路段动态特征中的交通组成因素研究

基于对交通流建模过程中将车型单一考虑为标准小汽车,忽略了现实交通量中车型的不同构成,本文通过对不同车型的动力特性进行标定,将其特性反映在跟车模型与车道变换模型中,并且利用交通仿真技术,分析了交通量构成对路段行程时间、车速、延误等动态特性指标的影响。     

汽车防追尾碰撞数学模型研究

为了提高车辆在高速行驶状态下的主动安全性能,研究了处于追尾行驶状态的本车与前车的运动学特征;针对前车的不同运动状态分别推导出了跟车距离的计算模型并分析了模型中3个关键参数的随机性和动态性,对制动迟滞时间提出了基于模糊推理的确定方法,对本车制动减速度和前车的运动加速度提出了比较实用的动态测算公式;另外,研究了防追尾碰撞的控制与执行,建立了动态调整安全制动停车距离的神经网络模型,提出了基于危险裕度判别的安全控制方法。     

考虑横向漂移的船舶路径跟踪自抗扰控制

针对船舶运动系统中内部动态不确定和外部干扰等问题,进行了欠驱动船舶路径跟踪的自抗扰方法研究。利用Backstepping设计参考航向角,并通过线性扩张状态观测器对流干扰和横向运动引起的横向漂移进行估计。其次,根据自抗扰算法对航向进行控制,采用线性扩张状态观测器对外界干扰及内部不确定项进行估计。最后仿真结果表明,在风流干扰下所设计的控制器仍能使船准确地跟踪上参考路径,验证了所提控制方案的有效性。     

Labview应用程序开发中使用支持文件相对路径的有关问题

用Labview开发涉及支持文件读写的应用程序时,常会出现调试正常而可执行文件运行时找不到支持文件路径的错误。基于一简单示例,分析了产生此问题的原因并提供了解决问题的思路。     

Human-like autonomous car-following model with deep reinforcement learning

This study proposes a framework for human-like autonomous car-following planning based on deep reinforcement learning (deep RL). Historical driving data are fed into a simulation environment where an RL agent learns from trial and error interactions based on a reward function that signals how much the agent deviates from the empirical data. Through these interactions, an optimal policy, or car-following model that maps in a human-like way from speed, relative speed between a lead and following vehicle, and inter-vehicle spacing to acceleration of a following vehicle is finally obtained. The model can be continuously updated when more data are fed in. Two thousand car-following periods extracted from the 2015 Shanghai Naturalistic Driving Study were used to train the model and compare its performance with that of traditional and recent data-driven car-following models. As shown by this study’s results, a deep deterministic policy gradient car-following model that uses disparity between simulated and observed speed as the reward function and considers a reaction delay of 1 s, denoted as DDPGvRT, can reproduce human-like car-following behavior with higher accuracy than traditional and recent data-driven car-following models. Specifically, the DDPGvRT model has a spacing validation error of 18% and speed validation error of 5%, which are less than those of other models, including the intelligent driver model, models based on locally weighted regression, and conventional neural network-based models. Moreover, the DDPGvRT demonstrates good capability of generalization to various driving situations and can adapt to different drivers by continuously learning. This study demonstrates that reinforcement learning methodology can offer insight into driver behavior and can contribute to the development of human-like autonomous driving algorithms and traffic-flow models.     

Investigating Causal Pathways Linking Site-Level Characteristics,Compliance, and Ecological Performance in Caribbean MPAs

A rapidly growing body of empirical evidence has linked improved ecological performance of social–ecological systems (SES) with better compliance with rules governing human activities. Researchers have proposed several key factors that motivate an individual to comply with resource use rules, including deterrence, social pressures, moral inclinations, and perceived legitimacy of rules and responsible authorities. Much of the research on compliance proposes causal mechanisms at the individual human level, but few studies test causal paths with empirical data particularly at the site level. This study tested a proposed pathway for explaining overall compliance rates and ecological performance in SES using data from structured surveys of community members, key informant interviews, policy documents, and coral reef surveys at marine protected areas (MPAs) and their associated human communities in the wider Caribbean. Findings indicate that compliance and other active MPA management interventions positively affect ecological performance, and site-level contextual factors (level of community development and being part of a political network) positively influence significant predictors of overall compliance rates, which include active MPA management and enforced punishments.     

考虑多人相互影响的舰船通道最短路径规划

将时间维引入既定舰船通道网络,把考虑多人相互影响的动态最优路径规划问题转化为时间依赖网络中的最优路径搜索问题。首先,论文的算法为所有人员随机生成走行路径,为了预测网络中的弧(路段)的走行时间,按人员速度从慢到快的顺序依次计算并记录人员到达路径中的各路段首节点的时刻,先记录到的人员将成为后记录到的人员的动态障碍。然后,将遗传算法与网络中弧的走行时间预测方法相结合,借助遗传算法的个体多样性天然地解决了人员走行任意性问题,因而获得了全局动态最优路径算法,并仿真计算了两人以及三人的最短时间路径;经与不考虑人员间相互影响时获得的最优路径相比较,论文的算法获得了人员遇到障碍时(或跟行或绕行)的最优走行路径。最后,借助时间依赖网络中的最优路径充要条件定理,说明了算法的有效性。     

浅析西门子BDPC系统在CTV上的 可行性

本文主要目的是分析西门子Blue Drive Plus C（随后简称BDPC）系统及其配置是否能够应用在深水动力定位原油输送船（随后称CTV）上。通过供需关系对比法,首先实现CTV实现海上自由航行、跟随定位、连接FPSO与～30万吨油轮之间浮管、加油跟随定位、紧急脱离、绿色环保运行等工况下的正常与应急储备动力需求,随后分析西门子BDPC直流系统及其运用技术,针对本项目提供的特殊配置,与CTV提出需求进行对比,结论是BDPC依靠它的直流发电、配电、逆变、储能等技术,可以满足CTV的配置需求,论证了BDPC系统运用在性能优越的深水原油输送装置CTV是可行的。     

An algorithm for finding reasonable paths in transit networks

In this paper, reasonable paths in transit networks are defined as possible paths that satisfy the acceptable time criterion and transfer‐walk criterion. A recursive algorithm for finding all of the reasonable paths in a transit network that does not involve a rapid increase in program run‐time with network size is presented. Realistic transit networks in Hong Kong and Guangzhou were selected as case studies of the different phases of the development of a trip planning system. Transport planning practitioners and potential users were invited to test the system to evaluate its performance. The results of the prototype evaluation were satisfactory, and the viability of the system as a useful tool for supporting decision‐making has been confirmed by the positive feedback that was obtained from survey questionnaires.