教练车用单片机控制副制动系统的设计

详细介绍教练车用单片机控制副制动系统的结构、工作原理、系统电路和程序。该系统结构简单，经济实用，提高了教练车的安全性。     

施工便道用漫水桥结构设计

为保证某漫水桥的正常通行,在考虑施工现场条件的基础上,设计了一种施工便道用漫水桥结构。介绍该漫水桥的结构设计,对其结构安全进行检算,并从桥位、桥面高程、桥型、桥台和行车道等方面总结设计思路。结果表明,该施工便道用漫水桥结构设计总体上安全可靠,将该漫水桥应用于现场施工,使用期间漫水桥结构稳定,效果良好。     

Fuzzy sliding mode control method for vertical motion of autonomous underwater gliders北大核心CSCD

［Objective］This paper proposes a fuzzy sliding mode controller based on T-S fuzzy logic for the vertical plane motion control of an autonomous underwater glider (AUG) with limited actuator capability. ［Methods］In the fuzzy sliding mode controller, the fuzzy switching rate is used to replace the switching rate in the fixed time controller to effectively suppress buffeting. The fuzzy switching rate is obtained by fitting the switching rate of the fixed time controller with T-S fuzzy rules. Based on the limited capabilities of AUG actuators, a saturation auxiliary system is designed to improve the actuator saturation effect. Finally, the performance of the system is verified by Lyapunov stability analysis and numerical simulation. ［Results］The results show that the AUG under the fuzzy sliding mode controller and the saturation auxiliary system can converge in finite time. The effectiveness of the fuzzy sliding mode controller and the saturation auxiliary system are verified by numerical simulation. ［Conclusions］By making comparisons with the fixed-time controller, it is verified that the two controllers have similar control performance, and the buffeting of the fuzzy sliding mode controller is lesser. © 2022 Journal of Clinical Hepatology. All rights reserved.     

基于分段学习模型的自动驾驶行为决策算法研究

在具有车道线的特定自动驾驶场景中,针对目前端到端的行为决策算法直接输入原始图像进行决策导致的网络模型迁移性差、预测精度欠佳、泛化能力不足等问题,提出一种基于分段学习模型的车辆自动驾驶行为决策算法。首先,基于GoogLeNet建立一种端到端的车道线检测网络模型,并引入车道中心线作为决策的重要线索提高算法的迁移能力,同时利用YOLOv3目标检测模型对本车道内前方最近障碍物进行位置检测;而后,经几何测量模型将两者检测结果转换成环境状态信息向量为决策做支撑;最后,构建基于长短期记忆(LSTM)网络的驾驶行为决策模型,根据编码的历史状态信息刻画出动态环境中车辆的运动模式,并结合当前时刻的状态推理得到驾驶行为参量。使用建立的真实驾驶场景数据集对模型分别进行训练、验证与测试,离线测试结果显示车道线检测模型的检测位置误差小于1.3%,车道内前方障碍物检测模型的检测精度达98%以上,驾驶行为决策网络模型表征预测优度的决定系数 大于0.7。为进一步验证算法的有效性,搭建了Simulink/PreScan联合仿真平台,多种工况下的仿真验证试验中多个评价指标均达到工程精度要求,实车测试的试验结果也表明该算法可实现复杂驾驶场景下平稳、准确无偏航的预测效果并满足实时性要求,且与传统端到端模式的算法相比,具有更好的迁移性和泛化能力。     

基于预期偏移距离的人机权值分配策略研究

针对智能车辆横向运动控制中驾驶员和辅助系统的控制权限冲突问题,本文中提出一种人机权值分配策略。采用车辆在预瞄点处的预期偏移距离(PDLC)衡量车道偏离危险度,预期偏移距离通过对预瞄偏差修正获取。权值分配函数设计时以PDLC为自变量,以保证驾驶员的权值为优先控制目标,以一定的横向运动控制精度为先决条件。在CarSim/Simulink联合仿真平台和CarSim/Labview RT硬件在环实验台上对提出的控制策略进行了实验验证和数据分析。结果表明,采用权值分配策略协调驾驶员和辅助系统的控制,可在有效跟踪理想道路中心线的前提下保证驾驶员的控制权值,降低其工作负荷以及纠正驾驶员的误操作行为。     

考虑车道荷载差异的多车道桥梁横向折减系数

为考虑车道荷载差异对多车道桥梁横向折减系数的影响，提出横向折减系数的多系数表达模式，并改进基于独立重复试验的相遇车道荷载方法以进行系数校核。首先，引入反映车道荷载差异性的车道修正系数和体现多车道荷载相遇概率的车道组合系数，形成横向折减系数的多系数表达模式。其次，改进目前常用的基于独立重复试验的相遇车道荷载方法，将车道交通流量及荷载分布的差异性涵盖进来。随后，参数化研究横向折减系数的关键影响参量，包括货车通过控制断面的平均时间、荷载重现期、车道交通量和车道荷载分布等。最后，基于某高速公路实测动态称重数据，给出采用该方法校核横向折减系数的详细过程，并与传统方法及JTG D60-2015规范方法所得结果进行对比。研究表明：所提出的改进相遇车道荷载方法，摈弃了传统方法中车重正态分布、车道荷载同分布、最大观测车重与均值3.5倍偏差关系等备受质疑的假设，是更一般性的解答。参数化分析表明：货车通过控制断面的平均时间和荷载重现期对结果影响很小，货车荷载模型及其在车道间差异性则影响很显著，说明对车道荷载分布规律进行准确建模的重要性。实测车道货车荷载数据统计发现：货车的交通量和荷载分布在车道之间具有明显的差异，传统方法和规范方法给出的横向折减系数均高估了实际情况，最大达19%，可能造成设计与管养资源的浪费；而基于该方法校核的横向折减系数，更能深入揭示各车道荷载的贡献规律，准确反映不同车道组合作用下的荷载横向折减规律，具有显著的工程应用价值。     

Identifying multiclass vehicles using global positioning system data

It has been previously evidenced that global positioning system (GPS) data can be used to distinguish passenger cars from delivery trucks. In this paper, a machine learning approach is proposed to use GPS data to identify multiclass vehicles, including passenger cars, single unit trucks, and multi-trailer trucks. The method is acceleration and deceleration-based since it considers the variations of acceleration and deceleration as the most effective features to classify vehicles. The overall classification result for the three vehicle classes is about 75%. The major challenge is to distinguish single unit trucks from multitrailer trucks due to their somewhat similar mobility patterns. The paper also explores the impacts of GPS sampling frequency on vehicle classification. It is found that the proposed multiclass vehicle classification can be reasonably conducted if the data are collected frequently enough (i.e., every five seconds or more frequently) to capture the major acceleration and deceleration processes. The proposed method can be considered as a low-cost and non-intrusive approach to collect vehicle class information and to potentially supplement the existing classification schemes in urban areas.     

Development and evaluation of bus lanes with intermittent and dynamic priority in connected vehicle environment

Exclusive bus lanes provide a very high level of priority for transit operations, especially for Bus Rapid Transit and Express service, but these lanes could be underutilized and be a source of extra capacity if they could be shared in an intelligent way. This article explores the benefits of providing intermittent priority, called bus lane with intermittent and dynamic priority, of these exclusive bus lanes. Intermittent and dynamic priority can be implemented by allowing vehicles to use the lane when Bus Rapid Transit or Express bus is not present. Drivers can be alerted when a bus is in the lane using either infrastructure-based signs, or in the future using infrastructure-to-vehicle, or connected vehicle communications. Some critical operating parameters for implementing bus lane with intermittent and dynamic priority system including clear distance, degree of saturation (volume-to-capacity ratio), connected vehicle penetration, and bus departure/headway frequency have been investigated in this paper.     

Performance evaluation of the inside intersection median-turn lane markings on the mobility and safety performance of signalized intersections in the Philippines and Japan

Signalized intersections are one of the key elements that play a vital role at road networks. The efficiency and safety levels of intersections can affect the operational performance of the whole system. In general, turning traffic, especially median-turning, has always been considered as the most problematic movement in the operation of intersections. This becomes more critical with high turning demand where exclusive turning lanes (single or double) can be assigned to provide larger capacities for these movements and to reduce conflicts with through traffic. However, improper treatment of median-turn lanes could create cross-maneuvering behavior which may limit the expected increase in capacity and create safety issues. Median-turning lane markings are commonly provided at intersections in Japan to guide drivers while turning which is expected to reduce the conflicts among turning traffic. Meanwhile, in the Philippines, exclusive median-turn lanes are installed at intersections without proper treatment which may contribute to the low mobility and safety levels. Therefore, this study evaluated the impact of inside intersection lane markings on the operation of median-turn lanes in terms of mobility and safety. The vehicle maneuver, speed and interactions between the turning traffic were utilized as essential components for the assessment. The empirical analysis shows that conflicting trajectories were present on double turn lanes without median-turn lane markings in the Philippines, which resulted to serious conflicts among the turning vehicles and negatively influenced the turning speed and saturation flow rate of the turn lanes. On the other hand, the turn lane markings in Japan, provided a positive impact to mobility and safety of the turning lanes. Moreover, it was also found that the geometric characteristics and traffic signal phasing scheme highly affects the capacity and safety condition of signalized intersections.     

基于工程分解的船体生产物流管理研究

船舶建造的物流特点表现为:物流量大、周期长且具有动态性。零件生产能力不足、理料效率低下和配送困难是国内造船企业在实施精益造船、建立现代造船模式的过程中普遍遇到的船体结构建造问题。通过对某船厂的生产布局进行分析,运用工程分解的理论对船体生产物流管理模式进行深入探讨,并提出新的结构流向设计原则,以期提高管理的精细化与数字化,最终形成规范的物流管理体系。