基于MapReduce的城市道路旅行时间实测计算

城市道路旅行时间计算一直是智能交通系统中研究的核心问题之一,准确高效的旅行时间计算可以有效地帮助道路管控,减少交通拥挤.然而面对巨大而且快速增长的城市道路交通检测数据,如何将分布式计算模式融合到传统的旅行时间计算问题中已成为一个亟待解决的问题.论文基于海量道路车牌识别数据,设计了基于MapReduce编程模型的城市道路旅行时间实测计算的算法.并利用Hadoop环境进行了实现,可以支持对自定义路段集下不同时间段道路旅行时间的计算.通过实验证明,相对于传统的旅行时间计算方式,在计算时间上基于MapReduce的旅行时间计算模式可以提高十倍以上.     

时空依赖的城市道路旅行时间预测

城市道路的旅行时间预测,对于路径规划以及交通管理至关重要.尽管旅行时间预测会受路段依赖、时空相关性以及其他因素的影响,但现有的方法并未考虑如何结合外部因素进行建模,因而可能会有引入错误信息、路段建模时忽略上下游路段间的依赖关系等问题,导致预测精度较差.鉴于此,提出了两阶段的旅行时间预测框架：首先,使用Skip-Gram模型对轨迹数据地图匹配后的路段序列进行编码,将其映射为低维向量,通过该编码方式避免引入错误信息的同时保留了路段间的上下游依赖信息.随后,基于路段编码模式整合天气、日期等外部因素,设计了基于深度神经网络的城市道路旅行时间预测模型.基于真实出租车轨迹数据集的对比实验结果表明,所提方法比对比算法具有更高的预测精度.     

基于道路摩擦系数估计的智能小车紧急避障策略研究

雾霾环境下驾驶员的视野受到限制,无法准确估计周围的环境信息,对行车安全具有重大影响.自主紧急制动(AEB)系统是一种重要的车辆主动安全功能,用来避免碰撞或减轻碰撞程度.通常,AEB系统利用一个碰撞时间TTC衡量与障碍物发生碰撞的危险程度.通常设计用于制动的TTC门槛值时假设道路摩擦系数为常数,然而,道路情况复杂多变,道路摩擦系数也是变化的,驾驶员在雾霾环境下更难准确估计道路摩擦系数.因此,开发了一个考虑不同摩擦系数对TTC门槛值影响的AEB控制策略.首先用一个复合滑移率轮胎模型来估计峰值道路摩擦系数,再用该系数计算TTC的门槛值,进而利用该Trc门槛值衡量与障碍物发生碰撞的危险程度.因为可以实时识别道路摩擦系数,提出的AEB策略可以自适应雾霾环境下不同的道路表面.仿真结果表明了该方法的有效性.     

隐马尔可夫模型的道路拥堵时间预测

道路车辆拥堵问题导致交通事故增加,降低了居民的出行效率,长时间的道路拥堵更是加重了环境污染,造成国家经济损失等诸多问题。为缓解城市道路交通的拥堵问题,提高出行效率,基于隐马尔可夫模型,针对已有道路拥堵时间数据进行采集与建模,并对该隐马尔可夫模型进行训练,通过算法计算与分析,预测未来一段时间的道路拥堵情况,为人们的出行提供拥堵时间预测,而后提出不同时段通过道路用时最短的最优路径。对韦尔奇算法进行改进,在原算法基础上增加考虑前n时刻状态。利用改进型韦尔奇算法,使得训练集参数更精确,达到预测精度更高的目的。实验结果表明,预测数据结果与真实数据相比,误差不超过3%,该模型预测结果具有较高准确性。     

基于信号控制的城市路网旅行时间计算模型

交通控制信号对交通流的影响是干扰实时交通数据计算准确性的重要因素。为此,提出一种基于信号控制的城市路网旅行时间计算模型。将城市道路的旅行时间分为2个部分,即路链有效旅行时间和路口延误时间,设计改进的信号控制延误模型用于计算路口延误时长,并给出路链合并算法。实验结果表明,该模型起点到终点的旅行时间误差率能降低5%~15%。     

基于道路风险评估的城市路网实时路径选择

为了缓解城市交通拥堵、避免交通事故的发生,城市路网的路径选择一直以来是一个热门的研究课题.随着边缘计算和车辆智能终端技术的发展,城市路网中的行驶车辆从自组织网络朝着车联网(Internet of vehicles,IoV)范式过渡,这使得车辆路径选择问题从基于静态历史交通数据的计算向实时交通信息计算转变.在城市路网路径选择问题上,众多学者的研究主要聚焦如何提高出行效率,减少出行时间等.然而这些研究并没有考虑所选路径是否存在风险等问题.基于以上问题,首次构造了一个基于边缘计算技术的道路风险实时评估模型(real-time road risk assessment model based on edge computing, R³A-EC),并提出基于该模型的城市路网实时路径选择方法(real-time route selection method based on risk assessment, R²S-RA). R3A-EC模型利用边缘计算技术的低延迟,高可靠性等特点对城市道路进行实时风险评估,并利用最小风险贝叶斯决策验证道路是否存在风险问...     

一种动态路段行程时间的预测模型

动态路段行程时问的预测是ITS动态最短路线选择的关键技术之一。根据对实际交通状况的分析,将路段行程时间分为三个部分,即自由行驶时间、排队等待时间和通过交叉 口时间。模型基于路段的基本信息及实时信息分别对这三部分时间进行预测,从而实现对整段路段行程时间的动态预测,精确度明显提高。     

基于计算实验的城市道路行程时间预测与建模

城市道路行程时间预测对于提高交通管控效果具有重要意义. 本文综合应用平行系统、集散波、误差反馈修正、多模型自适应控制及模型库动态优 化策略等方法与技术对间断流行程时间预测问题进行了研究. 首先,介绍了平行系统理论的基本原理及计算实验的基本方法; 然后,给出了基于平行系统理论的路段行程时间的预测模型, 设计了基于集散波的行程时间计算实验方法, 提出了多模型自适应行程时间预测并给出了模型动态优化策略. 最后,通过实验证明了本方法的有效性. 结果表明, 本文方法预测精度较高, 且能够对行程时间预测值进行持续优化, 可为后续的间断流行程时间预测研究提供借鉴.     

基于VR的驾驶仿真及安全教育系统开发

针对全国道路交通事故高发现状及传统驾驶安全教育方式单一、培训效果差的缺点,基于虚拟现实技术（VR）,在引发交通事故人为因素理论基础上,开发驾驶仿真及安全教育系统。系统基于Unity3D引擎,构建了基于道路实景数据的虚拟场景,并联合SUMO实现了道路交通流仿真,通过VR技术仿真驾驶环境及驾驶行为;基于碰撞检测原理,建立了关卡违规触发机制,编码自定义屏幕空间渲染方式模拟驾驶员视觉效果,并构建了基于图像的交通事故现场三维全景,从认知、感知层面培训驾驶员安全驾驶。实用性测试结果表明,系统实现了不同道路场景、气象条件与交通状况下的驾驶模拟及安全培训,增强了使用者的学习兴趣,提高了使用者驾驶安全素养,具有较强的实用性。     

Driver information system: a combination of augmented reality,deep learning and vehicular Ad-hoc networks

Improving traffic safety is one of the important goals of Intelligent Transportation Systems (ITS). In vehicle-based safety systems, it is more desirable to prevent an accident than to reduce severity of injuries. Critical traffic problems such as accidents and traffic congestion require the development of new transportation systems. Research in perceptual and human factors assessment is needed for relevant and correct display of this information for maximal road traffic safety as well as optimal driver comfort. One of the solutions to prevent accidents is to provide information on the surrounding environment of the driver. Augmented Reality Head-Up Display (AR-HUD) can facilitate a new form of dialogue between the vehicle and the driver; and enhance ITS by superimposing surrounding traffic information on the users view and keep drivers view on roads. In this paper, we propose a fast deep-learning-based object detection approaches for identifying and recognizing road obstacles types, as well as interpreting and predicting complex traffic situations. A single convolutional neural network predicts region of interest and class probabilities directly from full images in one evaluation. We also investigated potential costs and benefits of using dynamic conformal AR cues in improving driving safety. A new AR-HUD approach to create real-time interactive traffic animations was introduced in terms of types of obstacle, rules for placement and visibility, and projection of these on an in-vehicle display. The novelty of our approach is that both global and local context information are integrated into a unified framework to distinguish the ambiguous detection outcomes, enhance ITS by superimposing surrounding traffic information on the users view and keep drivers view on roads.     

Bi-level Decision-making Modeling for an Autonomous Driver Agent: Application in the Car-following Driving Behavior

ABSTRACT

Road crashes are present as an epidemic in road traffic and continue to grow up, where, according to World Health Organization; they cause more than 1.24 million deaths each year and 20 to 50 million non-fatal injuries, so they should represent by 2020 the third leading global cause of illness and injury. In this context, we are interested in this paper to the car-following driving behavior problem, since it alone accounts for almost 70% of road accidents, which they are caused by the incorrect judgment of the driver to keep a safe distance. Thus, we propose in this paper a decision-making model based on bi-level modeling, whose objective is to ensure the integration between road safety and the reducing travel time. To ensure this objective, we used the fuzzy logic approach to model the anticipation concept in order to extract more unobservable data from the road environment. Furthermore, we used the fuzzy logic approach in order to model the driver behaviors, in particular, the normative behaviors. The experimental results indicate that the decision to increase in velocity based on our model is ensured in the context of respecting the road safety.     

Efficient algorithm for automatic road sign recognition and its hardware implementation

The automatic detection of road signs is an application that alerts the vehicle’s driver of the presence of signals and invites him to react on time in the aim to avoid potential traffic accidents. This application can thus improve the road safety of persons and vehicles traveling in the road. Several techniques and algorithms allowing automatic detection of road signs are developed and implemented in software and do not allow embedded application. We propose in this work an efficient algorithm and its hardware implementation in an embedded system running in real time. In this paper we propose to implement the application of automatic recognition of road signs in real time by optimizing the techniques used in different phases of the recognition process. The system is implemented in a Virtex4 FPGA family which is connected to a camera mounted in the moving vehicle. The system can be integrated into the dashboard of the vehicle. The performance of the system shows a good compromise between speed and efficiency.     

出租车数据的城市道路网路段通行时间估计方法

城市路段通行时间估计能够更好地运营和管理城市交通。针对包含起点-终点位置,行程时间和距离信息的GPS行程数据,提出了一种城市道路网短时通行时间的估计模型。首先将城市道路网按照交叉路口分解为多个路段,并基于k-最短路径搜索方法分析司机行进路线。然后针对每一个路段,提出了双车道通行时间多项式关联关系模型,既能提升道路网通行时间精细度,又能避免因训练数据不足导致的路网通行时间过拟合问题。最后以最小化行程期望时间和实际行程时间之间的均方误差为优化目标,拟合道路网通行时间。在纽约出租车数据集上的实验结果表明,所提模型及方法相对于传统单车道估计方法能够更准确地估计城市道路网路段的通行时间。     

城市道路建设时序决策的鲁棒优化

为提高城市道路建设时序决策的鲁棒性,提出了城市道路建设时序决策优化的双 层规划模型。模型假定出行需求在一定范围内扰动,上层规划是在有限资金的约束下寻求各建设阶段的系统总出行时间与系统总出行时间对出行需求的灵敏度之间的综合最小值,下层规划为各建设阶段的随机用户均衡配流。文中推导出了系统总出行时间对出行需求灵敏度的计算式,并给出了模型的求解算法。最后以一个测试路网为例,对基于系统总出行时间、基于灵敏度、基于系统总出行时间与灵敏度综合出行时间的决策优化模型进行了计算分析,结果显示3种决策优化模型均可寻求到各自目标最优的城市道路建设时序,但在需求不确定的情景下基于灵敏度、基于系统总出行时间与灵敏度综合出行时间的决策优化结果更具鲁棒性。     

关于道路交通行车安全距离预测仿真

为了提高道路交通安全,针对行车安全距离的非线性带来的难以准确预测的问题,提出了一种临界行车安全距离的预测方法。以驾驶员驾驶风格类型、前车速度、后车速度、前车减速度为系统输入,以临界行车安全距离为系统的输出,应用最小二乘支持向量机(LS-SVM)建立预测模型。结合仿真软件采集到的样本数据进行训练,得到行车安全距离的预测结果,并与目前普遍采用的BP (Back Propagation)神经网络模型的预测结果进行了对比。实验结果表明,所提出的预测模型能准确地预测临界行车安全距离,且预测准确度明显优于BP神经网络。     

Research on the comprehensive traffic state evaluation model linked with drivers’ perception under the vehicle networking

Unified judgment standards and methods are often adopted to identify traffic state in certain road network based on traffic flow parameters. However, drivers often have different perceptions about the traffic state on different road sections, since their expectations on traffic state vary more or less from each other on different road sections. In particular, under the vehicle networking, out of considerations for safety and other relevant factors, requirements for the correlation and coordination of running vehicles have also raised significantly. Therefore, it is necessary to take driver’s perception about the driving conditions of certain road sections into consideration to adjust the release of traffic state. This paper has provided a comprehensive traffic state evaluation model linked with driver’s perception under the vehicle networking. The authors first establish an ANFIS model based on the T–S model and then conduct statistical analysis on drivers’ perceptions about certain traffic state. At last, the authors use the results of statistical analysis as regulatory factors to amend the parameters input through ANFIS. Through simulation, this paper has demonstrated that the model established has a high rate of convergence, a high identification precision and the generalization ability to conduct researches on the identification of traffic state.     

Cognition-based hierarchical en route planning for multi-agent traffic simulation

The goal of this study is to model drivers’ cognition-based en route planning behaviors in a large-scale road network via the Extended Belief-Desire-Intention (E-BDI) framework. E-BDI is a probabilistic behavior modeling framework based on agents’ own preferences of multiple attributes (e.g., travel time and its variance) and daily driving experiences. However, it is challenging to use the E-BDI framework for the demonstration of drivers’ en route planning behavior in a large-scale road network due to its high computational demand. To handle the computation issue, a hierarchical en route planning approach is proposed in this study. The proposed E-BDI-based en route planning approach consists of three major procedures: (1) network partitioning, (2) network aggregation, and (3) E-BDI-based en route planning. The Java-based E-BDI module integrated with DynusT^® traffic simulation software is developed to demonstrate the proposed en route planning approach in Phoenix, Arizona road network involving 11,546 nodes and 24,866 links. The demonstration results reveal that the proposed approach is computationally efficient and effective in representing various en route planning behaviors of drivers in a large-scale road network.     

基于波动方程的动态交通流网络模型研究

研究优化交通流量问题。从交通流量中获取实时旅行时间是现代智能交通系统模型的关键技术,针对传统模型构建复杂、计算时间长、难以提供实时旅行时间的缺点,构建出一种动态交通流网络分析模型,以计算实时性的旅行时间优化交通流量。在模型中首先使用LWR车流模型构建成起始值-边界条件连续方程式,采用高阶Runge-Kutta法计算路段上的流量、密度和运行速度,进而得到车辆运行某段距离所需要的旅行时间;再将这些旅行时间加总,则可求得全路段或路网的旅行时间。最后使用上面提出的的动态交通流网络模型对一小型的高速公路单车道交通流网络进行了仿真。仿真结果表明,上述模型可以加快动态交通流网络中旅行时间的求解速度,以达到提供实时信息的目标。     

Drivers' margins of safety considered as a focus for research on error

Accident statistics alone cannot provide a sound understanding of driver error, although they can assist the evaluation of remedial measures against errors and accidents. Roadside observation of drivers' errors can provide a valid index of their relative riskiness and of overall accident frequency, but only in route-specific applications. Field testing of hypotheses developed from theories of driver error is seen to be a far more valid and arguably more cost-effective method of improving road safety than relying on post hoc subjective assessments of error contributions to accident statistics. The distinction between driving task and envronmental factors which contribute to error production and those which constrain error correction is not well-documented in road accident studies. Yet it seems essential to make this distinction if we are to reach a sound understanding of research requirements in this field and hence identify and evaluate cost effective countermeasures against driver error. The bias which certain drivers appear to have towards inadequate safety margins is seen to provide an instructive theoretical framework for field studies of error production and error correction as contributory factors in traffic accident causation.     

Sharing Context-Aware Road and Safety Information

We present an integrated vehicular system for the collection, management, and provision of context-aware information on traffic and driver location. This system uses an integrated vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication paradigm enriched with an information management system. The infrastructure manages vehicle-detected safety hazards and other relevant information, adapting them to the vehicle's context and driver's preferences. This vehicular integrated system resembles the concept of a smart road.