitive nature, this method is not widely promoted. This paper proposed a novel method based on a BeiDou Navigation Satellite System (BDS), which has recently gained popularity in the field of transportation. First, position information at the centimeter-level was collected by the onboard terminal at a frequency of 5 Hz by using BDS-based continuously operating reference stations system (CORS) built along the highway. Second, a model was proposed to predict the vehicle outline conflict at the target moment. This was demonstrated in a 7 km section of an expressway in Xi'an. From the field experiment, 6 000 samples were collected. The results indicate:① In a static state, a centimeter precision level is achieved; ② At speeds of 80-100 km·h-1, a decimeter precision level is attained; ③ The standard deviation of the errors between the actual values and predicted values for the lateral and longitudinal distance can reach up to centimeter-level and decimeter-level respectively. The method proposed in this paper is feasible and extendable because of the high accuracy of the model. |
| |
| Keywords: | traffic engineering advanced driver assistance system experimental research high-precision position information vehicle outline conflict prediction active safety |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《中国公路学报》浏览原始摘要信息 |
|
点击此处可从《中国公路学报》下载全文 |
|
