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.     

An ontology-based framework to support performance monitoring in public transport systems

Managers of public transport systems have been facing for years the strategic challenge of maintaining high quality of transport services to improve the mobility of citizens, while reducing costs and ensuring safety and low environmental impact. A well-established way to evaluate the performance achieved by the system or by specific activities is to monitor Key Performance Indicators (KPI). However, existing management systems, which refer to flexible yet large and complex data models, provide a limited support to define and select relevant KPIs for the objectives at hand, and even the identification of whether and how the data model is capable to achieve a certain informative need is a critical and time-consuming task. This work is aimed to propose a framework to ease the development of a monitoring system in the public transport domain. The approach is based on the ontological representation of all the knowledge regarding indicators and their formulas, business objectives, dimension analysis and their relation with the Transmodel, the European reference data model for public transport information systems. On its top, a reasoning framework provides logic functionalities to interactively support designers in a set of common design tasks: the choice of the most suitable indicators for the performance monitoring needs at hand, the definition of new indicators and the identification of the minimal set of Transmodel modules needed to calculate them. A case study is included to discuss these applications, while an evaluation shows the feasibility of the approach.     

呼和浩特市优先发展城市公共交通探讨

通过对呼和浩特市城市公共交通现状的分析,指出呼和浩特市公共交通存在的问题,进一步说明保证呼和浩特市公共交通优先发展在缓减交通压力、提高城市公共客运效益、环境保护方面的重要意义,同时提出呼和浩特市优先发展公共交通应采取的措施.     

The effects of planned and unplanned transfers on public transport users' perception of transfer routes

Many authorities are investing in new infrastructure to improve the quality of public transport (PT) services in the hope to increase mode switch from cars. The goal is to provide users with an integrated multimodal PT network by facilitating transfers. There exists a lack in the clarity of the attributes which defines a planned transfer and the effects of planned transfers on users' willingness to use routes with transfers. The present study provides approximate effects of ‘planned’ and ‘unplanned’ transfers on PT users' decisions to use transfer routes. The study focuses on two attributes of ‘planned’ transfers, integrated physical connection of transfers and information integration. A user preference survey was undertaken in Auckland, New Zealand. Analysis of the results shows that physical integration is more important than information integration for current PT users. Results also suggest that information integration has a greater influence on users of transfer services which are more closely aligned to being ‘unplanned’.     

基于宜兴市居民出行特征的客运交通发展对策研究

居民出行调查是城市交通规划、建设和管理的基本依据．本文根据宜兴市2002年居民出行调查的有关数据,对人均出行次数、平均出行时耗、出行方式构成、出行目的和出行时间分布等进行了分析,并在此基础上,结合宜兴市交通现状及城市总体规划,从优先发展公共交通、重视自行车交通、控制摩托丰发展、加强交通管制、强化市场机制、加强企业管理和政府扶持相结合以及颁布公交行业服务标准等方面,提出了宜兴市客运交通发展的几点对策．     

公路客运市场退出影响因素实证分析

通过对广东省公路客运企业和政府主管部门的问卷调查,采用因子分析方法,从影响客运市场退出的相关主体及相关行为中筛选出关键变量．运用非参数检验过程,就各关键变量对退出的影响程度进行比较,最终得出驾驶员和承包经营人的相关行为对公路客运市场退出最具影响的结论,为政府部门进行行业管理和政策制定提供了实证依据．     

TOD交通走廊形成机理分析及经验借鉴——以美国阿灵顿县R-B走廊为例

TOD（Transit-Oriented Dovelopment,公共交通为导向的城市土地开发）交通走廊是以大运量的客运走廊为载体,通过选择若干站点进行TOD模式的开发,突破传统的单个节点开发形式,充分发挥规模效应,形成整合优势．本文以华盛顿特区西南部的阿灵顿县（Arlington County）的Rosslyn-Ballston轴状走廊（下文简称R—B走廊）为例,从宏观、中观和微观三个层次分析其形成机理,总结出TOD交通走廊建设的成功经验．     

布达佩斯城市交通发展战略与启示

分别从概况、战略目标、实施策略等3个方面对布达佩斯城市交通发展的总体战略及公共交通、道路网络、静态交通、步行及自行车交通、交通管理等发展战略进行了阐述。由此得出,居民出行线路上任何一种交通方式的低效都将破坏整个交通链的效率,只有整个交通链所有交通环节的和谐发展,才能够使公共交通在运输服务上较小汽车具有竞争力;只有综合利用公共交通与私人交通,才能使城市交通与土地利用相协调。最后,结合布达佩斯和国内实践经验,总结提出了一种包含人口、经济、地理、环境、技术、制度等要素的“六棱形”城市交通发展战略规划模型。     

北京市小汽车交通发展研究

首先对北京市小汽车发展的历程进行了回顾,接着分析了现状和特点,并对北京市小汽车发展引发的诸多经济社会问题进行了反思。认为当前小汽车发展是超前消费,超过了经济发展速度,与城市化发展不协调,小汽车在交通出行结构中所占比例过高,加重了交通拥堵,给资源、环境造成了沉重的负担和影响。从可持续发展的理念出发,北京市小汽车发展的总体策略应该是：适量拥有、合理使用。政府管理部门要树立“公交优先”的思想,采取各种管理手段缓解交通拥堵,减少交通事故。