Mapping method and device based on image group

Abstract

The embodiment of the invention discloses a mapping method and device based on an image group. The method comprises the following steps: acquiring multiple groups of road images when a vehicle travelsfor multiple times in the same position area and a positioning pose corresponding to each road image; performing data association among the semantic features in each group of road images to obtain associated semantic features belonging to the same road sign among each group of road images, and determining a first position of the associated semantic features in a world coordinate system; determining a reprojection error between the associated semantic feature and the corresponding semantic feature in each road image according to the first position and the positioning pose corresponding to eachroad image, determining a projection error sum value between the reprojection errors, and determining a second position of the associated semantic feature in the world coordinate system according tothe projection error sum value; and adding the associated semantic feature and the second position to the map. By applying the scheme provided by the embodiment of the invention, the accuracy of the constructed map information can be improved.

Description

technical field [0001] The present invention relates to the technical field of intelligent driving, in particular, to a method and device for constructing a map based on an image group. Background technique [0002] In the field of intelligent driving technology, vehicle positioning is an important part of intelligent driving. Usually, when the vehicle is driving, the vehicle pose can be determined according to the satellite positioning system. However, when the vehicle travels to a scene with weak or no satellite signal, in order to accurately determine the positioning pose of the vehicle, it can be positioned based on visual positioning. [0003] Vision-based localization builds on pre-built high-resolution maps. In the high-precision map construction scheme, the vehicle usually collects road images during driving and estimates the vehicle pose based on the inertial measurement unit (IMU), directly based on the vehicle pose and detected from the road image. The semantic...

AI Technical Summary

Problems solved by technology

However, because the data in a driving process is too simple, the position of the semantic features constructed based on such data will not be accurate enough, that is, the constructed map information is not accurate enough.

Images