基于HLA的智能交通信息试验平台研究

从智能交通信息平台的概念出发,研究利用已有的智能交通系统资源建立智能交通信息试验平台。首先对比分析了已有的平台体系结构开发方法,即面向过程的开发方法和面向对象的开发方法,两者都已不能适应复杂大系统的开发需求。通过引入计算机建模与仿真领域的HLA（高层体系结构）技术,创造性地提出了基于HLA的体系结构开发方法,结合信息试验平台需求分析,设计了智能交通信息试验平台的逻辑结构。平台分4层来组织,分别是基础数据层、数据库层、应用服务层和用户主体层,在若干接口程序的衔接下,各类联邦成员协同作用以实现信息平台的功能。结合现有的技术设备条件,实现了信息平台的物理连接。平台搭建后运营正常,已接入信息平台的ITS系统包括车载导航系统和城市商用车辆调度系统,并为其他子系统的接入预留了接口。     

智能车辆安全辅助驾驶技术研究近况

论述了安全辅助驾驶技术的研究现状、研究的必要性以及研究进展。安全辅助驾驶技术包括车道偏离预警与保持、前方车辆探测及安全车距保持、行人检测、驾驶员行为监测、车辆运动控制与通讯等。分析了各种传感器的优缺点及其在实际应用过程中存在的问题,基于单一传感器不能很好地解决安全辅助驾驶技术可靠性和环境适应能力的要求,应结合激光雷达技术解决图像模糊问题,利用红外传感器增强机器视觉识别的可靠性,未来的安全辅助驾驶技术应该采取多种传感器融合的技术,结合毫米波雷达和激光雷达系统具有深度测量精确的特点,将极大的推动汽车安全辅助驾驶系统的应用和推广。     

多尺度图像边缘检测技术在车辆识别中的应用

指出了图像边缘检测的定义及现有算法的不足,并对Canny边缘检测离散准则及该准则下的最优线性滤波器进行了论述。根据边缘检测离散准则,利用数值方法求出该准则下边缘检测的最优线性滤波器及对应的平滑算子,将其与Canny边缘检测技术相结合进行边缘检测,检测出不同尺度下多幅边缘图像,并根据小尺度下图像边缘细节信息丰富、边缘定位精度高,大尺度下图像边缘稳定、抗噪性好等特点,将检测到的多尺度边缘进行融合,得到抗噪性好的精确的单像素宽边缘;同时将该方法应用于汽车特征提取,检测到完整连续的汽车边缘图像,为车型识别提供条件。试验结果表明:该方法不仅能准确检测出图像边缘,而且能有效地抑制噪声,是边缘检测的实用方法。     

基于Q-learning和BP神经元网络的交叉口信号灯控制

解决单交叉口信号灯最优控制问题。提出了基于强化学习的信号灯控制系统结构,应用强化学习中Q学习,将信号灯最优控制问题转变成是否切换运行相位的决策问题,提出了采用BP神经元网络实现Q学习的信号灯控制系统。应用微观交通仿真软件PARAMICS进行仿真分析,结果表明该系统能够感知交通流变化,并能够自适应地调整信号灯切换策略,以达到最优的控制效果,该方法是可行的,与定时控制相比具有明显的优势。     

基于机器视觉的智能车辆前方道路边界及车道标识识别方法综述

为了全面了解国内外在基于机器视觉的智能车辆前方道路边界及车道标识识别领域的研究进展，文章介绍了近年来一些典型的基于机器视觉的道路边界及车道标识识别系统，对基于机器视觉的前方道路边界及车道标识识别方法进行了分类，对各大类方法中采用的不同技术进行了阐述，然后对基于机器视觉与其他传感器融合的识别方法进行了总结，最后就该领域的研究难点及发展趋势进行了简要论述。     

智能交通系统及其车辆自动控制技术

本文简要介绍了智能交通系统的概念、发展状况、基本内容和研究开发过程，进一步阐述了智能交通系统中的先进车辆控制技术。综述了汽车纵向控制、横向控制、自动驾驶及车列控制。最后，对我国如何发展智能交通系统和先进车辆自动控制技术进行了探讨。     

A study of the environmental impacts of intelligent automated vehicle control at intersections via V2V and V2I communications

This article presents a novel intersection traffic management system for automated vehicles and quantifies its impact on fuel consumption and greenhouse gas emissions of CO₂ relative to traditional traffic signal and roundabout intersection control. The developed intelligent traffic management (ITM) techniques, which are based on a spatiotemporal reservation scheme, ensure that vehicles proceed through the intersection without colliding with other vehicles while at the same time reducing the intersection delay and environmental impacts. Specifically, the spatiotemporal reservation scheme provides each vehicle a collision-free path that is decomposed into a speed profile along with navigational instructions. The integration of the developed microscopic traffic simulator with instantaneous emission model, provides improved assessments of the environmental impact of traffic control strategies at intersections. The simulator architecture integrates several ITM algorithms, vehicle sensors, V2V/V2I communications, and emission and fuel consumption models. Each vehicle is modeled by an agent and each agent provides information depending on the specific vehicle sensors. The ITM system is supported by V2V and V2I communications, allowing the exchange of information among vehicles and infrastructure. The data include the estimated vehicle position and speed. Compared with traditional traffic management techniques, the simulation results prove that the proposed ITM system reduces CO₂ emissions significantly. The research also shows that these reductions are more significant when the traffic flow increases.     

The impacts of changing flight demands and throughput performance on airport delays through the Great Recession

Several significant events between 2007 and 2009 impacted flight demands and the abilities of the three major New York area airports to handle demand. This paper assesses the results of applying a probabilistic simulation method – which isolates the individual contributions of changes in flight demand and changes in airport throughput performance to changes in flight delays – to diagnose how these different events may have caused operational changes at these airports, and in turn, how the results may be used to inform policies for appropriate countermeasures. The analysis revealed two key observations. Firstly, certain patterns in throughput performance shifts caused the most significant delays, and were more likely to have been caused by controller staffing issues rather than caps. Secondly, relatively constant average delays from one year to the next may result from significant demand drops accompanied by large throughput performance degradations at an airport. This suggests that not only operational limitations on capacity encourage airlines to reduce schedules, but that changed demands can also impact throughput performance. Overall, the analysis indicates that caps may not have provided their fully intended delay benefits. Although they successfully reduced overall flight demands at LGA and JFK, they also directly limited throughput performance at critical times, in turn limiting delay benefits. In addition, demands at the busiest times of the day appear to be relatively inelastic to these operational limitations, insofar as demand profiles at EWR and JFK remained “peaky” in 2008 and 2009. Also, the recession was largely responsible for reducing demands at the airports in 2009, but the delay benefits of this were dampened by a corresponding throughput performance degradation. Based on the above observations, a more direct demand management policy combined with policies that focus on maintaining high staffing capabilities at critical times of the day may be considered, to reduce the likelihood of major queue formation on days that do experience sustained demands. The results also suggest that a more flexible caps system, particularly during times of heavy queues, could be explored. Although airport practitioners have keen understandings of how their airports operate, without the support of quantitative analysis tools, it can be more difficult to argue the need for appropriate countermeasures. An analysis such as the one presented here can provide the detailed quantitative substantiation required to build cases for these targeted policy directives and infrastructure investments.     

Implementation of interactive transit information kiosks at New York City transit facilities: Analysis of user utilization and lessons learned

Transit Traveler Information Systems (TTIS) comprise a wide range of technologies that transit agencies use to provide reliable and timely transit-related information to customers. The touch-screen interactive information kiosk is an example of these emerging TTIS technologies. This paper examines the implementation of interactive touch-screen information kiosks, known as “On the Go!” Touch-Screen Travel Stations, at Metropolitan Transportation Authority-New York City Transit (MTA-NYCT) facilities in 2011. It analyzes data from passenger intercept surveys, from the kiosks’ built-in application usage logs and from field observations to understand actual passenger utilization of the kiosks and to assess the implications for transit agencies. The field observations also made it possible to obtain a profile of kiosk users, which sheds light on the concept of the “digital divide.” The findings, presented as lessons learned, can help agencies elsewhere develop guidelines and effective strategies for implementing similar interactive transit information systems.     

A new methodology for incident detection and characterization on surface streets

In this paper, a new methodology is presented for real-time detection and characterization of incidents on surface streets. The proposed automatic incident detection approach is capable of detecting incidents promptly as well as characterizing incidents in terms of time-varying lane-changing fractions and queue lengths in blocked lanes, lanes blocked due to incidents, and incident duration. The architecture of the proposed incident detection approach consists of three sequential procedures: (1) Symptom Identification for identification of incident symptoms, (2) Signal Processing for real-time prediction of incident-related lane traffic characteristics and (3) Pattern Recognition for incident recognition. Lane traffic counts and occupancy are the only two major types of input data, which can be readily collected from point detectors. The primary techniques utilized in this paper include: (1) a discrete-time, nonlinear, stochastic system with boundary constraints to predict real-time lane-changing fractions and queue lengths and (2) a pattern-recognition-based algorithm employing modified sequential probability ratio tests (MSPRT) to detect incidents. Off-line tests based on simulated as well as video-based real data were conducted to assess the performance of the proposed algorithm. The test results have indicated the feasibility of achieving real-time incident detection using the proposed methodology.