The Research And Realization Of Vehicle-road Collaborative Path Planning System

In recent years,the state has paid more and more attention to the de-velopment of intelligent connected vehicles.However,there are still some problems in the automatic driving technology based on single car intelli-gence,such as the complexity of on-board perception equipment,the large demand of calculation,the existence of visual blind area and so on.With the development of Internet of vehicles technology,the automatic driving technology based on Internet of vehicles provides new technical support for the development of automatic driving.With the support of the Internet of vehicles,vehicles can communicate with other vehicles and infrastructure around through V2X technology,to broaden the range of vehicle perception and refine the granularity of per-ception.With the support of edge computing,vehicles can also unload cer-tain complex computing tasks to the edge cloud,to reduce the demand for on-board computing capacity and assist vehicles to improve the quality of path planning and even trajectory planning.In this paper,the assistant driving method supported by edge compu-ting is studied,and a collaborative path planning method based on three-dimensional space-time traffic value map is proposed.In this method,the edge computing nodes gather the traffic elements information(such as ve-hicles,pedestrians,roadblocks,etc.)detected by vehicles and road test sensing devices,process the rich sensing information,and provide it to ve-hicles in the form of traffic value map.Based on the traffic value map,vehicles can directly make the global optimal driving decision,thus reduc-ing the demand for on-board computing power.Experiments show that the proposed path planning method has better performance in dynamic envi-ronment.On the basis of this method,this paper designs and implements a multi vehicle cooperative driving route planning service system to realize the optimal path planning function of vehicles in complex environment.