城市交通系统控制信号优化调度研究
|
摘要
本文从国内外智能交通系统的研究现状出发，重点研究交通信号的优化调度课题。研究的目标是在交通数据采集的基础上利用专家系统、神经网络、遗传算法等技术分析交通动态特征、形成全局优化调度指令，并利用模糊控制的方法实现交叉口的信号灯控制。文章针对现有城市交通信号控制系统的不足之处，提出了一种城市交通信号控制系统的设计方案。该方案采用模糊控制方法对交叉口交通信号进行控制，提出以当前相、后继相的车辆等待长度决定相位信号配时，并针对模糊神经网络提出FN-BP算法以改进传统BP算法收敛速度慢、处理精度不高的困扰。同时还详细研究了基于遗传算法的模糊神经网络智能控制方法，提出对城市交通系统控制信号进行全局优化调度的算法。在前述算法的指导下，依据面向对象软件开发思想，用Visual C++开发城市交通信号控制系统仿真系统S-UTCS。并通过对示例路网进行仿真，得出了合理的统计结果，证实设计的合理性和实现的有效性。实践证明，与传统BP算法相比，该算法具有收敛性能好、函数逼近精度高的优点。通过遗传算法的全局优化，可以在动态变化的交通环境下找到鲁棒性较好的控制参数，实现全局优化的面控。
According to the recent research on intelligent traffic control system at home and aboard, this dissertation lays a strong emphasis on the study of the traffic signal's global optimization control. Depending on getting the crossway transportation data, the intention of the study is to make use of the expert system, neural network and genetic algorithm to analyze the dynamic characteristic of transportation and to form the global optimization control order, and to realize the local signal lamp control by the method of utilization fuzzy control. A sort of project of an urban traffic signal control system is presented in the dissertation based on the weakness of intelligent traffic control system. A traffic signal control method using fuzzy logic controller in urban road intersection is proposed in the project, which uses current phase and next phase vehicles waiting queue to determine each phase time. And FN-BP algorithm is presented in allusion to fuzzy neural network in the dissertation, which improves the constringency speed and precision of the conventional BP algorithm. The dissertation also lays the emphasis on studying an intelligent transportation method to fuzzy neural network base on genetic algorithm, which can realize traffic signal optimization. According to the object-oriented software development concept and under the direction of the before-mentioned algorithm, Signal - Urban Traffic Control System is realized by VC++, which approves the rationality of system design and the validity of realization for getting the reasonable statistical result. As pointed out by the experimented results, this algorithm is more efficient than the conventional BP algorithm on the aspects of convergence performance and the ability of function approximation. Better control parameter of the highly robust capability can be found, which uses global optimization control based on the genetic algorithm under the dynamic transportation circumstances, and realizes the zone control globally.
According to the recent research on intelligent traffic control system at home and aboard, this dissertation lays a strong emphasis on the study of the traffic signal's global optimization control. Depending on getting the crossway transportation data, the intention of the study is to make use of the expert system, neural network and genetic algorithm to analyze the dynamic characteristic of transportation and to form the global optimization control order, and to realize the local signal lamp control by the method of utilization fuzzy control. A sort of project of an urban traffic signal control system is presented in the dissertation based on the weakness of intelligent traffic control system. A traffic signal control method using fuzzy logic controller in urban road intersection is proposed in the project, which uses current phase and next phase vehicles waiting queue to determine each phase time. And FN-BP algorithm is presented in allusion to fuzzy neural network in the dissertation, which improves the constringency speed and precision of the conventional BP algorithm. The dissertation also lays the emphasis on studying an intelligent transportation method to fuzzy neural network base on genetic algorithm, which can realize traffic signal optimization. According to the object-oriented software development concept and under the direction of the before-mentioned algorithm, Signal - Urban Traffic Control System is realized by VC++, which approves the rationality of system design and the validity of realization for getting the reasonable statistical result. As pointed out by the experimented results, this algorithm is more efficient than the conventional BP algorithm on the aspects of convergence performance and the ability of function approximation. Better control parameter of the highly robust capability can be found, which uses global optimization control based on the genetic algorithm under the dynamic transportation circumstances, and realizes the zone control globally.
引文
[1] Richardon B, Rodriquez D. ITS in developing countries: strategic opportunities. In ITS Quartely, 1997, 5(3): 7～15
[2] 陈森发，毛岚．城市交通信号灯模糊线控制及其仿真．系统仿真学报，2000，12(6)：698～770
[3] Bell M G H, Shield C M and Anderson J M. Assignment in the integration of urban traffic control and dynamic route guidance, in urban traffic networks dynamic flow modeling and control. Gurtner N H and Improta G. (Eds), Springer-Verlag, 1995:39～57
[4] West Germany. Intersection Without Signals. Proceeding of an International Workshop 16-18 March, 1998:30～40
[5] Zou Zhi jun. A study of capacity of major/minor priority T-intersection by means of computer simulation. China Journal of Highway and Transport, 2000, 13(3): 101～105
[6] Guguang HE, Noth G. Urban Traffic Control System - A General Analysis from the View of Control Theory. Transportation systems, 1997, 2:503～506
[7] 张锦，高世廉，蒲云．中国城市道路交通控制系统发展研究．西南交通大学学报，1997，32(1)：44～49
[8] C. P. Pappis, E. H. Mamdani. A Fuzzy Logic Controller for a Traffic Junction. IEEE Transaction on System, Man, and Cybernetics, 1977, 7(10):707～717
[9] Jee-Hyong Lee, Hyung Lee-Kwang. Distributed and Cooperative Fuzzy Controllers for Traffic Intersections Group. IEEE Transaction on System, Man, and Cybernetics, 1999, 29(2): 263～271
[10] Jee-Hyong Lee, Keon-Myung Lee and Hyong LeeKwang. Fuzzy Controller for Intersection Group[A] .Proceedings of 1995 IEEE/LAS Intersectional conference on Industrial Automation and Control: Emerging Technologies[C] . Piscataway, NJ, USA, 1995: 376～382
[11] K Wood. Urban traffic control, system review. TRL, 2000
[12] 黄卫，陈里得．智能运输系统(ITS)概论．北京：人民交通出版社，2001．6～110
[13] Rithchie S. G. and Cheu, R. L. Neural Network Models for Automated Detection of Non-Recurring Congestions. PATH Research Report UCB-ITS-PRR-93-5,1993a
[14] Samuel C, H. J. Panyne. Improved Freeway Incident Detection Algorithms. Public Roads, 1997, (6): 32～40
[15] 刘伟铭．高速公路系统控制方法．北京：人民交通出版社，1998．26～68
[16] 王亦兵．城市高速公路交通控制综述．自动化学报，1998，24(4)：484～496
[17] 李秀平，刘志勇等．多传感器信息融合的智能交通控制系统研究．信息与控制，2001，30(5)：460～464
[18] 徐吉万，徐冬玲．城市交通的计算机控制和管理．北京：测绘出版社，1988．46～78
[19] 荆便顺．道路交通控制工程．北京：人民交通出版社，1995．36～75
[20] 张殿业．论ITS的结构体系．西南交通大学学报，2000，35(2)：208～211
[21] 阙大顺．车辆检测器及其发展研究．交通与计算机，1999，17(2)：38～41
[22] 董国良等译．(日)社团法人：交通工学研究会编．智能交通系统．北京：人民交通出版社，2000．23～65
[23] Fretz Busch, Martin Fellendort. Automatic Incident Detection on Motorways. Traffic Engineering & Control, 1990, (4):221～227
[24] Cheu, R. L., Srinivasan. D. Development of new Incident Detection Methodologies for Expressways. Research Report GR6453, 1998b
[25] 徐冬玲等．交通系统的模糊控制及其神经网络的实现．信息与控制，1992，21(2)：74～78
[26] Xin Jin, Cheu, R. L. Development and Adaptation of Constructive Probabilistic Neural Network Incident-Detection. Transp Res, C, 2002, 10:121～147
[27] Hsiao, C. H., Lin, C. T., Cassidy, M. Application of Fuzzy Logic and Neural Networks to Automatically Detect Freeway Traffic Incidents. Journal of Transportation Engineering. ASCE, 1994, 120(5):753～771
[28] 陈洪，陈森发．单路口交通实时模糊控制的一种方法．信息与控制，1997，26(3)：227～233
[29] Takahashi. Unknown input observers for uncertain systems: a unifying approach. European Jour. Of Contr, 1999, 5(2):261～275
[30] 黄辉先，史忠科．城市单交叉路口交通流实时遗产算法优化控制．系统工程理论与实践，2001，13(3)：102～106
[31] 徐建闽，舒宁，尹宏宾．新型模糊控制算法在交叉路口信号控制中的应用．华南理工大学学报，2000，28(6)：1～5
[32] M. Hou, P. C. Muller. Design of observes for linear systems with unknown inputs. IEEE Trans. Automa. Contr. 1992, AC-37,871～875
[33] Chia-Chi Tsui. A new design approach to unknown input observers. IEEE Trans. Automa. Contr, 1996, 37(3): 464～468
[34] 张乃尧，阎平凡．神经网络与模糊控制．北京：清华大学出版社，1998．36～110
[35] 常虹，何丕廉．神经网络与模糊技术的结合与发展．计算机应用研究，2001，18(5)：4～6
[36] 焦李成．神经网络计算．西安：西安电子科技大学出版社，1993．26～45
[37] 毛国君，宋广军，杨名生．FNN上的反向传播学习算法．计算机应用与软件，1998，15(4)：34～38
[38] 阿孜古丽，牙会甫，左孝凌．一种改进的模糊神经网络学习算法．计算机应用与软件，2000，17(8)：21～31
[39] 武妍，施鸿宝．一种模糊规则动态调整BP算法中参数的方法．计算机研究与发展，1998，35(8)：689～693
[40] RalfSalomon, JLeovanHemmen. Acceleratingback-propaga-tionthroughdynamic self-adaption. NeuralNetworks, 1996,9(4):589～601
[41] Persaud B. N., Rail F. L., Hall L. M. Congestion Identification Aspects of the McMaster Incident Detection Algorithm. Transportation Research Record 1990, 1287, 157～175
[42] Payne, H. J., Helfenbein, E. D., Knobel, H. C. Development and Testing of Incident Detection Algorithms. Report NO. FHWA-RD-76-20. vol. 2.1976
[43] Specht, D. F. Probabilistic Neural Networks. Neural Networks, 1990: 109～118
[44] Ishak, S. S., AI-Deek, H. M. Freeway Incident Detection using Fuzzy ART. Proceedings of the Fifth International Conference on Application of Advanced Technologies in Transportation. 1998, 59～66
[45] 王永庆．人工智能原理与方法．西安：西安交通大学出版社，1998．46～60
[46] 李秀平，刘志勇，吴今培．平面交叉路口的神经网络自学习控制方案．信息与控制，2001，30(1)：76～79
[47] 王士同．神经模糊系统及其应用．北京：北京航空航天大学出版社，1999．16～88
[2] 陈森发，毛岚．城市交通信号灯模糊线控制及其仿真．系统仿真学报，2000，12(6)：698～770
[3] Bell M G H, Shield C M and Anderson J M. Assignment in the integration of urban traffic control and dynamic route guidance, in urban traffic networks dynamic flow modeling and control. Gurtner N H and Improta G. (Eds), Springer-Verlag, 1995:39～57
[4] West Germany. Intersection Without Signals. Proceeding of an International Workshop 16-18 March, 1998:30～40
[5] Zou Zhi jun. A study of capacity of major/minor priority T-intersection by means of computer simulation. China Journal of Highway and Transport, 2000, 13(3): 101～105
[6] Guguang HE, Noth G. Urban Traffic Control System - A General Analysis from the View of Control Theory. Transportation systems, 1997, 2:503～506
[7] 张锦，高世廉，蒲云．中国城市道路交通控制系统发展研究．西南交通大学学报，1997，32(1)：44～49
[8] C. P. Pappis, E. H. Mamdani. A Fuzzy Logic Controller for a Traffic Junction. IEEE Transaction on System, Man, and Cybernetics, 1977, 7(10):707～717
[9] Jee-Hyong Lee, Hyung Lee-Kwang. Distributed and Cooperative Fuzzy Controllers for Traffic Intersections Group. IEEE Transaction on System, Man, and Cybernetics, 1999, 29(2): 263～271
[10] Jee-Hyong Lee, Keon-Myung Lee and Hyong LeeKwang. Fuzzy Controller for Intersection Group[A] .Proceedings of 1995 IEEE/LAS Intersectional conference on Industrial Automation and Control: Emerging Technologies[C] . Piscataway, NJ, USA, 1995: 376～382
[11] K Wood. Urban traffic control, system review. TRL, 2000
[12] 黄卫，陈里得．智能运输系统(ITS)概论．北京：人民交通出版社，2001．6～110
[13] Rithchie S. G. and Cheu, R. L. Neural Network Models for Automated Detection of Non-Recurring Congestions. PATH Research Report UCB-ITS-PRR-93-5,1993a
[14] Samuel C, H. J. Panyne. Improved Freeway Incident Detection Algorithms. Public Roads, 1997, (6): 32～40
[15] 刘伟铭．高速公路系统控制方法．北京：人民交通出版社，1998．26～68
[16] 王亦兵．城市高速公路交通控制综述．自动化学报，1998，24(4)：484～496
[17] 李秀平，刘志勇等．多传感器信息融合的智能交通控制系统研究．信息与控制，2001，30(5)：460～464
[18] 徐吉万，徐冬玲．城市交通的计算机控制和管理．北京：测绘出版社，1988．46～78
[19] 荆便顺．道路交通控制工程．北京：人民交通出版社，1995．36～75
[20] 张殿业．论ITS的结构体系．西南交通大学学报，2000，35(2)：208～211
[21] 阙大顺．车辆检测器及其发展研究．交通与计算机，1999，17(2)：38～41
[22] 董国良等译．(日)社团法人：交通工学研究会编．智能交通系统．北京：人民交通出版社，2000．23～65
[23] Fretz Busch, Martin Fellendort. Automatic Incident Detection on Motorways. Traffic Engineering & Control, 1990, (4):221～227
[24] Cheu, R. L., Srinivasan. D. Development of new Incident Detection Methodologies for Expressways. Research Report GR6453, 1998b
[25] 徐冬玲等．交通系统的模糊控制及其神经网络的实现．信息与控制，1992，21(2)：74～78
[26] Xin Jin, Cheu, R. L. Development and Adaptation of Constructive Probabilistic Neural Network Incident-Detection. Transp Res, C, 2002, 10:121～147
[27] Hsiao, C. H., Lin, C. T., Cassidy, M. Application of Fuzzy Logic and Neural Networks to Automatically Detect Freeway Traffic Incidents. Journal of Transportation Engineering. ASCE, 1994, 120(5):753～771
[28] 陈洪，陈森发．单路口交通实时模糊控制的一种方法．信息与控制，1997，26(3)：227～233
[29] Takahashi. Unknown input observers for uncertain systems: a unifying approach. European Jour. Of Contr, 1999, 5(2):261～275
[30] 黄辉先，史忠科．城市单交叉路口交通流实时遗产算法优化控制．系统工程理论与实践，2001，13(3)：102～106
[31] 徐建闽，舒宁，尹宏宾．新型模糊控制算法在交叉路口信号控制中的应用．华南理工大学学报，2000，28(6)：1～5
[32] M. Hou, P. C. Muller. Design of observes for linear systems with unknown inputs. IEEE Trans. Automa. Contr. 1992, AC-37,871～875
[33] Chia-Chi Tsui. A new design approach to unknown input observers. IEEE Trans. Automa. Contr, 1996, 37(3): 464～468
[34] 张乃尧，阎平凡．神经网络与模糊控制．北京：清华大学出版社，1998．36～110
[35] 常虹，何丕廉．神经网络与模糊技术的结合与发展．计算机应用研究，2001，18(5)：4～6
[36] 焦李成．神经网络计算．西安：西安电子科技大学出版社，1993．26～45
[37] 毛国君，宋广军，杨名生．FNN上的反向传播学习算法．计算机应用与软件，1998，15(4)：34～38
[38] 阿孜古丽，牙会甫，左孝凌．一种改进的模糊神经网络学习算法．计算机应用与软件，2000，17(8)：21～31
[39] 武妍，施鸿宝．一种模糊规则动态调整BP算法中参数的方法．计算机研究与发展，1998，35(8)：689～693
[40] RalfSalomon, JLeovanHemmen. Acceleratingback-propaga-tionthroughdynamic self-adaption. NeuralNetworks, 1996,9(4):589～601
[41] Persaud B. N., Rail F. L., Hall L. M. Congestion Identification Aspects of the McMaster Incident Detection Algorithm. Transportation Research Record 1990, 1287, 157～175
[42] Payne, H. J., Helfenbein, E. D., Knobel, H. C. Development and Testing of Incident Detection Algorithms. Report NO. FHWA-RD-76-20. vol. 2.1976
[43] Specht, D. F. Probabilistic Neural Networks. Neural Networks, 1990: 109～118
[44] Ishak, S. S., AI-Deek, H. M. Freeway Incident Detection using Fuzzy ART. Proceedings of the Fifth International Conference on Application of Advanced Technologies in Transportation. 1998, 59～66
[45] 王永庆．人工智能原理与方法．西安：西安交通大学出版社，1998．46～60
[46] 李秀平，刘志勇，吴今培．平面交叉路口的神经网络自学习控制方案．信息与控制，2001，30(1)：76～79
[47] 王士同．神经模糊系统及其应用．北京：北京航空航天大学出版社，1999．16～88
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |