Urban Traffic Signal Control And Its Application

With the increasing demand of traffic, urban traffic congestion is more and more serious,which in turn triggeres a series of problems, such as vehicle delay, traffic accidents, energyconsumption and environmental pollution. These issues mentioned above can’t be resolvedfundamentally through expanding or building some new transportation infrastructures. Hence,adopting reasonable traffic control strategies becomes the main means, among which thebasic method is to establish efficient traffic control system and obtain scientific and rationaltraffic signal control scheme. Once some emergency events have happened, the transfer ofpeople from impacted areas to emergency shelters or medical assistance organizations asrapidly as possible plays an important role in reducing the negative impact of these events.However, widespread traffic congestion existed in most of the large and medium-sized citiesmake it difficult for emergency vehicles to reach their destination quickly and safely. So, it isvery essential to study traffic signal control method in emergency management.Comparing with traditional methods, Petri net has unparalleled advantages in describingconcurrency, resource sharing, conflicts and dynamics of urban transportation system. Somescholars have led Petri nets into the field of traffic signal control. But, there are still a lot ofplaces that need to be further refined and improved. Although some researchers andinstitutions have pointed out that a good signal timing plan can increase capacities of the roadon evacuation routes, and has a potentially important impact on evacuation behavior, littleformal research has focused on the problem of control traffic signals so as to assist evacuation.Based on this, the traffic signal control problem based on Petri nets, and emergency vehicletraffic signal control problem under unexpected events are mainly studied in this dissertation.Draw lessons from the results of previous studies, research has been carried out in thefollowing areas:(1) On the basis of analysis and summary of the relevant literatures at home and abroadof traffic signal control based on Petri nets, as well as traffic signal control in emergencymanagement, the shortcomings of them are pointed out and taken as the main researchcontents of this dissertation.(2) The inductive control and the optimal control of traffic signal at isolated intersectionare discussed. A hybrid Petri net model of isolated intersection in urban traffic network isestablished, and a Petri net model that control green time extension for each phase byinhibitor arcs according to the number of vehicles on the lanes of current phase is also given.In the given hybrid Petri net model, if a transition has two input places, the enable degree ofthis transition is decided by the markings of the two places, while the changes of the markingsof input places are also depending on the enable degree of this transition. Aiming at thiscomplex relationship between places and transition, the marking change law of this kind of places is analyzed. To optimize the green time of each phase, the formula of total vehicleresidence time is presented. By adopting discretization method, a mathematical model forminimizing average vehicle residence time is established. In the end, a simulation and resultsanalysis are executed.(3) To study the signal coordinated control problem of urban arterial street, according tothe modular modeling idea of Petri net, traffic signal display module and phase transitionmodule based on timed Petri net are designed, and the working principle of these two modulesare elaborated. The traffic signal control system given here is a phase sequence-changeablecontrol system. It is composed of three sub systems, say monitoring system, discriminatingsystem and current phase selecting system, and concrete control steps are presented. In orderto describe traffic flow on the link between two intersections accurately, each link is dividedinto three sub-sections. Based on this division, the determination method of smooth degree ofdownstream section is given by considering the relationship among various parameters ofcontinuous Petri net. Fuzzy control theory and fuzzy Petri net method are adopted todetermine current phase and its optimal green time duration. Finally, a simulation is carriedout.(4) Under unexpected events, emergency vehicles coming from different directionsarrive at the same intersection simultaneously and compete for green light phase. In this case,to reduce delay of emergency vehicle and improve efficiency of evacuation and rescue, trafficsignal preemption control is very essential. Hence, the coordinated signal priority control ofconflicting traffic flow at an isolated intersection with multi phases is studied. At a four-phasecontrolled intersection, taking the minimum delays of all vehicles at isolated intersection asthe control target, the priority of each phase is obtained by considering expected futurewaiting time and emergency vehicle volume rate comprehensively. Fuzzy Petri net approachis adopted to give the most suitable green time according to the queue and the emergencyvehicle volume rate of current phase and the next phase, and a concrete signal priority controlstrategy is presented.(5) A signal preemption control method based on route is proposed to reduce time delayof emergency vehicles. On the premise of having selected a concrete evacuation route,considering the location of emergency vehicle detector and some security intervalrequirements, the calculation methods of the earliest possible start time and the latest possiblestart time of green light at each intersection in the route are given. In order to ensure the safeand rapid transit of emergency vehicle, and reduce its impact on other society vehicles at thesame time, a multi-objective programming model, whose objectives are the minimal residencetime of emergency vehicle at all intersections and the maximal passing vehicle numbers, ispresented. A particle swarm algorithm for solving the given model is designed. Finally, Paretooptimal sets corresponding to different arriving time of emergency vehicle are obtained.