信息反馈机制在智能交通系统中的研究与应用
|
摘要
交通运输能力是衡量一个国家发展程度的重要标准。它不仅对国民经济发展起着至关重要的作用,而且也制约着现代化发展的进程。因此各国政府都在鼓励科学家研发新型的交通系统。在此背景下,智能交通系统应运而生。近年来,作为智能交通系统的核心部分,信息反馈策略越来越受到人们的关注。自从2000年Wahle等人提出第一种信息反馈策略以来,经过十余年的发展,目前已有数十种反馈策略。本文就是在前人研究的基础上,针对不同的实时交通信息和道路结构设计出不同的信息反馈策略。具体的研究工作如下：
一、针对非对称含瓶颈的一维道路模型设计出分段函数反馈策略。该策略不仅能给出拥堵簇在道路上不同位置时的权重,还能区分出拥堵簇在瓶颈内外的权重。结合之前人们提出的时间信息反馈策略、平均速度反馈策略、拥堵系数反馈策略和加权拥堵系数反馈策略,我们将这五种策略应用到非对称含瓶颈双通道单出口道路模型上进行模拟。结果显示相比于其它四种策略,新策略不仅是在交通流量、平均速度和车辆密度等方面具有良好的稳定性,而且还能提高道路的平均流量。
二、根据加权类信息反馈策略的构造原理,提出了指数函数反馈策略。我们首先是详细的分析了前人提出的两种加权类信息反馈策略,根据它们的共同特点设计出指数函数反馈策略。然后将该策略和前人提出的三种与拥堵系数相关的反馈策略应用在两种一维道路模型上进行模拟。结果显示,与其它策略相比,指数函数反馈策略具有明显的优势。尽管指数函数反馈策略相比于角度拥堵系数反馈在平均流量上的优势不是很明显,但是它的应用范围更广,而且计算和实施起来比较方便。根据模拟结果我们还发现角度拥堵系数反馈策略和指数函数反馈策略并不适用于封闭式的交通系统。随后我们将指数函数应用在二维网络道路模型上时也验证了这一结论。
三、信息反馈策略在二维网络道路模型上的应用。我们详细的论述了在二维网络道路模型上应用随机选道策略、指数函数反馈策略、平均速度反馈策略和拥堵系数反馈策略时出现的三种交通状态以及它们在各种状态下的临界密度和车辆分布情况。从模拟结果中可以看出,平均速度反馈策略和拥堵系数反馈策略在各态下的临界密度都要大于其它两种策略。随后,我们研究了交通灯周期和交通灯变换规则对系统的影响。发现随着交通灯周期的增加,其对系统的影响在逐渐减弱,并且当系统密度处于饱和态时采用拥堵系数反馈策略和逆时针的交通灯变换规则更有利于提高系统的运输能力。进一步地,我们还研究了道路长度和等面积下网络边长对系统的影响。根据模拟结果我们发现对于高度对称的网络模型,均匀(而非加权)的反馈道路信息将更有利于提高系统的交通流量。
四、基于二维网络道路模型设计出了一种全局搜索加权路径的反馈策略,即行驶时间加权路径反馈策略。该策略不但能有效的提高网络的交通流量,还能将车辆均匀的分布在系统中。
五、根据流量的定义提出了空间流量反馈策略和时间流量反馈策略,并详细的阐述了制定两种策略规则的原因。我们将这两种策略应用在对称双通道的道路模型上与拥堵系数反馈策略、预测信息反馈策略、车辆数反馈策略和加权拥堵系数反馈策略做对比。结果显示空间流量反馈策略在平均流量的数值方面是最优的。进一步地,我们还将上述六种策略应用在对称三通道三出口和非对称双通道双出口的道路模型上进行模拟。从模拟结果中可以发现空间流量反馈策略依然能保持它的优越性。
六、提出了一种基于局部信息的反馈策略：空位长度反馈策略。结合之前研究者们所提出的九种信息反馈策略,我们将这些策略统一应用到一维道路模型上进行模拟。从模拟结果中可以看出,在开放式的交通系统中,基于局部信息建立的反馈策略要明显好于基于全局信息建立的反馈策略。而在局部信息反馈策略中空位长度反馈策略无论是从道路的通行能力来讲,还是从策略实施的简便性来讲,都是最优的。最后我们总结出：对于开放式交通系统,离入口处越远的道路状况对道路的通行能力影响越小。因此我们只需要反馈入口处的局部信息就能提高道路的通行能力。
文章的最后,我们对全文的工作和创新点进行总结并对未来的研究工作提出展望。
The comprehensive capacity of transportation system is a key index to mea-sure the prosperity of a country. Not only is it crucial to the national economic development, but it also determines the modernizing progress of a country. There-fore, governments of various countries are encouraging scientists to develop the new types of traffic systems. Under this background, intelligent transportation system has come into being. In recent years, information feedback strategy serving as the critical part of intelligent transportation system has been treated with ever-increasing emphasis. Since Wahleet al. proposed the first information feedback strategy in2000, a dozen of them have been introduced after more than a decade. On the basis of previous study, my contribution in this field is to design various feedback strategies based on real-time traffic information and road structures. The research works are specified as follows:
1. In the light of the characteristics of asymmetrical two-route scenario with bottleneck and one exit, we designed the piecewise function feedback strategy. Be-sides that the new strategy can distinguish the weight of congestion clusters when they are in different positions on the road, it even gives out the distinction weight of clusters whether they are in the bottleneck or not. Combined with the travel time feedback strategy, mean velocity feedback strategy, congestion coefficient feedback strategy and weighted congestion coefficient feedback strategy raised by former re-searchers, we apply these four strategies into the asymmetrical two-route scenario with bottleneck and one exit for simulation. The simulation results suggest that piecewise function feedback strategy is the optimal one among all the feedback strategies. It outperforms others in terms of the stability of the traffic flux, average speed and vehicle density, and also exceeds others for the value of average flow.
2. Built on the design principles of weighted information feedback strategies, a new strategy is proposed, namely the exponential function feedback strategy. After analyzing two previous strategies, we find the reason why corresponding angle feedback strategy is superior to weighted congestion coefficient feedback strategy. Given that the sharp decay of the weighted coefficient is the key point, we propose the exponential function feedback strategy. This new strategy together with three other strategies related to the congestion coefficient are all applied into two kinds of one dimensional road model for simulation. The simulation results show that compared with other strategies, exponential function feedback strategy has distinct advantages in vehicle number and flux. Although the advantage of the average flow of the new strategy is not obvious at the side of the corresponding angle feedback strategy, the exponential function feedback strategy is applicable to more complex models and relatively convenient to employ. According to the simulation results, we even find that the exponential function feedback strategy and the corresponding angle feedback strategy are unsuited to be applied into the closed transportation systems. This conclusion will be came to again by the following research.
3. The application of feedback strategies in the two-dimensional road model. In this section, three different states, critical density as well as the distribution of vehicles are discussed in detail when the randomized choosing route strategy, expo-nential function feedback strategy, mean velocity feedback strategy and congestion coefficient feedback strategy are applied into the two-dimensional road model. The results display that the critical density of the mean velocity feedback strategy and congestion coefficient feedback strategy are higher than that of other two strategies. Next, the effect of the traffic light period and the traffic light rules are investigated. The results indicate that with increasing the traffic light period, its influence on the systems decreases. As the density is in the interval of saturation state, adopting the congestion coefficient feedback strategy with the anticlockwise traffic light rule will be more conducive to improve the transportation capability. In addition, we also study the effect of the route length and the side length of the traffic network. We concluded that in the highly symmetric network, equably feeding back the traffic information will help to improve the traffic flow.
4. On the basis of two-dimensional road model, a feedback strategy based on global information to search the minimum weighted path is introduced, called the route weighted by travel time feedback strategy. This new strategy can not only enhance the traffic flow, but have the vehicle evenly distributed as well.
5. Acting on the definition of traffic flux, we present the space flux feed-back strategy and time flux feedback strategy and explain the regulations of them. This two strategies together with congestion coefficient feedback strategy, predic- tion information feedback strategy and vehicle number feedback strategy as well as weighted congestion coefficient feedback strategy are applied into the symmetrical two-route model. And then, we obtain the simulations of the changing of vehicle number and flux according to time together with that of average flow according to the ration of dynamic vehicles. The results show that space flux feedback strategy enjoys great superiority over other strategies in terms of the average flow. With respect to other quantities such as the stability of vehicle number and flux, the new strategy is not inferior. Moreover, we applied these strategies into the symmetrical three-route model and asymmetrical two-route model with two exits. The results again demonstrate the superiority of space flux feedback strategy.
6. Depending on the local traffic information, we present the vacancy length feedback strategy which united with nine previous strategies are all applied in the symmetrical two-route scenario with two exits and the asymmetrical two-route sce-nario with one exit. The simulation results turn out that the superiority of the feedback strategies based on the local information over the ones based on the global information and the vacancy length feedback strategy is the best one among all the feedback strategies based on the local information in terms of both the improving capacity of transportation systems and convenience of its application. Finally, we conclusion that in the open traffic systems the further the vehicle is away from the entrance, the less influence its condition imposes on the traffic capability. In order to improve the capability, the information of the entrance only need to be considered.
In the end of this thesis, we summarized our research works and propose the outlook of future research.
一、针对非对称含瓶颈的一维道路模型设计出分段函数反馈策略。该策略不仅能给出拥堵簇在道路上不同位置时的权重,还能区分出拥堵簇在瓶颈内外的权重。结合之前人们提出的时间信息反馈策略、平均速度反馈策略、拥堵系数反馈策略和加权拥堵系数反馈策略,我们将这五种策略应用到非对称含瓶颈双通道单出口道路模型上进行模拟。结果显示相比于其它四种策略,新策略不仅是在交通流量、平均速度和车辆密度等方面具有良好的稳定性,而且还能提高道路的平均流量。
二、根据加权类信息反馈策略的构造原理,提出了指数函数反馈策略。我们首先是详细的分析了前人提出的两种加权类信息反馈策略,根据它们的共同特点设计出指数函数反馈策略。然后将该策略和前人提出的三种与拥堵系数相关的反馈策略应用在两种一维道路模型上进行模拟。结果显示,与其它策略相比,指数函数反馈策略具有明显的优势。尽管指数函数反馈策略相比于角度拥堵系数反馈在平均流量上的优势不是很明显,但是它的应用范围更广,而且计算和实施起来比较方便。根据模拟结果我们还发现角度拥堵系数反馈策略和指数函数反馈策略并不适用于封闭式的交通系统。随后我们将指数函数应用在二维网络道路模型上时也验证了这一结论。
三、信息反馈策略在二维网络道路模型上的应用。我们详细的论述了在二维网络道路模型上应用随机选道策略、指数函数反馈策略、平均速度反馈策略和拥堵系数反馈策略时出现的三种交通状态以及它们在各种状态下的临界密度和车辆分布情况。从模拟结果中可以看出,平均速度反馈策略和拥堵系数反馈策略在各态下的临界密度都要大于其它两种策略。随后,我们研究了交通灯周期和交通灯变换规则对系统的影响。发现随着交通灯周期的增加,其对系统的影响在逐渐减弱,并且当系统密度处于饱和态时采用拥堵系数反馈策略和逆时针的交通灯变换规则更有利于提高系统的运输能力。进一步地,我们还研究了道路长度和等面积下网络边长对系统的影响。根据模拟结果我们发现对于高度对称的网络模型,均匀(而非加权)的反馈道路信息将更有利于提高系统的交通流量。
四、基于二维网络道路模型设计出了一种全局搜索加权路径的反馈策略,即行驶时间加权路径反馈策略。该策略不但能有效的提高网络的交通流量,还能将车辆均匀的分布在系统中。
五、根据流量的定义提出了空间流量反馈策略和时间流量反馈策略,并详细的阐述了制定两种策略规则的原因。我们将这两种策略应用在对称双通道的道路模型上与拥堵系数反馈策略、预测信息反馈策略、车辆数反馈策略和加权拥堵系数反馈策略做对比。结果显示空间流量反馈策略在平均流量的数值方面是最优的。进一步地,我们还将上述六种策略应用在对称三通道三出口和非对称双通道双出口的道路模型上进行模拟。从模拟结果中可以发现空间流量反馈策略依然能保持它的优越性。
六、提出了一种基于局部信息的反馈策略：空位长度反馈策略。结合之前研究者们所提出的九种信息反馈策略,我们将这些策略统一应用到一维道路模型上进行模拟。从模拟结果中可以看出,在开放式的交通系统中,基于局部信息建立的反馈策略要明显好于基于全局信息建立的反馈策略。而在局部信息反馈策略中空位长度反馈策略无论是从道路的通行能力来讲,还是从策略实施的简便性来讲,都是最优的。最后我们总结出：对于开放式交通系统,离入口处越远的道路状况对道路的通行能力影响越小。因此我们只需要反馈入口处的局部信息就能提高道路的通行能力。
文章的最后,我们对全文的工作和创新点进行总结并对未来的研究工作提出展望。
The comprehensive capacity of transportation system is a key index to mea-sure the prosperity of a country. Not only is it crucial to the national economic development, but it also determines the modernizing progress of a country. There-fore, governments of various countries are encouraging scientists to develop the new types of traffic systems. Under this background, intelligent transportation system has come into being. In recent years, information feedback strategy serving as the critical part of intelligent transportation system has been treated with ever-increasing emphasis. Since Wahleet al. proposed the first information feedback strategy in2000, a dozen of them have been introduced after more than a decade. On the basis of previous study, my contribution in this field is to design various feedback strategies based on real-time traffic information and road structures. The research works are specified as follows:
1. In the light of the characteristics of asymmetrical two-route scenario with bottleneck and one exit, we designed the piecewise function feedback strategy. Be-sides that the new strategy can distinguish the weight of congestion clusters when they are in different positions on the road, it even gives out the distinction weight of clusters whether they are in the bottleneck or not. Combined with the travel time feedback strategy, mean velocity feedback strategy, congestion coefficient feedback strategy and weighted congestion coefficient feedback strategy raised by former re-searchers, we apply these four strategies into the asymmetrical two-route scenario with bottleneck and one exit for simulation. The simulation results suggest that piecewise function feedback strategy is the optimal one among all the feedback strategies. It outperforms others in terms of the stability of the traffic flux, average speed and vehicle density, and also exceeds others for the value of average flow.
2. Built on the design principles of weighted information feedback strategies, a new strategy is proposed, namely the exponential function feedback strategy. After analyzing two previous strategies, we find the reason why corresponding angle feedback strategy is superior to weighted congestion coefficient feedback strategy. Given that the sharp decay of the weighted coefficient is the key point, we propose the exponential function feedback strategy. This new strategy together with three other strategies related to the congestion coefficient are all applied into two kinds of one dimensional road model for simulation. The simulation results show that compared with other strategies, exponential function feedback strategy has distinct advantages in vehicle number and flux. Although the advantage of the average flow of the new strategy is not obvious at the side of the corresponding angle feedback strategy, the exponential function feedback strategy is applicable to more complex models and relatively convenient to employ. According to the simulation results, we even find that the exponential function feedback strategy and the corresponding angle feedback strategy are unsuited to be applied into the closed transportation systems. This conclusion will be came to again by the following research.
3. The application of feedback strategies in the two-dimensional road model. In this section, three different states, critical density as well as the distribution of vehicles are discussed in detail when the randomized choosing route strategy, expo-nential function feedback strategy, mean velocity feedback strategy and congestion coefficient feedback strategy are applied into the two-dimensional road model. The results display that the critical density of the mean velocity feedback strategy and congestion coefficient feedback strategy are higher than that of other two strategies. Next, the effect of the traffic light period and the traffic light rules are investigated. The results indicate that with increasing the traffic light period, its influence on the systems decreases. As the density is in the interval of saturation state, adopting the congestion coefficient feedback strategy with the anticlockwise traffic light rule will be more conducive to improve the transportation capability. In addition, we also study the effect of the route length and the side length of the traffic network. We concluded that in the highly symmetric network, equably feeding back the traffic information will help to improve the traffic flow.
4. On the basis of two-dimensional road model, a feedback strategy based on global information to search the minimum weighted path is introduced, called the route weighted by travel time feedback strategy. This new strategy can not only enhance the traffic flow, but have the vehicle evenly distributed as well.
5. Acting on the definition of traffic flux, we present the space flux feed-back strategy and time flux feedback strategy and explain the regulations of them. This two strategies together with congestion coefficient feedback strategy, predic- tion information feedback strategy and vehicle number feedback strategy as well as weighted congestion coefficient feedback strategy are applied into the symmetrical two-route model. And then, we obtain the simulations of the changing of vehicle number and flux according to time together with that of average flow according to the ration of dynamic vehicles. The results show that space flux feedback strategy enjoys great superiority over other strategies in terms of the average flow. With respect to other quantities such as the stability of vehicle number and flux, the new strategy is not inferior. Moreover, we applied these strategies into the symmetrical three-route model and asymmetrical two-route model with two exits. The results again demonstrate the superiority of space flux feedback strategy.
6. Depending on the local traffic information, we present the vacancy length feedback strategy which united with nine previous strategies are all applied in the symmetrical two-route scenario with two exits and the asymmetrical two-route sce-nario with one exit. The simulation results turn out that the superiority of the feedback strategies based on the local information over the ones based on the global information and the vacancy length feedback strategy is the best one among all the feedback strategies based on the local information in terms of both the improving capacity of transportation systems and convenience of its application. Finally, we conclusion that in the open traffic systems the further the vehicle is away from the entrance, the less influence its condition imposes on the traffic capability. In order to improve the capability, the information of the entrance only need to be considered.
In the end of this thesis, we summarized our research works and propose the outlook of future research.
引文
[1]韩正.以公共交通为主导推进上海城市交通建设[J].城市公交规划与管理,1999(6)：2-4.
[2]戴世强,冯苏苇,顾国庆.交通流动力学：它的内容、方法和意义[J].自然杂志,1997(4)：196-201.
[3]徐东云,张雷,兰荣娟.城市交通拥堵的背景变换分析[J].城市问题,2009(3)：46-49.
[4]Helbing D, Hennecke A, Shvetsov V, et al. MASTER:macroscopic traffic sim-ulation based on a gas-kinetic, non-local traffic model[J]. Transportation Re-search Part B,2001(35):183-211.
[5]贾述评,房筱莉.日本智能交通系统的发展[J].吉林交通科技,2006(3)：67-68.
[6]陈旭梅,于雷,郭继孚,等.美国智能交通系统ITS的近期发展综述[J].中外公路,2003,23(2)：9-12.
[7]中辛.美、日、欧洲ITS研究开发项目[J].东北汽车运输,2000(4)：24-26.
[8]范建军,缪立新,史其信.ITS在亚太地区的发展[J].中国物流与采购,2003(3)：24-25.
[9]吴俊.智能交通系统中的信息处理关键技术研究[D].天津：天津大学,2006.
[10]Kuospe W. Single-vehicle data of highway Ixaffic:Microscopic description of traffic phases. Physical Review E,2002(65):056133.
[11]贾斌,高自友,李克平,李新刚.基于元宝自动机的交通系统建模与模拟[M].北京：科学出版社,2007.
[12]Lighthill M J, Whitham G B. On kinematic waves:II. a theory of traffic flow on long crowded roads [J]. Proceedings of the Royal Society A,1955(229):317-345.
[13]Richards P I. Shock waves on the highway[J]. Operations Research,1956(4): 42-51.
[14]Payne H J. Models of freeway traffic and control[M]. La Jolla:Simulation Council.1971.
[15]Jiang R, Wu Q S, Zhu Z J. A new continuum model for traffic flow and nu-merical tests[J]. Transporation Research Part B,2002(36):405-419.
[16]Siebel F, Mauser W. On the fundmental diagram of traffic flow[J]. SIAM Jour-nal on Applied Mathematics,2006(66):1150-1162.
[17]Siebel F, Mauser W. Synchronized flow and wide moving jams from balanced vehicular traffic[J]. Physical Review E,2006(73):066108.
[18]Paveri-Fontana S L. On Boltzmann-like treatments for traffic flow. A critical review of the basic model and an alternative proposal for dilute traffic analysis. Tranpration Research,1975(9):225-235.
[19]Helbing D. Derivation and empirical validation of a refined traffic flow model[J]. Physica A,1996(233):253-282.
[20]Nagatani T. Kinetic segregation in a multilane highway traffic flow[J]. Physica A,1997(237):67-74.
[21]Treiber M, Hennecke A, Helbing D. Derivation, properties and simulation of a gas-kinetic-based nonlocal traffic model[J]. Physical Review E,1999(53):239-253.
[22]Hoogendoorn S P, Bovy P H L. Generic gas-kinetic traffic systems modeling with applications to vehicular traffic flow[J]. Transprotation Research Part B, 2001(35):317-336.
[23]Reuschel A. Vehicle movements in a platoon [J]. Oesterreichisches Ingenieur-Archir,1950(4):193-215.
[24]Helbing D, Tilch B. Generalized force model of traffic dynamics[J]. Physical Review E,1998(58):133-138.
[25]Nagayama A et al. Effect of looking at the car that follows in an optimal velocity model of traffic flow[J]. Physical Review E,2002(65):016112.
[26]Ishibashi Y, Fukui M. Temporal variations of traffic flow in the Biham-Middleton-Levine model[J]. Journal of the Physical Society of Japan,1994(63): 2882-2885.
[27]Chung K H, Hui P M, Gu G Q. Two-dimensional traffic flow problems with faulty traffic lights[J]. Physical Review E,1995(51):772-774.
[28]Fukui M, Oikawa H, Ishibashi Y. Flow of cars crossing with unequal velocities in a two-dimensional cellular automaton model[J]. Journal of the Physical Society of Japan,1996(65):2514-2517.
[29]Chopard B, Luthi P O, Queloz P A. Cellular automata model of car traffic in two-dimensional street networks[J]. Journal of Physics A,1996(29):2325-2336.
[30]Wolfram S. Statistical mechanics of cellular automata[J]. Reviews of Modern Physics,1983(55):601-644.
[31]Knospe W, Santen L, Schadschneider A, Schreckenberg M. Towards a realistic microscopic description of highway traffic[J]. Journal of Physics A,2000(33): 477-485.
[32]Jiang R, Wu Q. Spatialltemporal pattcflls at an isolated on-ramp in a new cel-lular automata model based on three-phase trafhc theory [J]. Journal of Physics A,2003,36(381):8197-8213.
[33]Jiang R, Wu Q. First order phase transition from free flow to synchronized now in a cellular automata model[J].,2005(46):581-584.
[34]高坤.从基本图方法到三相交通流——交通流元胞自动机模型理论研究[D].合肥：中国科学技术大学,2008.
[35]孙晓燕.交通流复杂动态特性的元胞自动机模型研究[D].合肥：中国科学技术大学,2010.
[36]孙舵.道路交通流元胞自动机模型中的相变现象研究[D].合肥：中国科学技术大学,2011.
[37]丁中俊.元胞自动机交通流模型中的相变现象和解析研究[D].合肥：中国科学技术大学,2012.
[38]Nagel K, Schreckenberg M. A cellular automaton model for freeway traffic[J]. Journal of Physique I France,1992(2):2221-2229.
[39]Barlovic R, Santen L, Schadschneider A, Schreckenberg M. Metastable states in cellular automata for traffic flow[J]. European Physical Journal B,1998(5): 793-800.
[40]Wahle J, Bazzan A L C, Klugl F, Schreckenberg M. Decision dynamics in a traffic scenario[J]. Physica A,2000(287):669-681.
[41]Lee K, Hui P M, Wang B H, Johnson N F. Effects of announcing global infor-mation in a two-route traffic flow model[J]. Journal of the Physical Society of Japan,2001(70):3507-3510.
[42]Wang W X, Wang B H, Zheng W C, Yin C Y, Zhou T. Advanced information feedback in intelligent traffic systems[J]. Physical Review E,2005(72):066702.
[43]Dong C F, Ma X, Wang G W, Sun X Y, Wang B H. Prediction feedback in intelligent traffic systems[J]. Physica A,2009(388):4651-4657.
[44]Dong C F, Ma X, Wang B H, Sun X Y. Effects of prediction feedback in multi-route intelligent traffic systems[J]. Physica A,2010(389):3274-3281.
[45]Dong C F, Ma X, Wang B H. Advanced information feedback strategy in in-telligent two-route traffic flow systems [J]. Science China-Information Sciences, 2010(53) 11:2265-2271.
[46]Dong C F, Ma X, Wang B H. Effects of vehicle number feedback in multi-route intelligent traffic systems[J]. International Journal of Modern Physics C,2010, 21(8):1081-1093.
[47]Dong C F, Ma X, Wang B H. Weighted congestion coefficient feedback in intelligent transportation systems[J]. Physics Letters A,2010(374):1326-1331.
[48]Dong C F, Ma X. Corresponding Angle Feedback in an innovative weighted transportation system[J]. Physics Letters A,2010(374):2417-2423.
[49]Dong C F, Paty C S. Application of adaptive weights to intelligent information systems:An intelligent transportation system as a case study [J]. Information Sciences,2011(181):5042-5053.
[50]Dong C F, Ma X. Dynamic weight in intelligent transportation systems:A comparison based on two exit scenarios[J]. Physica A,2012(391):2712-2719.
[51]Li M, Ding Z J, Jiang R, Hu M B, Wang B H. Traffi c flow in a Manhattan-like urban system[J]. Journal of Statistical Mechanics,2011:P12001.
[52]Brockfeld E, Barlovic R, Schadschneider A, Schreckenberg M. Optimizing traf-fic lights in a cellular automaton model for city traffic[J]. Rhysical Review E, 2001(64):056132.
[53]Wang W X, Wang B H, Yin C Y et al. Traffic dynamics based on local routing protocol on a scale-free network[J]. Rhysical Review E,2006,73(2):026111.
[54]Wang W X, Yin C Y, Yan G, et al. Integrating local static and dynamic information for routing traffic[J]. Physical Review E,2006,74(1):016101.
[55]Ling X, Hu M B, Jiang R, Wang R L, Cao X B, Wu Q S. Pheromone routing protocol on a scale-free network[J]. Physical Review E,2009(80):066110.
[56]Ling X, Hu M B, Jiang R, Wu Q S. Global dynamic routing for scale-free networks[J]. Physical Review E,2010(81):016113.
[57]Ling X, Hu M B, Du W B, Jiang R, Wu Y H, Wu Q S. Bandwidth allocation strategy for traffic systems of scale-free network[J]. Physics Letter A,2010(374): 4825-4830.
[58]Hu M B, Ling X, Jiang R, Wu Y H, Wu Q S. Dynamic hysteresis phenomena in complex network traffic[J]. Physical Review E,2009(79):047101.
[59]Arenas A, Iacute, Az-Guilera A et al. Communication in Network with Hier-archical Branching[J]. Physical Review Letters,2001,86(14):3196-3199.
[60]Yan G, Zhou T, Hu B. Efficient routing on complex networks[J]. Physical Re-view E,2006,73(4):046108.
[2]戴世强,冯苏苇,顾国庆.交通流动力学：它的内容、方法和意义[J].自然杂志,1997(4)：196-201.
[3]徐东云,张雷,兰荣娟.城市交通拥堵的背景变换分析[J].城市问题,2009(3)：46-49.
[4]Helbing D, Hennecke A, Shvetsov V, et al. MASTER:macroscopic traffic sim-ulation based on a gas-kinetic, non-local traffic model[J]. Transportation Re-search Part B,2001(35):183-211.
[5]贾述评,房筱莉.日本智能交通系统的发展[J].吉林交通科技,2006(3)：67-68.
[6]陈旭梅,于雷,郭继孚,等.美国智能交通系统ITS的近期发展综述[J].中外公路,2003,23(2)：9-12.
[7]中辛.美、日、欧洲ITS研究开发项目[J].东北汽车运输,2000(4)：24-26.
[8]范建军,缪立新,史其信.ITS在亚太地区的发展[J].中国物流与采购,2003(3)：24-25.
[9]吴俊.智能交通系统中的信息处理关键技术研究[D].天津：天津大学,2006.
[10]Kuospe W. Single-vehicle data of highway Ixaffic:Microscopic description of traffic phases. Physical Review E,2002(65):056133.
[11]贾斌,高自友,李克平,李新刚.基于元宝自动机的交通系统建模与模拟[M].北京：科学出版社,2007.
[12]Lighthill M J, Whitham G B. On kinematic waves:II. a theory of traffic flow on long crowded roads [J]. Proceedings of the Royal Society A,1955(229):317-345.
[13]Richards P I. Shock waves on the highway[J]. Operations Research,1956(4): 42-51.
[14]Payne H J. Models of freeway traffic and control[M]. La Jolla:Simulation Council.1971.
[15]Jiang R, Wu Q S, Zhu Z J. A new continuum model for traffic flow and nu-merical tests[J]. Transporation Research Part B,2002(36):405-419.
[16]Siebel F, Mauser W. On the fundmental diagram of traffic flow[J]. SIAM Jour-nal on Applied Mathematics,2006(66):1150-1162.
[17]Siebel F, Mauser W. Synchronized flow and wide moving jams from balanced vehicular traffic[J]. Physical Review E,2006(73):066108.
[18]Paveri-Fontana S L. On Boltzmann-like treatments for traffic flow. A critical review of the basic model and an alternative proposal for dilute traffic analysis. Tranpration Research,1975(9):225-235.
[19]Helbing D. Derivation and empirical validation of a refined traffic flow model[J]. Physica A,1996(233):253-282.
[20]Nagatani T. Kinetic segregation in a multilane highway traffic flow[J]. Physica A,1997(237):67-74.
[21]Treiber M, Hennecke A, Helbing D. Derivation, properties and simulation of a gas-kinetic-based nonlocal traffic model[J]. Physical Review E,1999(53):239-253.
[22]Hoogendoorn S P, Bovy P H L. Generic gas-kinetic traffic systems modeling with applications to vehicular traffic flow[J]. Transprotation Research Part B, 2001(35):317-336.
[23]Reuschel A. Vehicle movements in a platoon [J]. Oesterreichisches Ingenieur-Archir,1950(4):193-215.
[24]Helbing D, Tilch B. Generalized force model of traffic dynamics[J]. Physical Review E,1998(58):133-138.
[25]Nagayama A et al. Effect of looking at the car that follows in an optimal velocity model of traffic flow[J]. Physical Review E,2002(65):016112.
[26]Ishibashi Y, Fukui M. Temporal variations of traffic flow in the Biham-Middleton-Levine model[J]. Journal of the Physical Society of Japan,1994(63): 2882-2885.
[27]Chung K H, Hui P M, Gu G Q. Two-dimensional traffic flow problems with faulty traffic lights[J]. Physical Review E,1995(51):772-774.
[28]Fukui M, Oikawa H, Ishibashi Y. Flow of cars crossing with unequal velocities in a two-dimensional cellular automaton model[J]. Journal of the Physical Society of Japan,1996(65):2514-2517.
[29]Chopard B, Luthi P O, Queloz P A. Cellular automata model of car traffic in two-dimensional street networks[J]. Journal of Physics A,1996(29):2325-2336.
[30]Wolfram S. Statistical mechanics of cellular automata[J]. Reviews of Modern Physics,1983(55):601-644.
[31]Knospe W, Santen L, Schadschneider A, Schreckenberg M. Towards a realistic microscopic description of highway traffic[J]. Journal of Physics A,2000(33): 477-485.
[32]Jiang R, Wu Q. Spatialltemporal pattcflls at an isolated on-ramp in a new cel-lular automata model based on three-phase trafhc theory [J]. Journal of Physics A,2003,36(381):8197-8213.
[33]Jiang R, Wu Q. First order phase transition from free flow to synchronized now in a cellular automata model[J].,2005(46):581-584.
[34]高坤.从基本图方法到三相交通流——交通流元胞自动机模型理论研究[D].合肥：中国科学技术大学,2008.
[35]孙晓燕.交通流复杂动态特性的元胞自动机模型研究[D].合肥：中国科学技术大学,2010.
[36]孙舵.道路交通流元胞自动机模型中的相变现象研究[D].合肥：中国科学技术大学,2011.
[37]丁中俊.元胞自动机交通流模型中的相变现象和解析研究[D].合肥：中国科学技术大学,2012.
[38]Nagel K, Schreckenberg M. A cellular automaton model for freeway traffic[J]. Journal of Physique I France,1992(2):2221-2229.
[39]Barlovic R, Santen L, Schadschneider A, Schreckenberg M. Metastable states in cellular automata for traffic flow[J]. European Physical Journal B,1998(5): 793-800.
[40]Wahle J, Bazzan A L C, Klugl F, Schreckenberg M. Decision dynamics in a traffic scenario[J]. Physica A,2000(287):669-681.
[41]Lee K, Hui P M, Wang B H, Johnson N F. Effects of announcing global infor-mation in a two-route traffic flow model[J]. Journal of the Physical Society of Japan,2001(70):3507-3510.
[42]Wang W X, Wang B H, Zheng W C, Yin C Y, Zhou T. Advanced information feedback in intelligent traffic systems[J]. Physical Review E,2005(72):066702.
[43]Dong C F, Ma X, Wang G W, Sun X Y, Wang B H. Prediction feedback in intelligent traffic systems[J]. Physica A,2009(388):4651-4657.
[44]Dong C F, Ma X, Wang B H, Sun X Y. Effects of prediction feedback in multi-route intelligent traffic systems[J]. Physica A,2010(389):3274-3281.
[45]Dong C F, Ma X, Wang B H. Advanced information feedback strategy in in-telligent two-route traffic flow systems [J]. Science China-Information Sciences, 2010(53) 11:2265-2271.
[46]Dong C F, Ma X, Wang B H. Effects of vehicle number feedback in multi-route intelligent traffic systems[J]. International Journal of Modern Physics C,2010, 21(8):1081-1093.
[47]Dong C F, Ma X, Wang B H. Weighted congestion coefficient feedback in intelligent transportation systems[J]. Physics Letters A,2010(374):1326-1331.
[48]Dong C F, Ma X. Corresponding Angle Feedback in an innovative weighted transportation system[J]. Physics Letters A,2010(374):2417-2423.
[49]Dong C F, Paty C S. Application of adaptive weights to intelligent information systems:An intelligent transportation system as a case study [J]. Information Sciences,2011(181):5042-5053.
[50]Dong C F, Ma X. Dynamic weight in intelligent transportation systems:A comparison based on two exit scenarios[J]. Physica A,2012(391):2712-2719.
[51]Li M, Ding Z J, Jiang R, Hu M B, Wang B H. Traffi c flow in a Manhattan-like urban system[J]. Journal of Statistical Mechanics,2011:P12001.
[52]Brockfeld E, Barlovic R, Schadschneider A, Schreckenberg M. Optimizing traf-fic lights in a cellular automaton model for city traffic[J]. Rhysical Review E, 2001(64):056132.
[53]Wang W X, Wang B H, Yin C Y et al. Traffic dynamics based on local routing protocol on a scale-free network[J]. Rhysical Review E,2006,73(2):026111.
[54]Wang W X, Yin C Y, Yan G, et al. Integrating local static and dynamic information for routing traffic[J]. Physical Review E,2006,74(1):016101.
[55]Ling X, Hu M B, Jiang R, Wang R L, Cao X B, Wu Q S. Pheromone routing protocol on a scale-free network[J]. Physical Review E,2009(80):066110.
[56]Ling X, Hu M B, Jiang R, Wu Q S. Global dynamic routing for scale-free networks[J]. Physical Review E,2010(81):016113.
[57]Ling X, Hu M B, Du W B, Jiang R, Wu Y H, Wu Q S. Bandwidth allocation strategy for traffic systems of scale-free network[J]. Physics Letter A,2010(374): 4825-4830.
[58]Hu M B, Ling X, Jiang R, Wu Y H, Wu Q S. Dynamic hysteresis phenomena in complex network traffic[J]. Physical Review E,2009(79):047101.
[59]Arenas A, Iacute, Az-Guilera A et al. Communication in Network with Hier-archical Branching[J]. Physical Review Letters,2001,86(14):3196-3199.
[60]Yan G, Zhou T, Hu B. Efficient routing on complex networks[J]. Physical Re-view E,2006,73(4):046108.