【作者】 杨立才；

【导师】 贾磊；

【作者基本信息】 山东大学 ， 控制理论与控制工程， 2005， 博士

【摘要】 随着社会经济的不断发展与城市化进程的加快,城市人口和机动车辆日益增加,城市道路交通的拥挤与阻塞已经成为世界大中城市普遍存在的现象,由此带来的交通拥堵、交通事故、能源浪费以及成倍增加的汽车废气排放造成的环境污染等问题,不仅严重地制约着城市和社会经济的可持续发展,同时也严重地影响着城市居民的生活质量。 智能交通系统是解决现代社会交通需求与供给矛盾的重要途径之一。实施智能交通系统工程不仅有利于提高交通的安全性、生产效率与效益,而且关系到土地资源与能源的合理利用、环境污染与噪声的改善,乃至国民经济的持续发展和社会经济效益的全面提高。 智能交通系统是一个发展中的交叉学科,涉及到交通运输系统的各个层面。城市交通是交通系统最为重要的组成部分,城市交通智能控制是实施智能交通系统工程的首要任务,也是目前交通控制工程领域研究的热点问题之一。城市道路交通控制是一个复杂的系统工程问题,涉及到交通工程、自动控制、系统工程、优化调度等自然科学和工程技术的众多学科。作为多学科交叉的研究领域,城市交通智能控制技术的发展依赖于自然科学和工程技术的最新研究成果。把自然科学的最新研究成果和工程技术的最新方法引入城市交通智能控制,可以进一步完善城市交通控制的理论体系及应用,解决日益严重的城市交通难题。这对满足社会需求、推动国家和社会的进步以及学科的发展,无疑都具有十分重要的意义。 城市道路交通是一个有人参与的时变复杂大系统,集成了人、车、路和环境等各种复杂因素,存在高度的复杂性、时变性和不确定性。基于精确数学模型的传统控制方法难以有效解决复杂的现代城市交通问题,基于人工智能的智能控制技术是解决城市交通问题的有效途径。 本文基于模糊逻辑、混沌优化、人工神经网络、人工免疫系统和粗集更多还原

【Abstract】 With the increasing development of social economy and urbanization, the urban population and vehicles increase rapidly. Traffic congestion and traffic jam have become prevalent problems for metropolis all over the world. Traffic accident, energy wasting, air pollution caused by exhaust gas and other problems resulted from traffic congestion and traffic jam not only seriously restrict the sustainable development of social economy, but also severely influence the urban living environment.Intelligent transportation system is one of the important ways to solve the antinomy of traffic demand and supply in the modem society. The application of intelligent transportation system will not only improve the transportation safety, production efficiency and revenues, but also connect with land resources and energies exploitation, environment improvement, and national economic and social revenue development.As a developing multidisciplinary subject, intelligent transportation system relates to many facets of the traffic transportation systems. Urban road traffic is the most important part of the traffic transportation systems. Urban traffic intelligent control is the first step to implement the intelligent transportation systems engineering, and also a hot issue in control engineering and traffic engineering fields. Urban road traffic control is a complex systems engineering problem. It involves many subjects of science and technology, such as traffic engineering, automatic control, systems engineering, and optimization scheduling. Urban traffic intelligent control is a multidisciplinary research subject, and the development of urban traffic intelligent control technology depends on the newest research outcomes of science and technology.Applying the newest research of science and technology to urban traffic control systems can consummate traffic control theories and solve increasingly serious urban traffic problems, which is of the most important significance for meeting the social demand, accelerating the progress ofnation and society, and driving the development of subject.Urban road traffic system, as a time-dependent complex great system into which integrates human, vehicles, roads, environment and other complex factors, is of high complexity, time-dependence and randomicity. It is difficult for traditional control methods based on precise mathematical models to solve complex modern urban traffic problems, and the intelligent control technology based on artificial intelligence is a feasible way to solve urban traffic problems.Based upon fuzzy logical, chaos optimization, artificial annual networks, artificial immune systems, rough sets and other artificial intelligence technologies, this dissertation makes a comprehensive and deep research on urban traffic signal control strategies, dynamic route guidance system and the concerned traffic control technologies. After reviewing the traditional traffic control studies, this dissertation puts forward several advanced traffic control ideas, algorithms and models. The main contents in this dissertation include:1) Reviewing the urban road traffic control technology in detail, and deducing the main research contents after making a discussion on the difficulties and problems in this field.2) The fuzzy control of adjacent intersections is studied. A traffic coordination control algorithm for two adjacent intersections using hierarchical fuzzy logic and the idea of adjusting membership functions automatically using chaos optimization are put forward. The two-stage fuzzy controller with this algorithm is designed and simulation results show that the average delay of vehicles in intersections is less using this algorithm than that using the conventional control strategies.3) Concerning the fact that the span between some natural intersections of urban traffic trunk roads is smaller than usual, this dissertation brings forward a new idea to integrate two or three natural intersections whose span is smaller than usual to a built-up intersection, and the concept of "Big Intersection" as well as the control model and algorithm for it is given. In this research, the traffic coordination control algorithm for two adjacent intersections using hierarchical fuzzy logic and the idea of adjustingmembership functions automatically using chaos optimization are generalized to urban trunk traffic control systems, and the urban trunk traffic fuzzy control mechanism using "Big Intersection" is established.4) The urban road traffic route guidance algorithm is studied. To improve the performance of Kth shortest route algorithms in dynamitic route guidance systems, this dissertation suggests a practical dynamitic route guidance stagery, in which a new algorithm based on the metaphors of vertebrate immune system and the ideals of intelligent optimization is proposed. Combined the urban traffic network models built via " extended node method" , this dissertation studies the ATth shortest route search problems in urban dynamitic route guidance systems using the artificial immune optimization algorithm, and discusses the interrelated issues of dynamic route guidance systems. The experiment results show that this route guidance stagery is effective and advanced.5) Combined rough sets theory with fuzzy reasoning together, a new method for modeling traffic fuzzy control systems is put forward to meet the needs of designing fuzzy controllers with high performance in intelligent transportation systems. This method extracts the fuzzy control rules from historical traffic data by knowledge reasoning in rough sets theory and solves the bottleneck problems for modeling urban traffic signal fuzzy control systems. A rough fuzzy model of four-phase intersections is constructed using the method given in this dissertation, and the rough fuzzy modeling method of multi-intersections with more practicabilities than isolated intersections is studied.6) To solve the problems of programming traffic intervals automatically for time of day control scheme in urban traffic control systems, an artificial immune data clustering algorithm based on metaphors from the vertebrate immune systems is put forward. This algorithm has been successfully used in making the time of day schemes of urban traffic control systems. It can get over the irrational intervals from manually programming methods and the hierarchical clustering algorithms based on genetic algorithm. This work supplies a new idea for programming time of day intervals and making urbantraffic control schemes.7) The traffic flow forecasting method based on the artificial neural networks and rough sets is studied. A new forecasting idea with the " similar intervals" is put forward to improve the flow forecasting performance. Combined rough sets with orthogonal wavelet networks together, a new traffic flow forecasting model is put forward, and the model has been successfully used for urban traffic flow forecasting applications. Integrating the excellent performances of wavelet networks and rough networks, combaining with the idea of " similar intervals" , the model can be used for the applications of real time flow accurate forecasting.Finally, we make a conclusion and propose the future research directions in this field.更多还原