Research On The Real-time Vehicle Routing Problem Within A Specific Region

With the high elaboration development of the electronic map data and the rapid development of the Internet of things,sensors and Internet technology,it is possible to make a research on the unmanned aerial vehicle.Using unmanned driving instead of human driving,in the process of driving the vehicle with the computer instead of the human brain to analyze and deal with all kinds of road conditions,thereby avoiding or reducing the traffic accidents caused by driver fatigue and errors.Path planning is one of the key issues in unmanned vehicles.In this paper,we study the problem of driving scenes in a specific area,propose a vehicle routing scheme in a specific region and develop a test software package.Specific work as follows:The path planning problem in a specific area is a planning of the traffic path between the starting and ending points in a complex urban road.In this paper,we present a solution of global optimization and local optimization with multiple indexes.Firstly,the real map is transformed into a feature grid map,and the map information is extracted and simplified.Secondly,according to the characteristics of the grid map information,the travel time,travel distance and driving oil consumption are used as the evaluation index to select the optimal path,and the Floyd algorithm is used to plan the global optimal path in a specific region.Finally,the local path planning problem is defined for a selected global planning path and the local path is solved by improved A*algorithm.Set Xi'an Lianhu as the center,we selected respectively 15km x 15km,12km×12km,9kmx9km,6kmx6km,3km×3km part of the map data to make a simulated experiment.The experiment results show that the path planning time using the Floyd algorithm can be controlled within 60s and the local path planning time of the specific path can be controlled in 100ms.The experimental results show that the proposed scheme can effectively achieve the vehicle routing planning in a specific region.Based on the proposed solution,a software package is designed and developed to solve the vehicle routing planning problem in a specific region.The package is developed using the Unity3d engine and Visual Studio2012.The package is run on a mobile workstation(The fourth generation of Intel core i7-4710MQ 2.50GHz,memory is 16GB)installed with the WindowslO operating system.There are mainly five function modules in the software package,including simulation of raster map,plan of global path,establishment of time matrix,establishment of fuel consumption matrix and plan of local path.In the module of simulation of raster map,provide functions like customizing the scale of the map and creating raster map.In the module of plan of global path plan of global path,mainly solve optimal path for vehicles among the intersections aimed at the information in static indicators of sections.In the module of establishment of time matrix,mainly solve the shortest time for vehicles among intersections in specific areas,aimed at the drive time of the path.In the module of establishment of fuel consumption matrix,aimed at the fuel consumption on the path,mainly solve the minimum fuel consumption for vehicles among intersections in specific areas.In the module of plan of local path,accomplish the plan on a single path.To sum up,this paper studies the vehicle routing planning problem in a specific area.In this paper,a new method is proposed to solve the problem of global optimization and local optimization with multi index and get better planning results,which provides a new way of thinking to solve the vehicle routing planning problem in a specific area.