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.     

A planning tool for evaluating vehicles miles travelled and traffic safety forecasts of growth management scenarios: A case study of Baton Rouge and New Orleans

This study describes an adaptable planning tool that examines potential change in vehicle miles travelled (VMT) growth and corresponding traffic safety outcomes in two urbanized areas, Baton Rouge and New Orleans, based on built environment, economic and demographic variables. This model is employed to demonstrate one aspect of the potential benefits of growth management policy implementation aimed at curbing VMT growth, and to establish targets with which to measure the effectiveness of those policies through a forecasting approach. The primary objective of this research is to demonstrate the need to break with current trends in order to achieve future goals, and to identify specific policy targets for fuel prices, population density, and transit service within the two study regions. Models indicate based on medium growth scenarios, Baton Rouge will experience a 9 percent increase in VMTs and New Orleans will experience 10 percent growth. This translates to corresponding increases in crashes, injuries and fatalities. The paper provides forecasts for planners and engineers to consider an alternative future, based on desired goals to reduce VMTs and therefore improve safety outcomes. A constrained-forecast model shows a cap on VMTs and crash rates is achievable through policy that increases fuel prices, population density and annual transit passenger miles per capita at reasonable levels through a growth management approach.     

城市大型医院周边道路交通问题剖析及改善对策——以深圳市为例

为了合理科学组织医院周边及内部交通,避免重现目前深圳大型医院周边普遍存在的交通拥堵、停车困难、进院车辆排队、的士乘坐不便等交通问题,以深圳市为例,分析现状大型医院周边存在的道路交通问题和产生原因,提出完善周边道路交通设施、大力发展公共交通、合理规划医院布局、建立科学有效的交通组织体系等改善对策.     

我国城市防洪排涝地下深隧规划设计与施工方法

为了建设符合中国国情的地下排水深隧,解决城市内涝问题,消除"城市看海"现象,采用调研分析的研究方法,通过调研国外城市排水深隧建设的成功案例,借鉴成熟经验并结合我国国情探索出我国城市排水深隧在规划设计、施工技术等方面的建设模式。主要结论如下:1)深隧规划必须坚持雨污分流原则,防涝标准应设为百年一遇;2)深隧层位应建于地下30~50 m的位置,隧道管线应布设在城市易涝区,并且将修建深隧、改造浅管和清淤河道衔接起来,形成一套完整的防洪排涝系统,让深隧的作用得以充分发挥,才能解决城市内涝问题;3)排水深隧要注重防水设计,防止雨水污染地下水;4)为提高深隧的经济效益,应将深隧与地下快速路的规划结合起来,功能上多样化。     

奥运与北京城市轨道交通规划建设

北京奥运会的召开促使北京社会经济发展进入了新阶段,奥运的高标准、高要求推动了城市基础设施的建设,特别是引领北京城市轨道交通建设进入了飞速发展时期。简要论述了奥运前和近期北京城市轨道交通建设规划概况,并阐述了可持续发展、与城市重大功能区发展一体化、交通一体化的轨道交通规划理念。介绍了奥运轨道交通线网覆盖范围,在此基础上提出了奥运交通圈层划分,并重点对奥运支线的线路运营组织方案做了分析。最后,指出要加强对奥运轨道交通车站周边交通衔接设施的使用,进一步扩大公共交通系统的服务范围。     

高速公路直线路段突起路标间隔设置方法

基于驾驶人的视觉惰性、视觉警示频率和感觉特性,分析了突起路标的视觉警示作用,确定了突起路标的闪现频率,计算与修正了突起路标的初始间隔。应用ADAMS/Car仿真软件模块,建立了道路、车辆和突起路标模型,将车型设置为大型车和小型车,大型车的车速设置为60、80、100km·h-1,小型车的车速设置为80、100、120km·h-1,大型车和小型车的方向盘转角均设置为1°、3°、5°。分别对提出的12m间隔与现行规范推荐的15m间隔各进行243次仿真,分析振动警示效果。仿真结果表明:在12m间隔下,平均碾压率为93.1%,小型车和大型车对应的平均警示率分别为41.7%和5.6%;在15m间隔下,平均碾压率为93.7%,小型车和大型车对应的平均警示率分别为33.3%和28.9%;12、15m间隔对小型车都具有较好的振动警示效果,15m间隔对大型车的振动警示效果更好。当高速公路夜间交通流较大或需要加大交通设施的夜间安全效果时,可选12m间隔提供良好的视觉连续性和视觉警示效果;考虑施工维护的方便性和经济性或者大型车比例较大时,可选择15m间隔。     

《深圳市综合交通体系规划((2013—2030)》编制探索——以更具竞争力、可持续的综合交通体系促进城市转型发展

通过回顾评价深圳市历次综合交通体系规划的得失,分析城市转型对于综合交通体系的挑战以及新形势下综合交通体系规划的定位。以《深圳市综合交通体系规划(2013—2030)》为例,对规划编制的内容和方法开展研究,具体包括:转型导向的交通发展目标与指标体系、一体化的综合交通体系组织、立足区域协作的整体交通战略、与组团结构相匹配的交通分区组织、与城市功能布局的协调互动、注重协同的交通管控政策,以及应对大数据的智能交通策略等。最后介绍规划编制组织特点并提出城市转型期综合交通体系规划的进一步思考。     

深圳市货运节点功能重组与空间重构

当前在产业转移升级趋势明显、土地资源加剧紧张的状况下,如何将经济部门的物流产业及园区规划、运输部门的行业发展规划、城市规划部门的综合交通体系规划三者协调统一,是各大城市亟须解决的重点和难点问题。对此,《深圳市综合交通体系规划(2013—2030)》开展了以重新组织城市货运空间为重点的货运规划实践。首先对深圳市货运系统发展现状进行分析,阐述货运系统在用地、运输方式、场站布局、货运通道等方面存在的问题。其次,梳理深圳市货运交通规划面临的行政管理、规划编制和落实、规划方法适应性、物流用地空间等方面的挑战。最后,以物流功能集聚为出发点,拓展货运枢纽的功能和属性,提出枢纽分级、立体化布局、公共空间共享专营等规划方法与实施机制。     

武汉市综合交通体系规划构想

结合武汉市新一轮城市总体规划、土地利用规划和综合交通体系规划"三规"同步修编,如何做好新时期综合交通体系规划修编工作成为当前亟须思考的问题。首先对城市综合交通体系规划的地位、作用及存在的不足进行分析。回顾武汉市已开展的三轮综合交通体系规划实践,指出既有规划在一体化发展、枢纽衔接与合作、土地开发模式、交通与空间发展的协调性等方面的不足。结合武汉市新一轮综合交通体系规划修编,从交通引领发展、建设枢纽城市、交通规划转型、优化交通结构等方面阐述需要重点研究并解决的问题。规划修编强调综合交通体系规划、城市总体规划、土地利用规划"三规"高效衔接与互动融合,通过一系列政策保障城市交通行动计划的推进。     

高速艇水翼减阻方案及翼滑艇阻力估算方法

文章介绍了高速艇上水翼减阻的原理以及三种不同类型的高速艇上加装水翼的技术方案及其达到的减阻效果,并给出了滑行艇首部加装水翼（即翼滑艇）后整船阻力的估算方法。基于三维非线性涡格法,建立了单独水翼/水翼组合体/多水翼系统的水动力性能理论计算方法,计算结果与试验结果吻合较好,可作为翼滑艇阻力估算中单独水翼水动力性能的计算方法。算例结果表明,文中的方法可用于单独水翼/水翼组合体/多水翼系统和滑行艇加装减阻水翼的初步技术方案设计。