基于可靠性的绿色智能导航系统初步研究

Traffic congestion is a recurrent problem in densely populated cities such as Bei Jing. It has considerable impact on economic productivity, network reliability, environment and safety. However, due to the topography of most big cities, there are hardly any feasible sites for further expansion of existing road network. To alleviate the recurrent traffic congestion problem, recent attention has been given to develop intelligent transportation systems (ITS). There are a wide range of ITS for providing updated traffic information and route guidance by mode and traffic monitoring and control on urban roads etc. In this project, a reliability-based green road guidance system is proposed. Network reliability, especially travel time reliability issues have recently become very important and critical in making the ITS more efficient. In view of this, the travel time reliability is analyzed with consideration of congestion toll and recurrent congestion. On this basis, an efficient solution algorithm is proposed with the objective of minimizing the emission, in order to get the set of the shortest and the most reliable paths. Finally, the proposed models and algorithms are applied in real cases to recommend the reliable green path set with given on-time arrival probability.

随着经济发展，交通需求不断增长，交通拥堵问题已经成为世界各国城市发展面临的重要问题。大范围交通拥堵，不仅造成交通网络效率下降，而且使得尾气排放剧增，导致严重的环境污染。由于地域限制，很多城市的路网难以扩张，只能提高已有交通设施的利用率。智能交通系统，将现代信息技术和交通控制技术相结合，可以显著提高交通基础设施的使用效率，减少出行的盲目性，被认为是解决道路交通拥堵、减少尾气排污等交通问题的有效方法之一。本研究将以智能交通系统为平台，重点开发基于可靠性的绿色出行路径导航系统。首先，该系统会分析拥堵收费及日常交通拥堵对出行时间的影响，得到出行时间的不确定性。在此基础上，以减少汽车尾气排放为目标建立模型并设计求解算法，寻找出行时间短且可靠度高的出行路径集合。项目工作后续会应用开发的模型和算法进行实例研究，根据出行者不同需求，提供最环保可靠的出行路径集，使其在可以接受的交通延迟范围内到达目的地。

实际交通出行中，由于受到各方面随机事件影响，交通路网的供给和需求都可能是不确定的，这种不确定性使得行程时间也呈现随机性变化。不同出行者风险偏好不同，当行程时间随机变化时，会选择不同路径出行，确保可以按时到达目的地。项目从不同类型出行者对行程时间可靠性的要求，结合不同出行目的出行者的风险厌恶程度，提出负效用确定性等价作为行程时间可靠性的度量指标。并在该指标的基础上，研究了出行者路径选择规则。并在给定择路规则下建立交通网络用户配流均衡模型，引入可变车道建立混合双层网络设计模型，分别设计算法对所建立模型进行求解。项目较好的解决了前述问题，为智能交通系统设计提供支持。