Track planning method and device, medium and unmanned device

Abstract

The invention discloses a track planning method and device, a medium and an unmanned device. The track planning method comprises the steps of collecting obstacle data of obstacles around a road where the unmanned device runs; for different set projection models, determining cost parameters corresponding to the projection models according to the obstacle data of the obstacles; according to the obtained cost parameters corresponding to the different projection models, selecting a used projection model for the obstacle from the preset different projection models; determining projection coordinates of the obstacle projection in a road coordinate system corresponding to the running road by using the selected projection model; and planning a driving track for the unmanned device according to the projection coordinates of the obstacle. Therefore, the appropriate projection model is selected for the difficulty of planning the driving track for the unmanned device, the planned driving track can be ensured to reasonably avoid the obstacles, planning resources are saved, and the efficiency of planning the driving track for the unmanned device is improved.

Description

technical field [0001] This specification relates to the field of unmanned driving technology, in particular to a trajectory planning method, equipment, medium and unmanned equipment. Background technique [0002] With the development of artificial intelligence technology, computer technology and Internet of Vehicles technology, unmanned driving technology has emerged as the times require. With the unmanned driving technology, unmanned vehicles have also received widespread attention. [0003] In order to ensure the driving safety of unmanned vehicles, it is necessary to provide reasonable and effective driving paths for unmanned vehicles. A trajectory planning module is usually deployed in an unmanned vehicle, and the trajectory planning module is mainly used to plan a local driving path for the unmanned vehicle. [0004] When planning the trajectory of an unmanned vehicle, it is necessary to consider the obstacles around the unmanned vehicle to ensure that the planned tr...

AI Technical Summary

Problems solved by technology

However, in practical applications, it is found that using the closest point projection method to determine the projected coordinates of obstacles on these complex roads, it is easy to have a large distance between the determined projected coordinates and the actual position of the obstacle corresponding to the position coordinates in the road coordinate system. Therefore, it consumes more computing resources to plan the driving trajectory for the unmanned vehicle based on the projected coordinates, and the planned driving trajectory is not necessarily optimal, and even seriously affects the subsequent driving of the unmanned vehicle ( For example: causing driverless vehicles to go nowhere later)

Images