ORB Algorithm Applied to Detection, Location and Grasping Objects

Grasping objects is a highly used activity in the industry but is often done without the aid of vision, limiting the use to few controlled applications. This paper proposes a practical and fast way to grasp objects with a robotic arm. Using a RGB-D visual sensor, the Kinect, it is proposed to use it...

Full description

Saved in:
Bibliographic Details
Published in:2018 Latin American Robotic Symposium, 2018 Brazilian Symposium on Robotics (SBR) and 2018 Workshop on Robotics in Education (WRE) pp. 176 - 181
Main Authors: Conceicao, Andre G.S., Oliveira, Daniel M., Carvalho, Maria Paula
Format: Conference Proceeding
Language:English
Published: IEEE 01-11-2018
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Grasping objects is a highly used activity in the industry but is often done without the aid of vision, limiting the use to few controlled applications. This paper proposes a practical and fast way to grasp objects with a robotic arm. Using a RGB-D visual sensor, the Kinect, it is proposed to use its point cloud to locate an object pose. To find the object, we propose a feature based method, being the feature extractor the ORB algorithm. Together with RANSAC, it is going to find the object position in the RGB image, and find the pixels coordinates in the point cloud. After the object detection, we propose a way to grasp based on the estimated position by the point cloud and the robotic arm position. To develop the system, ROS and OpenCV are used. We demonstrate, through experiments, the efficiency of the proposed system in a real world application.
DOI:10.1109/LARS/SBR/WRE.2018.00040