Urban Road Traffic Signal Control And Optimization Algorithms

Traffic is necessary for people to live and develop. With the rapid development of urban and prevalence of cars, urban traffic congestion and accidents, noise and pollution that are resulted from traffic congestion have become serious social problems increasingly. Traffic has become universal difficult problem that puzzle the governments around the world. Traffic engineering practice indicates that building road properly is needed, at the same time, it is essential ways of solving traffic problems to regulate and control traffic flow through software properly and scientifically, to exert potential of current traffic networks fully, to make traffic flow move in an order, and to develop Intelligent Transportation System. Urban traffic is an important branch of transportation system. Traffic control, especially control of traffic signal light is a dominating part of urban traffic research.Control and optimization of traffic signal has become a primary branch of urban traffic control research since traffic signal light was used in the early 20th century, and now it is still one of the hot topics in the field of urban traffic control. Traffic signal control is carrying out in the intersection, which can separate incompatible traffic flows on time and space, make transportation safety of the intersection get guarantee and improve ability of urban traffic networks fully.Control strategy research that based on traffic model can provide a lot of beneficial new ideas and means for traffic control practice. This paper set up macroscopic traffic flows model of trunk control and district control respectively, deduce corresponding property indexes, predict traffic flow through mathematic model, and combine fuzzy control to realize traffic control. In addition, optimization methods of traffic control are studied inthis paper, the improved Immune-Genetic Algorithm is used to realize optimization of traffic control strategy and perfect result is acquired. The main creative achievements are as follows:1 Based on green wave control and induction control, a two-level fuzzy control algorithm about double-intersection is proposed. Lengths of vehicle queues in each phase are used as the basic inputs of the first-level fuzzy controller, and temporary green prolong time is reasoned via the first-level fuzzy controller. Temporary green prolong time and surplus time of current term are used as the basic inputs of the second-level fuzzy controller, and actual green time is reasoned via the second-level fuzzy controller.2 During the course of optimization in two-level fuzzy control algorithm, chaos optimization algorithm is applied in designing membership function of fuzzy controllers. Comparing with methods of manual adjust and data from experience, membership functions are scientific, so that the result is close to optimal value.3 Conventional Immune-Genetic Algorithm is improved in two ways.l) Memory library is extended. 2) New generation antibody colony is more reasonable. Then it is used to optimize property indexes of trunk control and district control.4 According to urban traffic trunk control, macroscopical traffic flow model is set up. This paper brings forward a kind of method of predicting traffic flow in which traffic flow at the moment before different phase difference and Correcting Coefficient Method are both used to predict traffic flow. At first, uniform term of the trunk is established by Perfect Period formula which is proposed by Webster. Follow this, weighted total vehicle number that is detained in each intersection of the trunk is computed to acquire optimal property index, then traffic flow forecast is run. At last the improved Immune-Genetic Algorithm is adopted to optimize multi-object. The simulation resultstestify that relatively satisfactory results is obtained5 Based on urban traffic district control, four-phase mathematics model of district control is set up. At first, uniform period of the district is established through fuzzy controller. After that, weighted total vehicle number that is detained in each intersection of the district is computed to acquire optimal property index. Then, traffic flow at the moment before different phase difference is used to predict traffic flow of each intersection at the moment. In the end, the improved Immune-Genetic Algorithm is adopted to optimize multi-object so as to realize optimization and control of urban district traffic.Detailed research is made from traffic signal control indouble-intersection to those in trunk and district control in this paper.Aiming at different traffic problems, relevant control models andoptimization methods are established. Through computer simulation,satisfying achievements of the initial stage are obtained.