智能网联车辆混行交通流中灯语意图识别模型研究

为使混行交通流下智能网联车辆(Connected and Automated Vehicles, CAV)实现对人工驾驶车辆(Human-driven Vehicle, HV)前照灯灯语意图(Vehicle Headlights Intention, VHI) 的识别，弥补车对车(Vehicle to Vehicle, V2V)和鸣笛意图识别技术的不足，更好地与HV交互沟通，提出CAV对HV的VHI识别模型.模型包括：灯光感知、光数据处理、VHI识别3个模块，灯光感知模块通过RGB(Red-Green-Blue, RGB)和HSV(Hue-Saturation-Value, HSV)颜色空间感知前照灯(Vehicle Headlights, VH)，采用KLT(Kanade-Lucas-Tomasi Tracking，KLT)和车辆匹配算法定位跟踪发出灯语的HV；光数据处理模块采用光通道增益算法计算光辐射通量变化； VHI识别模块基于双层隐马尔可夫模型(Double-layer Hidden Markov Model，DHMM)辨识VH 闪烁次数和HV行驶状态，实现VHI识别.在3种灯语示意典型场景下的实验结果表明：1 s内 VH感知准确率为96.8%，定位跟踪精度小于1°，VHI识别率为96.6%，满足混行交通环境下 CAV对HV驾驶意图的识别要求，基本保证实时性，为混行交通流中CAV自动驾驶决策提供理论依据.     

The sequenced activity mobility simulator (SAMS): an integrated approach to modeling transportation, land use and air quality

The persistence of environmental problems in urban areas and the prospect of increasing congestion have precipitated a variety of new policies in the USA, with concomitant analytical and modeling requirements for transportation planning. This paper introduces the Sequenced Activity-Mobility Simulator (SAMS), a dynamic and integrated microsimulation forecasting system for transportation, land use and air quality, designed to overcome the deficiencies of conventional four-step travel demand forecasting systems. The proposed SAMS framework represents a departure from many of the conventional paradigms in travel demand forecasting. In particular, it aims at replicating the adaptative dynamics underlying transportation phenomena; explicitly incorporates the time-of-day dimension; represents human behavior based on the satisficing, as opposed to optimizing, principle; and endogenously forecasts socio-demographic, land use, vehicle fleet mix, and other variables that have traditionally been projected externally to be input into the forecasting process.     

抽动障碍儿童的5种行为问题分析

目的比较短暂性抽动障碍、慢性抽动障碍/Tourrettes综合征和正常对照儿童的5种行为问题。方法选择来我院儿童保健门诊就诊的169例短暂性抽动障碍儿童(病例1组)、44例慢性抽动障碍/Tourrettes综合征儿童(病例2组)和130例正常对照(对照组)儿童进行研究。用相关软件和量表对儿童智力、焦虑、注意缺陷多动障碍、睡眠障碍和自残行为进行测查;用自行设计的调查问卷收集儿童的一般情况、发病情况以及家庭成员患病情况。结果病例1组儿童与对照组儿童5种行为问题发生率比较,无统计学差别;病例2组儿童行为问题发生率高于其他两组儿童(P<0.05);病例1组和病例2组一级亲属抽动障碍的发生率分别为8.28%和9.09%(P>0.05),明显高于对照组的0.77%(P<0.05)。结论慢性抽动障碍/Tourrettes综合征儿童存在较多的行为问题,抽动障碍家族史是该病的危险因素。     

浅谈农村公路专项调查

根据交通部对农村公路建设新要求,为了合理分配农村公路建设资金,全面准确监测农村公路建设情况,交通部决定开展农村公路专项调查。详细阐述了丹东市农村公路专项调查对象和内容、工作开展的全过程及结果。     

基于PRO/E模型的潜器稳性计算

潜器稳性校核的基础是准确确定潜器的吃水线及浮心、重心位置.本文基于船舶稳性计算理论,结合潜器PRO/E三维模型,探索一种利用行为建模技术计算潜器稳性的方法.与传统计算方法相比,行为建模自动迭代计算方法的效率与准确性均有所提高,也为船舶其他CAD三维软件稳性计算提供参考.     

HEV制动意图识别的研究

基于大量的工况数据建立辨识模型,实现了对制动意图的准确识别.在此基础上优化了再生制动的控制策略.仿真结果表明,通过制动意图识别可有效地优化混合动力汽车再生制动的控制策略,从而进一步改善混合动力汽车制动时的驾驶感觉和燃油经济性.     

A study of motorcyclist's idling stop behavior at red lights

Abstract

Motorcycle activity in Asian economies is a significant contributor to carbon dioxide (CO₂) emissions, both when moving and when idling at traffic lights. This paper investigates Taiwanese motorcyclists’ behavior of turning off the idling engine while stopping at traffic lights based on the theory of planned behavior (TPB), which recognizes that the achievement of voluntary change behavior can be identified by knowing an individual's attitudes (or behavioral intentions [BIs]) in the context of social norms (SN). A structural equation model system is used to identify candidate causal links between attitudes, SN, BI and behavior related to the idling stop behavior of motorcyclists. A partial least squares (PLS) model is built to correct the covariance matrix, given the relatively small sample size. Results suggest that attitudes, SN and perceived behavioral control, are significant determinants of idling stop BI at red lights.     

基于大理市个体旅游交通行为分析

通过问卷调查的方式,获得大理市旅游交通的基础数据,采用统计分析的方法,得出系统客观的实证研究成果。在抽样调查的基础上,通过考察旅游者自身属性、出行信息以及与出游相关因子,运用量化的方法对所得结果进行分析,从而得出基于个体的旅游交通行为一般规律。     

基于制动意图的电动汽车复合制动系统制动力分配策略研究

文中对制动意图的识别和不同制动意图所要求的不同的制动力分配策略进行研究.通过试验,验证了制动意图识别算法和制动力分配策略的可行性和有效性,为复合制动系统的应用奠定了基础.     

识别驾驶人意图的车道偏离防避有界控制

为了防止车辆偏离车道导致交通事故的发生和避免车道偏离防避系统（Lane Departure Avoidance Systems,LDAS）对驾驶人行为不必要的干预,提出基于中心区操纵特性阈值法和基于D-S（Dempster-Shafer）证据理论的车辆偏离车道驾驶人意图识别准则,并运用CarSim/Simulink联合仿真对比2种识别准则的有效性。建立转向盘角速度为输入的车路模型,设计LDAS滑模转向控制器,基于预瞄点的侧向偏移量和横摆角速度设计LDAS的期望横摆角速度观测器,并与基于道路曲率和预瞄点侧向偏移量的期望横摆角速度的LDAS进行性能对比。运用相平面法确定保证LDAS车辆稳定性的前轮转向角最大值,并基于CarSim/LabVIEW RT硬件在环试验平台验证基于BP神经网络训练获得D-S证据理论的初始概率赋值的驾驶人意图决策算法的有效性。结果表明：所提出的识别准则能够及时识别车辆偏离车道时的驾驶人意图,为LDAS控制器介入赢得了宝贵的时间,所设计的期望横摆角速度观测器具有很好的稳定性,所提出的方法能够有效避免车辆偏离车道。