Loading [MathJax]/extensions/MathMenu.js
Application of data mining techniques to monitor the network-controllable robot's performance | IEEE Conference Publication | IEEE Xplore
Scheduled Maintenance: On Tuesday, 23 September, IEEE Xplore will undergo scheduled maintenance from 1:00-5:00 PM ET (1800-2200 UTC). During this time, there may be intermittent impact on performance. We apologize for any inconvenience.

Application of data mining techniques to monitor the network-controllable robot's performance


Abstract:

This study presents the data mining techniques to monitor and classify the network-controllable robot's performance. The robot is a part of the networked production syste...Show More

Abstract:

This study presents the data mining techniques to monitor and classify the network-controllable robot's performance. The robot is a part of the networked production system, which can be accessed and controlled via the Internet. In this study, the robot's repeatability is the main focus of the performance variables, of which being gauged with the use of a networked vision system and a precision calibration grid. The collected data are analyzed and made available over the network. The ability to monitor and analyze the robot's important performance variables presents many benefits. For example, the remote operator can decide whether the robot is suitable for a new assembly task. The data mining techniques utilized in this study effectively classify the repeatability data. The visual display of classification results is also presented.
Date of Conference: 06-09 December 2011
Date Added to IEEE Xplore: 29 December 2011
ISBN Information:

ISSN Information:

Conference Location: Singapore

I. INTRODUCTION

At Ajou University, a networked robotic assembly cell for the production of automotive starter gear housing units has been developed and installed in the Web-enable Enterprise Systems Lab. The total cost including the equipment and the customized design is about $85,000, while the construction time was over two months. The main goal is to create an Internet-controllable production line, which can be controlled and monitored over the network. The ubiquitous accessibility to the production cell presents many benefits. The operators can upload new instructions and part programming from anywhere at anytime. In the event of anomalies, the operators can diagnose and troubleshoot without having to travel to the site. The cell represents highly advanced industrial equipment with the real components from the automotive industry. The main constituents are connected on the LAN through their IP/Gateway addresses, consisting of two Yamaha SCARA (selective compliance assembly robot arm) robots (YK-220X & YK-350X), a Yamaha single-axis robot (F1030-BK1050) three robot controllers (two RCX 142s, one SR1-X) a programmable logic controller (PLC), a Smart Image sensor (DVT 545C vision system), a Sony PZT network camera (SNC-RZ25N), Omron optical sensors, two Festo pneumatic grippers for the SCARA robots, one Festo pneumatic clamp and a part carrying system, Festo vacuum control switches, a touch screen LCD interface unit, and various fixture devices. More detailed descriptions on the cell and other closely related studies are in the literature [1], [2]. Figure 1 illustrates the overall outline of the cell. Figure 2 shows the overall assembly sequence. The robotic assembly cell at Ajou University. Operating sequence of assembly cell.

Contact IEEE to Subscribe

References

References is not available for this document.