车辆导航系统中基于三维电子地图的单GPS车道级定位方法研究

在对三维电子地图建模的基础上,提出了一种基于单GPS的虚拟差分、高程辅助定位和碰撞检测误差消除相结合的车道级车辆定位方法.该方法将车辆定位分为两次地图匹配过程,即首次通过虚拟差分技术和高程辅助定位技术对GPS定位参数进行修正,提高定位精度,然后在三维电子地图中结合碰撞检测技术实现车道级匹配.通过实地DGPS跑车对比实验,证明该方法完全能够满足车道级路径引导的要求.     

基于地空信息融合的无人机交通状态感知方法研究

现有的无人机（UAV）交通状态感知方法，主要针对宏观交通状态参数的获取，同时尚未克服UAV自运动对交通参数检测精度的影响，难以满足智能交通系统对于高精度微观交通参数的应用需求。为此，提出一种基于地空信息融合的UAV交通状态感知方法，该方法包括：地空信息融合模型、道路关键点（IKP）检测及跟踪、车辆目标检测及追踪算法和交通状态参数提取及估计。其中，地空信息融合模型利用地基信息（IKP世界坐标）与空基信息（IKP像素坐标）进行最优化融合，并通过自适应IKP追踪算法与自适应UAV位置偏移判断算法实时更新模型参数，以此克服UAV自运动对车辆轨迹精度的影响，进而获取可靠的车辆级（瞬时速度、车头间距和车头时距）与车道级（车道动态密度、车道流量和空间平均车速）交通状态参数。利用提出的感知方法获取实地拍摄视频的车辆级交通参数并进行了分布检验，同时比较了基于不同交通流模型的车道级参数估算方法。结果表明：该方法在车辆检测的mAP@0.5指数超过90%，同时提取的车辆轨迹相对完整，获取的车辆级和车道级交通状态参数也符合实际交通流状况。最后，将该模型应用于实地道路的交通拥堵检测及交通事件检测，该研究结果为UAV在现代交通感知和管理中的应用提供了一种理论和技术参考。     

A proposed vision and vehicle-to-infrastructure communication-based vehicle positioning approach

It is essential to obtain accurate location of vehicles for new applications of Intelligent Transportation Systems. To remedy the defects of present Global Positioning System and vehicle-to-infrastructure (V-I) positioning technology, a new positioning approach based on vision and V-I communication is proposed. This approach aims at lane-level positioning with lower cost than conventional ones. In this approach, the position of the vehicle is represented by its lateral position (the lane number) and longitudinal position (the distance from entrance of the road) in a course coordinate system along the road; the specific lane the vehicle is occupying (the lane number) can be judged using the information of lane lines detected by vision systems; then the distance to the vehicle is obtained by a Road Side Unit (RSU) during the V-I communication; and the longitudinal position is calculated. The error of the approach on typical operating conditions is analyzed, indicating that the new approach can achieve the accuracy of less than 0.31 m for straight road and 0.58 m for typical arc road with ultra-wideband communication and ranging technologies and rational arrangement of RSUs. The feasibility of this approach is presented.     

基于事故特征耦合影响的城市道路交通事故影响分析

城市道路交通事故发生后,由于事故车辆占用车道,使得车辆通行的车道数目减少,道路的通行能力降低,造成排队和交通拥堵,对交通运行产生一定的影响.以双向6车道的城市道路为例,运用Vissim仿真软件模拟交通事故下的交通运行,分析车流量、占道类型、事故持续时间以及借道超车4种因素下的交通影响.结果表明,流量越大、事故持续时间越...     

Autonomous vehicle self-localization based on abstract map and multi-channel LiDAR in urban area

Accurate vehicle self-localization is significant for autonomous driving. The localization techniques based on Global Navigation Satellite System (GNSS) cannot achieve the required accuracy in urban canyons. On the other hand, simultaneous localization and mapping (SLAM) methods suffer from the error accumulation problem. State-of-the-art localization approaches adopt 3D Light Detection and Ranging (Lidar) to observe the surrounding environment and match the observation with a priori known 3D point cloud map for estimating the position of the vehicle within the map. However, storing the massive point cloud needs immense storage on the vehicle, or it should be stored on servers, which makes the simultaneous downloading of the map by multiple vehicles another challenge. In this study, rather than employing the point cloud directly as the prior map, we focus on the abstract map of buildings, which is easy to extract, and at the same time apparently observable by Lidar. More especially, we proposed vehicle localization methods based on two different abstract map formats representing urban areas. The first format is the multilayer 2D vector map of building footprints, which represents the building boundaries using vectors (lines). The second format is the planar surface map of buildings and ground. These map formats share the same idea that the uncertainty (deviation) of each vector or planar surface is calculated and included in the map. Later in the localization phase, the observed data from Lidar is matched with the abstract map to obtain the precise location of the vehicle. Experiments conducted in a dense urban area of Tokyo show that even though we significantly shrank the map size, we could preserve the mean error of the localization.     

基于驾驶行为的城市道路车辆主观换道模型研究

车辆换道行为因其运行环境复杂,所涉及的交通因素众多,容易引起交通冲突,从而降低道路交通系统的安全性.对车辆换道行为的动态特性及其对车流运行的影响机理进行建模与研究,对提高交通系统的运行效率有重要意义.基于城市道路车辆换道行为的特征,改进了元胞自动机模型细化车辆换道过程;考虑驾驶员特性、车辆类型的影响,采用模糊推理理论描述驾驶员的换道决策,进而建立了城市道路驾驶员主观换道模型.通过将实测交通流数据与仿真输出数据进行对比,验证模型的有效性.结果表明,所建立的模型输出结果与实测数据的误差较小,说明模型具有一定的有效性.     

面向智能网联汽车定位的协同地图匹配算法

为实现智能网联环境下低成本、高精度的车辆定位, 研究了基于自适应遗传Rao-Blackwellized粒子滤波的协同地图匹配算法。利用联网车辆的定位信息和道路约束条件消除公共偏差, 提高车辆定位精度。将自适应遗传算法引入到粒子滤波的重采样过程中, 增加粒子的多样性, 解决传统粒子滤波算法中容易出现的“粒子退化”和“粒子耗尽”问题。通过仿真实验与传统粒子滤波以及卡尔曼平滑粒子滤波下的定位结果进行了对比, 同时分析了不同联网车辆数目对定位精度的影响。通过实际测试验证了算法在实际应用中的定位效果。实测结果表明: 以典型十字路口为例, 在联网车辆数目为4的情况下, 协同地图匹配算法的定位误差范围为1.67 m, 分别为原始GNSS定位以及单车地图匹配定位结果的41.03%和56.80%。同时, 该算法的统计定位精度(CEP)达到1.06 m, 比GNSS原始定位精度提高了2.52 m, 具有较好的定位效果。      

Probe vehicle lane identification for queue length estimation at intersections

Vehicles instrumented with Global Positioning Systems, also known as GPS probe vehicles, have become increasingly popular for collecting traffic flow data. Previous studies have explored the probe vehicle data for estimating speeds and travel time; however, there is very limited research on predicting queue dynamics from such data. In this research, a methodology was developed for identifying the lane position of the GPS-instrumented vehicles when they are standing in the queue at signalized intersections with multiple lanes, particularly in the case of unequal queue. Various supervised and unsupervised clustering methods were tested on data generated from a microsimulation model. Among the tested methods, the Optimal Bayes Rule that utilizes probability density functions estimated using bivariate statistical mixture models was found to be effective in identifying the lanes. The methodology for lane identification was tested for queue length estimation. This research confirms that the lane identification is an important step required prior to the queue length estimation. The accuracies of the models for lane identification and queue length estimation were evaluated at varying levels of demand and probe vehicle market penetrations. In general, as the market penetration increases, the accuracy improves as expected. The result shows that 40% market penetration rate is adequate to reach about 90% accuracy.     

Integration of GPS and monocular vision for land vehicle navigation in urban area

For land vehicle navigation in urban area, Global Positioning System (GPS) receivers often suffer from the lack of positioning accuracy, availability, and continuity due to insufficient number of visible satellites and multipath errors. To mitigate this problem, this paper proposes an efficient hybrid positioning method combining a single frequency GPS receiver and a monocular vision sensor. The proposed method is advantageous in that it requires only low-cost hardware and no external map aiding. Compared with existing vision-based methods, the proposed method directly measures absolute heading angle based on the images of straight road segments. For the reason, the proposed method is resilient to multipath errors. The performance of the proposed method is evaluated by the experiments with field-collected real measurements; one with good satellite visibility and the others with poor satellite visibility. Comparison with existing positioning methods demonstrates the feasibility of the proposed method in urban area.     

1种基于车辆时空图的车辆异常行为检测方法

对快速路车辆异常行为的检测有助于防止或及时处理交通事故,缓解交通拥堵,保障出行的安全和效率。采用多高斯背景模型提取前景运动车辆及其中心点并利用Kalman滤波算法跟踪运动车辆。在此基础上,得到各个车道上车辆的行驶时空图,通过车辆时空图对车辆行为进行轨迹分析,根据时间序列上车辆位置的变化检测车辆逆行,通过车间距和车辆位置状态信息检测车辆碰撞。实验表明,该方法能较好地识别出车辆异常行为。     

In-depth understanding of lane changing interactions for in-vehicle driving assistance systems

Lane-changing events are often related with safety concern and traffic operational efficiency due to complex interactions with neighboring vehicles. In particular, lane changes in stop-and-go traffic conditions are of keen interest because these events lead to higher risk of crash occurrence caused by more frequent and abrupt vehicle acceleration and deceleration. From these perspectives, in-depth understanding of lane changes would be of keen interest in developing in-vehicle driving assistance systems. The purpose of this study is to analyze vehicle interactions using vehicle trajectories and to identify factors affecting lane changes with stop-and-go traffic conditions. This study used vehicle trajectory data obtained from a segment of the US-101 freeway in Southern California, as a part of the Next Generation Simulation (NGSIM) project. Vehicle trajectories were divided into two groups; with stop-and-go and without stop-and-go traffic conditions. Binary logistic regression (BLR), a well-known technique for dealing with the binary choice condition, was adopted to establish lane-changing decision models. Regarding lane changes without stop-and-go traffic conditions, it was identified based on the odd ratio investigation that he subject vehicle driver is more likely to pay attention to the movement of vehicles ahead, regardless of vehicle positions such as current and target lanes. On the other hand, the subject vehicle driver in stop-and-go traffic conditions is more likely to be affected by vehicles traveling on the target lane when deciding lane changes. The two BLR models are adequate for lane-changing decisions in normal and stop-and-go traffic conditions with about 80 % accuracy. A possible reason for this finding is that the subject vehicle driver has a tendency to pay greater attention to avoiding sideswipe or rear-end collision with vehicles on the target lane. These findings are expected to be used for better understanding of driver’s lane changing behavior associated with congested stop-and-go traffic conditions, and give valuable insights in developing algorithms to process sensor data in designing safer lateral maneuvering assistance systems, which include, for example, blind spot detection systems (BSDS) and lane keeping assistance systems (LKAS).     

基于车轴图谱的城市道路汽车运行轨迹特性分析

为探究城市道路行车轨迹与路侧之间的横向距离对车辆运行的影响,提高驾驶员行车安全,在某市滨海路进行汽车运行轨迹样本采集试验,使用AxleLight RLU11系列路侧交通数据采集系统分车道采集试验路段汽车运行轨迹样本,利用SPSS Statistics对试验路段不同车道车辆运行轨迹样本进行数据处理,绘制不同行车道运行车辆横向距离的累积频率曲线,计算得到汽车运行轨迹与路侧的横向距离D85,通过绘制行驶车辆距路侧的横向距离直方图,得到不同车道的车辆分布规律。结果显示,驾驶员大多数偏向选择在内侧车道运行。根据试验路段内外2条车道车辆横向距离和运行轨迹特性,可为城市道路交通安全设施的设置提供理论依据,以期提高城市道路交通运行安全。     

自动驾驶汽车基础测试场景构建研究（双语出版）

针对L2/L3级自动驾驶汽车的仿真测试和封闭场地测试认证需求，结合现有L2/L3级自动驾驶汽车量产车型的主要功能特点，提出自动驾驶汽车基础测试场景群的构建方法。首先针对指定的道路交通环境，分析主车和周围交通参与者可能的相对位置和运动方向的组合，确定复杂场景群。其次分别以主车功能所确定的各个可能运动方向，依此与各干扰车辆的可能运动方向（包括任一干扰车辆不存在的情形）进行组合，组合时采用PICT组合测试工具，并添加必要的运动约束条件，选择参数组合覆盖标准自动生成全部的组合场景群。最后结合场景筛选规则，筛选出具有测试价值的覆盖各个层级及功能的基础测试场景群。采用场景构建方法，对于主车处于三车道中间车道的路段场景和无红绿灯的十字路口场景，分别构建62种和33种基础测试场景。根据驾驶人行为特性、交通规则、汽车在城市、郊区和高速公路工况下的典型车速、加减速度、横向加速度、交通事故和自然驾驶数据库的有关场景数据等，设计主车换道工况的测试用例。采用模型预测控制框架建立主车局部路径规划和控制仿真模型，并对主车危险换道场景进行仿真。研究结果表明：主车在邻车道前车大减速的情况下实现了减速换道并避免了与本车道前车和邻车道前后车的碰撞，同时跟踪到期望跟车间距，验证了该换道测试用例的有效性。     

城市道路机动车道宽度研究

在分析国内机动车道宽度取值标准和国内交通流特性的基础上,采用波良可夫模型计算不同等级道路车辆的侧向摆动距离,结合车道上最具代表性的机动车辆车身宽度,计算得出不同等级道路、不同位置的推荐机动车道宽度。结果表明：适当缩减机动车道宽度可在满足行车安全的同时节约城市道路建设用地,提高慢行交通运行的通达性和安全性,推荐的车道宽度为城市道路建设中机动车道的合理选区提供了参考依据。     

交通宁静化理念下公交站点处两轮车专用道一体化设计

两轮自行车、电动自行车和摩托车专用道一体化的改造方案是为解决城市中心摩托车和非机动车争夺路权带来的交通问题的一种探索和尝试。专用道的设置不可避免也要对公交车站进行改造，交通比较复杂。为有序引导站点处的交通流，保障乘客安全，首先介绍了路段两轮自行车、电动自行车和摩托车专用道一体化方案的改造原因和思路，然后结合城市交通宁静化理念，对直线式停靠站和港湾式停靠站处的专用道提出了几种设计方案，并分析其优缺点和用地条件。经过对比，推荐公交站点处采用曲线式两轮车专用道设计方案，最后还配套了一些交通宁静化措施。     

Intersection traffic energy management system (ITEMS): formulation and validation

Urban transportation and traffic face significant challenges as cities continue to grow. Traffic intersections present unique challenges for city planners. Several studies have proposed diverse approaches to managing intersection traffic. However, these typically fail to reduce both energy and trip time for all vehicles involved. This paper proposes the intersection traffic energy management system (ITEMS). By using speed regulation of two vehicles approaching an intersection on crossroads, ITEMS is able to resolve intersection conflicts. This obviates the need for either vehicle to stop. Two-vehicle and four-vehicle scenarios were simulated. It was found that ITEMS reduced the total energy about 22% and the trip time across the intersection by almost 16% for every scenario compared to a traditional fixed-timed traffic light intersection. ITEMS was also found to be superior than always giving a green light to urban transit vehicles. Furthermore, with ITEMS, the energy and trip time of each vehicle in the scenario was reduced when compared to other strategies.     

城市车路协同系统下实时交通状态评价方法

交通状态评价方法能够为交通管理系统提供可量化的实时路网信息,为动态引导交通流、缓解交通拥堵提供依据。受限于交通路网的时变性和评价过程的主观性,目前传统评价方法的精度时常无法满足需求。基于城市车路协同系统动态获取路网信息优势,提出一种利用车路信息融合的实时交通状态评价方法。首先,定义了一种网联汽车与路侧终端间的无线交互方式,并确定数据协议以保证实时车辆数据的准确性;其次,从实时数据中选取平均通过时间、平均停车次数、平均停怠时间作为一级评价指标进行模糊综合,应用多算子对计算的一级评价结果构成二级交通状态评价指标,并根据层次分析法确立指标权重,同时根据仿真和试验结果建立适用于各级道路参数的可变隶属度规则,从而融合动态车辆数据与静态路段参数,计算得出交通状态评价结果与评分;最后,由网联汽车、车载终端、路侧终端和无线通信模块搭建实际协同测试系统对该方法进行了试验验证。试验结果表明：测试系统所得到的路段实时交通状态评价得分与对应的交通状态变化趋势一致,能够准确体现城市车路协同环境下的交通状态特点。该评价方法运用信息融合方法提高了交通状态评价结果的实时性与客观性,同时为车路协同技术应用于实时交通诱导,缓解城市交通拥堵提供了理论依据。     

存在车辆干扰的车道线识别

为避免道路上行驶的其它车辆对车道线识别的干扰,提出了一种结合车辆识别的车道线识别方法。融合雷达数据,车辆识别模块首先在图像中识别出车辆占据的区域;对于每一个车道线识别模块挑出的车道线候选点进行判断,去除处于车辆区域的车道线点;如果有效车道线点数目不足,则利用卡尔曼滤波的跟踪结果,确定符合最小风险函数的车道线位置。经过多种工况下的试验验证,该方法能够稳定地对车道线进行识别,准确地提取车道线参数,并且算法对车辆干扰有良好的抵抗能力。     

基于车联网的汽车自动换道方法研究

为弥补现有汽车自动换道时未考虑周围车辆运动状态变化以及舒适性差和通行效率低等方面的不足,同时提出了在车联网条件下的汽车自动换道方法,主要包括动态轨迹的规划、前馈与反馈相结合的PQ跟踪控制策略两部分,开展Carsim和simulink联合仿真以及实车验证结果表明,与传统非动态自动换道方法相比,该方法能有效解决在换道过程中周围车辆车速变化及车辆突然闯入等情况的难题,明显提高了换道过程中的安全性,由数据分析可知在保证车辆舒适性、稳定性的同时,换道时间缩短了20%,有效提高了车辆换道效率。     

基于注意力Seq2Seq网络的高速公路交织区车辆变道轨迹预测

针对交通状态复杂的高速公路交织区域,经验丰富的驾驶人能够通过正确地推断周围车辆的未来运动进行及时的车道变换,这对于实现安全高效的自动驾驶至关重要,然而目前的自动驾驶车辆往往缺乏这种预测能力。为此,基于深度学习理论,提出了一种结合注意力机制和编-解码器结构的交织区车辆强制性变道轨迹预测方法,利用Next Generation Simulation（NGSIM）数据集提取车辆变道过程中的关键特征,并引入碰撞时间（Time to Collision,TTC）和避免碰撞减速度（Deceleration Rate to Avoid a Crash,DRAC）2种风险指标,将变道车辆及其周围车辆视为一个整体状态单元,同时补全状态单元内部不同车辆在横向和纵向上的时空状态特征,从而更有效地刻画车辆间的动态交互行为;然后将不同观测车辆的连续窗口序列输入基于长短期记忆网络（Long Short-term Memory,LSTM）的编-解码器,预测交织区车辆变道的未来运动轨迹,通过添加软注意力模块,使模型能够集中聚焦于影响车辆在不同时刻下位置变化的关键信息,再现了真实交通场景下车辆的变道行为。试验验证表明：基于注意力机制的编-解码器模型与当前流行的卷积长短期记忆网络、极限梯度提升树等模型相比具有更高的轨迹预测精度,在长时域的变道轨迹拟合上有显著的优越性,为辅助和自动驾驶领域的发展提供了新思路。