Implementation, improvement, and testing of robotic vision algorithms
Chew, Ivan Sze Wei
Date of Issue2017
School of Mechanical and Aerospace Engineering
Robotics Research Centre
The report describes point cloud and RGB pose estimation systems implemented in robotics applications. Speed improvements for point cloud based pose estimation were developed and their results discussed. A fiducial marker, RGB-D based pose estimation was developed, proving to be the superior alternative. The concept is explained and the development process described in detail. In-depth accuracy analysis of the camera system used was also performed and detailed.
Final Year Project (FYP)
Nanyang Technological University