Research On Real-time Route Planning Of Cooperative UAV-intelligent Platoons System

With the development of economy,the number of automotive vehicles is increasing day by day.Traffic jams often occur,which will prolong travel time and even seriously affect daily work.The problem of route planning has always been a hot issue.By avoiding congested streets and choosing an optimal route,travel time can be greatly reduced.The main contents and achievements of this paper include:(1)The UAV is introduced into road traffic,and the UAV-intelligent platoons collaborative system is proposed,which combines vehicle navigation system and UAV,and has two ways to obtain information.One is to collect information from roadside units of the intelligent transportation system,the traffic control center processes,generates,publishes and sends the information to vehicle.The other is that the UAV takes the road video and the vehicle processes the images to get information.According to the two ways of obtaining information,two working modes are named: vehicle-road collaborative mode and UAV-intelligent platoons collaborative mode,then the workflow of the system is introduced.(2)The UAV-Intelligent platoons collaborative mode is the cooperation of UAV and vehicles.The UAV takes videos over the road and transmits them to vehicle,then,the images are processed by the vehicle.In the aspect of image processing,using background difference method and Canny edge detection method to judge the existence of vehicles,then the queue length before the intersection is calculated by transforming the length relationship between the unit pixel of the picture and the actual distance.Then a new calculation method of road weight is put forward,that is,using the obtained queue length,and considering the vehicles driving in different lanes and the situation of multiple queues,the road weight is calculated.According to the road weight,Dijkstra algorithm is used to calculate a route with the shortest travel time.(3)The vehicle-road collaborative mode is to obtain information from intelligent transportation system.The mode obtains two kinds of information: the queue length before the intersection and the signal light display when the vehicle just enters the road.In this mode,a new calculation method of road weight is proposed,that is,according to the two kinds of information,and combined with the multiple signal light display when the vehicle arrives at the intersection,driving in different lanes and multiple queues,road weight is calculated.Likewise,the Dijkstra algorithm is also used to calculate a route with the shortest travel time.(4)Three scenarios are used to verify the route planning method in two working modes.The results of MATLAB simulation are consistent with the theoretical analysis,which verify the effectiveness of the algorithm.