Abstract
With the increasing demand for intelligent robot applications over the years, SLAM technology is also developing rapidly as an essential part of robot perception modality. SLAM is a vast system that involves multi-domain knowledge such as image pre-processing, matching, and camera pose estimation. In this paper, the basic architecture of SLAM based on binocular vision is elaborated, and some applications of its methods are presented in conjunction with new methods in recent years. The primary approach is to separate the front-end and back-end and achieve good tracking and map-building results using a multi-threaded approach. The front-end uses the SHI-Tomasi algorithm in OpenCV to detect the feature points, triangulation to calculate the landmark depth, and Lucas-Kanade optical flow method for tracking from one frame to the next and computing the pose transformation. G2O is used in the SLAM back-end to optimize the poses and nonlinearities. The analysis of SLAM based on binocular vision also provides an essential reference for people new to visual SLAM.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
