Monocular vision-combined RGB-D SLAM method

Abstract

The invention discloses a monocular vision-combined RGB-D SLAM method. The method comprises the following steps of: respectively carrying out feature extraction, feature matching, motion estimation and optimization on data obtained by a depth sensor by adopting a monocular vision method and an RGB-D depth vision method; establishing a monocular local position map and an RGB-D local position map; carrying out three-dimensional map fusion on the monocular local position map and the RGB-D local position map by a robot in a motion process so as to construct a global map; and carrying out closed loop detection on the fused position maps and carrying out optimization to obtain an optimum global map. The RGB-D SLAM method disclosed by the invention is high in timeliness, strong in adaptability and good in stability, and can be used for solving the problem that RGB-D depth sensors cannot obtain depth information or are insufficient in obtaining of the depth information.

Description

technical field [0001] The invention relates to a RGB-D SLAM method combined with monocular vision. Background technique [0002] Vision sensors can provide compact, accurate, non-invasive and easy-to-understand visual information. At the same time, in addition to providing rich visual information, vision sensors are inexpensive and cost-effective. At present, in the field of mobile robots, monocular vision methods and RGB-D SLAM methods are commonly used. [0003] The monocular vision SLAM method refers to the method of using one step and three turns to calculate the depth information by using the triangulation method in the state of robot movement, and put the depth information into the map to construct a three-dimensional map environment model, but in the actual process Among them, the depth calculation consumes a lot of computing resources and the speed is relatively slow. [0004] Since the depth sensor was applied to the SLAM field, more and more developers have shif...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a RGB-D SLAM method combined with monocular vision, which is a RGB-D visual SLAM method with high real-time performance, strong adaptability and good stability, and solves the problem that the RGB-D depth sensor cannot acquire Obtaining depth information or obtaining insufficient depth information

Images