Application of map semantic information fusion in end-to-end autonomous driving decision planning
In order to improve the safety of autonomous driving and realize the high-precision decision planning of end-to-end autonomous driving,this article studies the fusion of map semantic information integration in end-to-end autonomous driving decision-making planning.The basic data of the mainstream autonomous driving is obtained,and the convolution operation is introduced to extract the image features.Combined with map semantic information fusion,end-to-end decision model is established to realize multi-modal multi-task end-to-end decision planning.Through examples,it is proved that the new decision planning method can provide high precision running speed and steering corner for autonomous driving,and promote the improvement of operation safety of autonomous driving.
map semantic information fusionautonomous drivingplanningdecision-makingend-to-end
万方数据
维普
