High-precision positioning and mapping method based on visual edge features

Abstract

The invention provides a high-precision positioning and mapping method based on visual edge features, relates to the field of machine visual positioning, and aims to improve the description stability of re-projection errors by using the robust edge features and using the edge features less affected by illumination variation as information sources in a visual odometer. Meanwhile, in order to improve the description stability of a re-projection error, a geometric error and luminosity error fusion mode is used for describing an error generated by edge information, the robustness of pose estimation and the environmental adaptability of an algorithm are improved, on the basis that an initial pose is obtained, initial correction is conducted on a point cloud, and then the ICP algorithm is used to accurately match the point cloud of the key frame and the point cloud of the current frame to obtain a final pose, so that the positioning accuracy of the algorithm is improved.

Description

technical field [0001] The present disclosure relates to the field of machine vision positioning, in particular to a high-precision positioning and mapping method based on visual edge features. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] With the development and improvement of robot technology, vision-based robot positioning technology has gradually become a research hotspot in the key technology field of robots. Currently commonly used visual positioning sensors include monocular cameras, binocular cameras, and depth cameras. [0004] The inventors found that in the current edge positioning method, by detecting whether the captured image includes stairs, when the captured image includes stairs, extract the edge information of the stairs according to the captured image, and then perform edge positioning based on the edge information...

AI Technical Summary

Problems solved by technology

[0004] The inventors found that in the current edge positioning method, by detecting whether the captured image includes stairs, when the captured image includes stairs, extract the edge information of the stairs according to the captured image, and then perform edge positioning based on the edge information of the stairs to obtain the edge positioning information. However, there is still the problem of uneven distribution of features, resulting in errors, which seriously affect the effect of pose estimation, and it is difficult to meet the current environmental adaptability and accuracy requirements for pose estimation.

Images