基于BP神经网络的道路交通事故预测

道路交通事故预测是交通研究的一个重要课题.以我国交通安全状况为研究对象,依据我国道路交通事故的特点,利用神经网络具有自学习、自组织、自适应能力特性,运用神经网络的方法及我国多个年度道路交通事故统计数据,建立了道路交通事故神经网络宏观预测模型,预测精度符合道路交通事故预测的要求.     

基于新老司机道路选择行为的交通流均衡研究

针对新上路司机数量的爆发式增长带来的交通流均衡变化问题,本文在对司机进行分类和道路选择行为分析的基础上,应用决策树方法构建了基于新老司机道路选择行为的交通流均衡模型,并用实际案例研究了主要参数的敏感性。研究发现:路况通过联合系统车流量的分配情况、车速预期和司机学习过程等因素共同对司机比例产生影响;当某一条道路达到均衡时,车流量对司机比例的敏感程度受到路况差异的影响;司机在驾驶经验、对道路的熟悉程度以及学习能力等方面存在差异,但这种差异仅存在于对交通信息的感知方面。该研究有助于加深对城市交通流变化规律的认识和理解,对于发展出有效的交通管制措施有积极的理论价值和现实意义。     

基于实际路网情境的配送车辆调度优化

在实际路网情境下结合车道数、车道宽度、路口信号灯设置等路网物理特性,构建了考虑综合交通阻抗的多车型车辆调度模型,提出了两阶段求解策略:第1阶段设计了改进A-star精确解算法用于计算客户时间距离矩阵;第2阶段针对实际路网的特征设计了混合模拟退火算法求解调度方案。以大连市某配送中心运营实例进行路网情境仿真试验,结果表明:改进A-star算法较改进Dijkstra算法具有更短的路径搜索时间;混合模拟退火算法求解结果较实际调度方案优化了13.1% 的综合成本;路网增流、区域拥堵和路段禁行三类路网情境均能对配送方案的车辆配置、路径选择、客户服务次序、作业时间和违约费用等5方面内容产生干扰,调度计划的制定需要详细考虑这些因素的变化。     

基于深度学习LSTM对交通流状态的预测

当今道路交通状态对城市管理和人们出行愈加重要,影响着人类生活的方方面面.以深圳交通为研究对象,由基础车辆数据和道路坐标构建了路网系统,从车辆速度和密度两个方面导出了交通流状态评价指数TSI.利用深度学习长短期记忆神经网络(LSTM)对车辆速度和密度两个指标进行预测,并通过对比极限学习机(ELM),时间序列(ARMA)和BP神经网络,进行仿真实验,结果表明相对于传统预测模型,所采用的LSTM网络具有更优的预测精确度和对远期预测的稳定性.最后利用预测结果计算出更能直观反映出道路交通拥堵情况的TSI指数,为人们提供了准确的交通状态预测.     

不同行为风格下路段行人与机动车行为博弈分析

针对路段过街行人与机动车的博弈决策行为,考虑不同交通行为风格下行人的过街特征和驾驶人的驾驶行为习惯等影响因素,开展路段行人交通行为风格调查,掌握不同交通行为风格下路段行人的过街特征,构建非合作动态人车博弈模型.通过分析不同策略下行人与驾驶人的收益,求解纳什均衡,得到不同策略下行人与驾驶人的最优策略.为解决路段行人与机动车冲突提供新的思路.     

Scheduling of road vehicles in sugarcane transport: A case study at an Australian sugar mill

Pressure to remain internationally competitive has forced Australian sugar mills to reduce capital and operational costs. Improved scheduling of road transport vehicles provides one such opportunity, as it would reduce vehicle queue and mill idle times and hence the number of vehicles needed. It is difficult for mill traffic officers to produce good transport schedules manually due to the need to service a large number of harvesters in different locations. To address this issue, research was undertaken participatively with a sugar milling company in Australia to produce and implement a mixed integer programming model that represents the road transport operations. Two meta-heuristics were applied to find a solution to the model, leading to potential cost savings of AU$240,000 per year versus schedules produced manually by the mill traffic officer. The model was also applied to explore regional planning options for a more integrated harvesting and transport system.     

Stability of vehicles under nonstationary crosswind excitation

Strong crosswind gusts have great influence on the stability of railway and road vehicles. They may lead to accidents and also to a discomfort for the road vehicle driver. Risk assessment for overturning of railway and road vehicles is usually calculated based on the stationary situation or at least on wind-tunnel experiments that are mostly carried out with a static vehicle model. Nonstationary excitation due to wind turbulence occurs if the vehicle accelerates or decelerates. Increasing vehicle speed relative to wind speed will move the energy content of the spectrum to a higher frequency range. It has been realized that nonstationary wind has a great influence on vehicle stability especially when the vehicle speed is high. Thus in order to assess the overturning risk in a more realistic way, a nonstationary wind model together with its interaction with the vehicle should be taken into consideration.This paper proposes a nonstationary wind turbulence model for the investigation of crosswind stability of ground vehicles. A wind model with nonstationary turbulence as well as the wind effects to the moving vehicle in a nonstationary situation (acceleration/deceleration) is described. Nonstationary aerodynamic forces are considered together with the interaction between the moving vehicle system and the wind turbulence. Failure probabilities are computed and reliability analyses are carried out. (© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Car path tracking in traffic flow networks with bounded buffers at junctions

This article deals with the modeling for an individual car path through a road network, where the dynamics is driven by a coupled system of ordinary and partial differential equations. The network is characterized by bounded buffers at junctions that allow for the interpretation of roundabouts or on-ramps while the traffic dynamics is based on first-order macroscopic equations of Lighthill-Whitham-Richards (LWR) type. Trajectories for single drivers are then influenced by the surrounding traffic and can be tracked by appropriate numerical algorithms. The computational experiments show how the modeling framework can be used as navigation device.     

Dynamic system optimal performances of shared autonomous and human vehicle system for heterogeneous travellers

ABSTRACT

Autonomous vehicles (AV) can solve vehicle relocation problems faced by traditional one-way vehicle-sharing systems. This paper explores the deterministic time-dependent system optimum of mixed shared AVs (SAV) and human vehicles (SHV) system to provide the benchmark for the situation of mixed vehicle flows. In such a system, the system planner determines vehicle-traveller assignment and optimal vehicle routing in transportation networks to serve predetermined travel demand of heterogeneous travellers. Due to large number of vehicles involved, travel time is considered endogenous with congestion. Using link transmission model (LTM) as a traffic flow model, the deterministic time-dependent system optimum is formulated as linear programming (LP) model to minimize the comprehensive cost including travellers’ travel time cost, waiting time cost and empty vehicle repositioning time cost. Numerical examples are conducted to show system performances and model effectiveness.     

道路维修施工路段交通分流信控优化模型

针对道路维修施工期间常采用的部分路面封闭施工且利用辅路进行分流的情形,探讨了交通分流信控优化模型。借助交通流波动理论,分析了施工路段及其前后车流拥挤排队及疏散特征和规律,分析了对车流进行控制需满足的约束,并分析了车流的延误计算公式。以总的车辆行驶时间最小化目标,原路径及分流路径的绿时分配及信号周期为优化参数,考虑交通分流控制的各种约束,建立了道路施工路段交通分流信控优化模型。分析了该模型属于非凸问题,因此提出了一种近似求解最优解的办法。通过一个示例对模型和求解算法进行了验证,并对一些规律性结果进行了分析。     

基于判别分析的骑推行交通事故鉴定研究

道路交通事故鉴定对于交通事故责任认定和法庭举证具有重要作用.采集某市近三年125起交通事故案例数据,建立机动车、自行车、人体总计73个调查变量,运用SPSS软件开展骑推行事故鉴定研究.通过相关性分析发现车座、机动车类型等6个变量与鉴定结论相关.开展线性判别分析研究,结果表明直接对相关变量进行判别分析可以快速获得最佳判别效果.当函数选入的自变量为车座时,交叉验证准确率最高可达72.8%,说明采用数据挖掘的方法来鉴别交通事故中的行为方式具有一定可行性.     

Location-allocation models for traffic police patrol vehicles on an interurban network

This research investigates the traffic police routine patrol vehicle (RPV) assignment problem on an interurban road network through a series of integer linear programs. The traffic police RPV’s main task, like other emergency services, is to handle calls-for-service. Emergency services allocation models are generally based on the shortest path algorithm however, the traffic police RPV also handles other roles, namely patrolling to create a presence that acts as a deterrence, and issuing tickets to offenders. The RPVs need to be located dynamically on both hazardous sections and on roads with heavy traffic in order to increase their presence and conspicuousness, in an attempt to prevent or reduce traffic offences, road accidents and traffic congestion. Due to the importance of the traffic patrol vehicle’s location with regard to their additional roles, allocation of the RPVs adheres to an exogenous, legal, time-to-arrival constraint. We develop location-allocation models and apply them to a case study of the road network in northern Israel. The results of the four models are compared to each other and in relation to the current chosen locations. The multiple formulations provide alternatives that jointly account for road safety and policing objectives which aid decision-makers in the selection of their preferred RPV assignments. The results of the models present a location-allocation configuration per RPV per shift with full call-for-service coverage whilst maximizing police presence and conspicuousness as a proxy for road safety.     

基于权的最小平方法的山区道路交通安全度模糊综合评价

针对山区道路交通特点,综合考虑影响山区道路交通安全度的人、车、路、环境等主要因素,运用层析分析法构建道路交通安全度评价指标体系,基于权的最小平方法建立模糊综合评价模型,并利用加权平均原则对其进行综合评价.以云南省罗富路"者桑"段为例,结果表明该模型可以有效的反映道路的安全状况,且权的最小平方法可以实现道路交通安全度主要影响因素权重系数的精确分配,合理反映各评价指标的重要性,为提高道路的交通安全性奠定基础.     

A simulation-based assessment of route guidance benefits under variable network congestion conditions

The purpose of this study is to establish a quantitative relationship between network congestion and travel-time reduction benefits of a real-time route guidance user service. The approach of the study is to employ the INTEGRATION traffic simulation model and a 2,000 link network based on the Detroit, Michigan roadway system in a series of experiments. While holding the capacity of the roadway fixed, the value of route guidance is evaluated over a range of increasing demand levels. Network demand patterns and trip characteristics are comparable to current national averages. Measures of congestion such as average system commute speed either match or exceed current national averages. Congestion metrics measured for the lightest demand scenario match most current empirical national average data, while the heaviest demand scenario appears roughly comparable with 1994 Tokyo conditions.Results from this study indicate that route-guided vehicles benefit regardless of level of congestion, however, the amount of trip time savings achieved is highly dependent on network congestion conditions. Average benefits for route guided vehicles over unguided vehicles in the A.M. peak period range between 8–26% depending on overall traffic volume.The results indicate a two-part linear relationship between route guidance benefit and network congestion. As congestion increases in the network, benefits of route guidance increase until average network speed drops below 20mph. Beyond that point, benefits decline (but remain positive). This 20mph threshold in our network is the point where the dynamically growing and shifting mass of queued vehicles around bottlenecks begins to impede access to alternative routes for guided vehicles network-wide.In a related experiment, route guided vehicles that receive reliable data on network conditions (including incidents or demand variation) gain 3–9% travel time savings over unguided vehicles that follow optimal routes based on average time-variant network congestion conditions. Route guided vehicles may exploit information about unexpected delays in the network related to incidents as well as variability in daily traffic patterns. This experiment was conducted to isolate the value of route guidance with respect to experienced commuter traffic, rather than an aggregated model of driver behavior including both familiar and unfamiliar drivers.The preliminary results of this study have implications for ITS benefit assessment. First, the benefits of route guidance are directly related to the level of recurrent congestion in a network. Thus, a near-term poor market for route guidance may evolve over time into a good market for these services. Likewise, a good market for a route guidance user service may deteriorate if overall network congestion reaches very high levels. Second, a route guidance user service provides benefits compared to both a model of aggregated unguided traveler behavior and a model of experienced commuter behavior, regardless of congestion levels. Third, route guided vehicles are demonstrated to gain benefit by avoiding the worst congestion in the network. This minimization in day-to-day variability in commute time may be the most significant benefit of the route guidance system for the familiar driver.     

Model of road traffic accidents based on root primary causes

An offshoot of an existing general model of road traffic accidents (MRTA) is presented. The general MRTA contains an unwieldy number of terms for practical purposes. A reduction in the number of terms becomes necessary to render the model suitable for numerical computations. The method used to achieve this through grouping primary causes is shown. Some potential applications of the model are also outlined. Computations performed using this model yielded results which compared favourably with available results from other practical sources.     

A model of the relay valve used in an air brake system

The mathematical model governing the response of the relay valve in an air brake leads to a hybrid system in which different governing equations apply to different phases of the response of the air brake. To accurately describe the brake’s response characteristics it is imperative to take into account this hybrid structure, and it is to this aspect of the problem that this paper is addressed. The safe operation of any vehicle on the road depends, amongst other things, on a properly operating brake system. Most commercial vehicles such as trucks, tractor–trailers, buses, etc., are equipped with an air brake system. Any defect in a brake system can degrade its performance seriously and can lead to accidents. It is desirable and also important to develop systems that can control and diagnose air brake systems in order to both sustain and improve their performance. One approach to develop such systems is by obtaining a model of the air brake system and then using the same in the design process. The air brake system currently used in commercial vehicles can be broadly divided into a pneumatic subsystem and a mechanical subsystem. One of the main components in the pneumatic subsystem is the relay valve which operates the brakes on the rear axles of a tractor and the axles of a trailer. A relay valve has different modes of operation and the pressure response of the relay valve can be naturally described as the response of a hybrid system. In this article, we develop a hybrid dynamical model to predict the pressure response of the relay valve. An air brake testing facility has been set up at Texas A&M University and this model will be corroborated against experimental data obtained from the same.     

主路多车道无信号交叉口混合交通流的通行能力

Highway capacity is defined as maximum volume of traffic flow through the particular highway section under given traffic conditions,road conditions and so on.Highway construction and management is judged by capacity standard.The reasonable scale and time of highway construction,rational network structure and optimal management mode of highway network can be determined by analyzing the fitness between capacity and traffic volume.All over the world,highway capacity is studied to different extent in different country. Based on the gap acceptance theory,the mixed traffic flow composed of two representative vehicle types heavy and light vehicles is analyzed with probability theory.Capacity model of the minor mixed traffic flows crossing m major lanes,on which the traffic flows fix in with M3 distributed headway,on the unsignalized intersection is set up,and it is an extension of minor lane capacity theory for one vehicle-type and one major-lane traffic flow.     

Optimizing winter road maintenance operations under real-time information

This paper introduces a real-time optimization model that can be used by maintenance managers to develop and evaluate alternative resources allocation plans for winter road maintenance operations. The model takes into account a wide range of road and weather condition factors such as road network topology, road class, weather forecasts, and contractual service levels, and produces a vehicle dispatch schedule that is optimal with respect to operating costs and quality of service. The model is then used in an analysis on a realistic case to illustrate the potential impact of improved information on winter maintenance operations.