Research On Dynamic Timing Optimization Algorithm For Urban Traffic Lights

The number of vehicles in the city has been increasing and the existing trafficresources are limited which causes the more problems of urban traffic congestion.Theintersection is the key point in the urban traffic network.The reasonable guidance andcontrol of traffic flow in the intersection have become the main ways to improve theurban road traffic capacity.And nowadays constant timing control modes,manualcontrol modes as well as single-objective optimization methods are unable to meet thechanging requirements of urban traffic development in real-time,efficiency,integrity andso on.In this paper,to overcome the shortcomings and deficiencies of existing controlprograms,related researches are conducted.Firstly,an algorithm of dynamic timing based on multi-objective optimization forurban single intersection traffic lights is proposed.That adopts the method ofmulti-objective optimization to calculate signal cycle according to traffic data,and thenthe period of green light is increased within the real-time in order to achieve dynamictiming optimization by means of fuzzy control theory.The queue length is used as thefuzzy control variable.Compared with traditional timing means,the experimental resultsof the two phase single intersection show that the algorithm combined the dynamicdistribution strategy with multi-objective optimization method can have a goodadvantage over reducing vehicle delay and number of stops and so on.Secondly,themultiple intersections on the trunk road are studied,and then a dynamic coordinationalgorithm based on number methods is proposed.The number method is used to get thedifferences between their initial phases.With the real-time traffic volume,the cycle andthe green time both end with the trunk road are dynamically adjusted.And then thesystem cycle and the differences of the phases are dynamically regulated.The momentthe program is given with each intersection.And a multi-objective optimization modelfor mixed traffic environments is established to adjust to the mixed traffic volume.Thealgorithm is verified by real-time traffic simulation.The results show that the average delay and number of stops of the system has been significantly improved,and the trafficcapacity has been improved greatly.Finally,the exploratory study about regional coordination for urban traffic controlis conducted,and the algorithm that a dynamic regional division of the sub-region isstarted and a coordinated control strategy based on priority of the main road among theintersections within the sub-region is executed.The regional signal coordination controlis decomposed into two parts,the complexity is effectively reduced and the trafficcontrol capacity of the entire region is improved.A summary of the full paper of the work is given and the still deficiencies arepointed out,and the prospective of further development is described in the end.