Research On The Control Algorithm Of Virtual Intelligent Traffic Lights In The Internet Of Vehicles

The popularization of Internet of Vehicles has a huge influence on the traffic,one of them is traffic light,it's the V2 M communication in Internet of Vehicles that makes traffic data could be collected timely,what's more,the data is also used to calculate traffic lights phase position.Shifting traffic lights from the outside to the inside of the car eliminates.The IoV enhance the real-time control capability of physical traffic light and solve the inaccuracy of traffic lights phase position design and the problem about extremely high maintenance costs.The exsiting algorithms of virtual traffic lights collect information using direct communication,the designing of vehicle priority in all lanes only uses one parameter,that is,the moment of reaching intersection,transportation power would be turned over to the next intersection only according to the whole group of vehicles have passed.However,the algorithm above is neither feasible nor achieving the best throughput capacity.In order to adapt common traffic and increase intersection throughput capacity,The main contributions of this article are as follows:1.Based on the data collection algorithm of the queuing chain,design a three-lane queuing chain model;First of all,a three-lane queuing chain model was designed,we consider a single vehicle as one node in the chain,regard vehicle which enters a new lane as a new node which is added into the chain,we delete one node in the chain when the corresponding vehicle leaves the lane.We consider the first node of the chain as aggregation node vehicle,this node collects and manages the data of the whole chain and calculates traffic light phase position by using the data above in intersections.2.Put forward an independent intersection virtual traffic light phase algorithm;We propose a phase position algorithm named single virtual traffic lights(SVTL)which takes advantage of three parameters including waiting time of every single vehicle,length of the chain and number of vehicles in the chain to make the passing priority,so that vehicle queue which has the highest priority gains passing priority.Based on the confliction vector table in intersection,we find a lane which doesn't conflict with vehicle queue which has the highest priority,at the same time,both lanes turn into green time,as a result,the algorithm could increase throughput capacity.The algorithm eliminates phenomena such as vehicles wait around in intersection,decreases waiting time and avoids unnecessary parking case.3.Propose virtual traffic light phase algorithm based on multi-segment combination.Multi-section virtual traffic lights,or simply “MVTL” for short,employs plenty of fundamental communication facilities in IoV to collect vehicles communication information and communication of V2 V to gather data of single lane.The aggregation node of vehicle uploads information which is collected by utilizing communication of V2 I periodically to server.Just before aggregation node gets close to intersection,server would send traffic lights phase position computed to vehicles,thus real-time virtual traffic lights would come true.The result also considers the downstream lane which influenced by intersection,eases the blocking by controlling the duration before vehicle enters the congestion.The result of experiments demonstrates that when road is sparse,compared with CTL and existing virtual traffic algorithm,SVTL has better performance.Furthermore,when the traffic is heavy,MVTL has better consequence,80% of vehicles can reach an average speed of 10 m/s or more.