信号交叉口行车风险场建立及车辆通行行为优化

随着汽车逐步向智能化、网联化发展,智能网联车辆逐步进入实际应用阶段。进行智能网联车辆的通行行为优化,对提升驾驶安全性和行车效率,避免事故发生和交通拥堵至关重要。车辆在通过交叉口时将受到很多环境及运动因素的影响,而现有的通行优化模型难以准确表达各类因素共同作用下的行驶环境。为此,基于风险场理论建立由环境场和运动场组成的信号交叉口行车风险场,表征信号交叉口中每点的实时行车风险程度,从而引导车辆驶向风险值低点,并提供下一步长的位移及速度指引,实现车辆的动态轨迹优化及速度控制。典型场景下的仿真结果表明：在优化模型的控制下单车的信号交叉口通行效率明显提升,其中直行方向车辆单车平均通行效率提升最高,平均提升6.35%,通过对交叉口面积内所有车辆进行通行行为优化,交叉口通行效率提升了9.3%,这表明所建模型可以准确表达交叉口行车环境并优化车辆通行行为。研究结论可应用于自动驾驶车辆的交叉口通行控制,并为网联环境下的行车环境表达和安全驾驶控制提供模型基础。     

面向驾驶模拟器的高等级公路三维视景建模与实现

随着汽车驾驶模拟器在道路交通仿真中的广泛应用,驾驶员对视景的逼真度也相应地提出了较高的要求。视景仿真系统作为模拟器的一个子系统,其逼真度和实时的动态显示是影响道路交通仿真的关键因素。文中介绍了高等级公路三维视景的建模过程、实现方法,以及地形的简化技术。设计成果表明,这种方法对于利用驾驶模拟器进行道路交通仿真等方面的研究是可行的。     

管道主要风险的失效频率数据分析

对其他国家地区管道运输发展中管道所发生的事故进行研究,尝试对一些事故报告数据进行分析,研究它们的特性.借鉴其他管道系统的事故原因,列出了管道类型初步分类应考虑的条件和面临的主要风险.认为对管道类别应该有区别地划分.     

船上实施听力保护的技术要求及应用

船舶噪声对船员的听力损伤是众所周知的,甚至发生部分船员的永久性听力损失。国际海事组织第91届海安会通过了新的《船上噪声等级规则》,围绕实施噪声规则,分析了听力保护的技术要求和正确选用听力保护器的技术依据,探讨了佩戴听力保护器对语言交流和感受听觉信号造成的影响。强调必须开展听力损伤的风险评估,并对识别听力损伤的特殊风险因素提出了建议。     

基于船舶领域和MMG的最晚施舵点/CRI研究

提出了一种基于MMG和船舶领域的LSP计算模型,以及改进的空间碰撞危险度（space collision risk index,SCRI）/时间碰撞危险度（time collision risk index,TCRI）模型,并进行了计算机仿真.结果表明：水动力模型精度可满足研究与实践要求;基于二分法的LSP模型能快速、可靠收敛;LSP和SCRI/TCRI模型更符合海员通常做法和避碰规则的公认理解.     

Examining the influence of firm performance on business risk-taking and the mediation effect of scale of operations in the container terminal industry

Container shipping and its related service sectors help accelerate globalization of the world economy. This industry has been experiencing rapid growth, prompting container terminal operators to increase their handling capacity in response. Providing container terminal services requires substantial capital investment in physical assets such as cargo handling facilities and information systems. On the other hand, operating container terminals is a long-term investment that typically spans several business cycles. Hence prudent asset management using appropriate tools is critical for container terminal operators to sustain their businesses. Generally, due to risk-adverseness, investors are unwilling to take more risk in their investment unless they can reap a higher return. Contrary to this argument, this study finds no direct influence of better firm performance as a proxy of higher return on business risk-taking by container terminal operators. Instead, scale of operations is positively associated with business risk-taking, suggesting that container terminal operators with a larger scale of operations are willing to take more business risk.     

A new composite decision support framework for strategic and sustainable transport appraisals

This paper concerns the development of a new decision support framework for the appraisal of transport infrastructure projects. In such appraisals there will often be a need for including both conventional transport impacts as well as criteria of a more strategic and/or sustainable character. The proposed framework is based on the use of cost-benefit analysis featuring feasibility risk assessment in combination with multi-criteria decision analysis and is supported by the concept of decision conferencing. The framework is applied for a transport related case study dealing with the complex decision problem of determining the most attractive alternative for a new fixed link between Denmark and Sweden – the so-called HH-connection. Applying the framework to the case study made it possible to address the decision problem from an economic, a strategic, and a sustainable point of view simultaneously. The outcome of the case study demonstrates the decision making framework as a valuable decision support system (DSS), and it is concluded that appraisals of transport projects can be effectively supported by the use of the DSS. Finally, perspectives of the future modelling work are given.     

基于满意度的多目标约束模糊控制规则库的建立及应用

本文针对难以建模的非线性系统,结合有关满意度优化理论,给出了基于模糊控制系统的多目标搜索规则库的方法,并利用所建立规则采用Mamdani算法来逼近控制问题的最优解。同时把该方法应用在具有强非线性特点的车辆自动驾驶问题上,通过对模糊控制规则的量化细分搜索出车辆自动驾驶的规则库。     

公路空间视距计算方法与检测技术

为了精确计算公路在三维空间中的实际视距,提出了空间视距的概念,通过延伸应用空间两点通视原理,建立了空间视距计算和分析的数学模型。基于计算机仿真技术开发出对空间视距的实时检测技术,通过将设计视距、运行视距和空间视距3种视距对比分析,给出了发现视距不足的方法。与早前的视距计算方法和研究相比,该方法对正常路面、行车道空间视距计算准确,能够计算和分析公路路侧附属设施对视距的影响,为公路几何设计和安全评价提供了依据。     

基于DSP的嵌入式驾驶疲劳监测系统研究

为了进行有效的、实时的、非接触式的驾驶员疲劳检测,采取NHTSA和FHWA推荐的非接触式疲劳检测方法PERCLOS,利用TI公司的专用图像处理DSP芯片TMS320DM642,构建了车载的基于DSP的嵌入式驾驶疲劳监测系统。首先使用肤色模型算法迅速定位人脸,然后利用累积差分帧和Hough变换等实时图像处理技术来检测、跟踪眼睛,分析眼睛的状态和提取眼睛的特征参数,计算PERCLOS值、判断疲劳程度并采取报警或其他措施。试验结果表明,系统的算法简单、快速、鲁棒性强。