Research On Intelligent Intersection Control Protocol And Route Planning Method Based On Autonomous Vehicles

The management of intersections is one of the most challenging tasks in traffic control.At present,intersections are managed by stop signs or traffic lights,these strategies aim to manage traffic flow,improve the efficiency and safety of intersections.With the rapid development of China's economy and the continuous improvement of people's life quality,the number of private vehicles has grown exploded in recent years,more and more vehicles are pouring into the urban traffic flow,the disadvantages of traffic light strategy begin to appear.In many cases,even if there are no vehicles in the intersection,the traffic light strategy will require vehicles wait at the entrance of the intersection,resulting in frequent traffic jams,affecting the economic benefits of the city and the quality of life of residents seriously,causing a large number of economic losses.With the rapid development of artificial intelligence,visual computing,wireless communication technology,autonomous technology is becoming more and more mature,autonomous vehicles and Intelligent Transportation System(ITS)has become one of the hot topics.The advantage of autonomous vehicles is that they can cooperate with each other,or interact with the control center of the intersection,which opens up the possibility of intelligent intersection management.In the near future,autonomous vehicles will become the core of urban traffic,many researchers have foreseen this future and put forward many intelligent intersection management strategies.However,there are some shortcomings in these control strategies,such as can not control in real time,the model is not optimized enough,so it is of great significance to put forward more efficient intelligent control strategies.In this paper,we propose an intelligent management protocol for the management of autonomous vehicles passing intersections,and provides route recommendation service for autonomous vehicles.Firstly,this paper puts forward the requirements of intelligent management protocol,defines the types of vehicle communication messages and the allowed operations,provides priority determination and speed adjustment strategies,reduces the average waiting time of vehicles at intersections greatly.Secondly,we apply SMOTE,RF and GBDT algorithms to predict the average waiting time of vehicles at intersections.We use SMOTE algorithm to process the initial data set to generate the balanced data set,and then use RF and GBDT algorithms to train model and predict the waiting time.Compared with training on the initial data set,using the synthesized data set can get a more efficient and accurate result.Thirdly,we propose iEigenAnt algorithm to solve the route planning problem.We prove the efficiency of iEigenAnt algorithm by mathematical analysis,experimental results and case study.Afterwards,we use iEigenAnt algorithm to find multiple shortest paths problem in the traffic network.Finally,we recommend route for autonomous vehicles based on the shortest driving distance or the shortest driving time,i.e.the minimum sum of the driving time on the road and the waiting time at each intersections.The main contributions of this paper are as follows:· Proposed an intelligent management protocol for the management of autonomous vehicles at intersections.Compared with the traditional traffic light management strategy,it can reduce the waiting time of vehicles at the intersection greatly and improve the traffic efficiency of intersection.· Using SMOTE,RF,GBDT algorithms to predict the waiting time of autonomous vehicles at intersection under different traffic flow in this protocol,which improves the accuracy of prediction and obtains better prediciton results.· Propose an improved iEigenAnt ant colony algorithm to solve the problem of finding the shortest route.Firstly,we prove the efficiency of this algorithm in solving combinatorial optimization problem by mathematical analysis,experimental results and case study.Then,we apply this algorithm on route planning problem to find multiple short paths.Finally,recommend route for autonomous vehicles according to the standards of the shortest driving time or the shortest driving distance,which imporves the efficiency of urban traffic.