An intelligent object manipulation framework for industrial tasks

Artur Saudabayev, Yerbolat Khassanov, Almas Shintemirov, Hüseyin Atakan Varol

2013

Closed access Metadata (CSL-JSON)
DOI 10.1109/icma.2013.6618173
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This paper presents an intelligent object manipulation framework for industrial tasks, which integrates a sensor-rich multi-fingered robot hand, an industrial robot manipulator, a conveyor belt and employs machine learning algorithms. The framework software architecture is implemented using a Windows 7 operating system with RTX real-time extension for synchronous handling of peripheral devices. The framework uses Scale Invariant Feature Transform (SIFT) image processing algorithm, Support Vector Machine (SVM) machine learning algorithm and 3D point cloud techniques for intelligent object recognition based on RGB camera and laser rangefinder information from the robot hand end effector. The objective is automated manipulation of objects with different shapes and poses with minimum programming effort applied by a user.