Please use this identifier to cite or link to this item:
http://hdl.handle.net/123456789/4613
Title: | Autonomous Edge Detection Improvement using Adapting Gradient Methodology for the Automation of Robotic Teaching Process |
Authors: | Samsi Md Said, Amir Sharizam Ismail Ishkandar Baharin |
Keywords: | Contour tracking TCP Single Sensor Logic Adapting Gradient Programming Automation |
Issue Date: | 19-Nov-2013 |
Abstract: | Two Sensors is being used in on and off schem to automate the industrial robot automatic edg detection process. In dual sensor logic th ascending, descending and flat curve can b identified immediately and the next movement ca be planned. This work is trying to improve fro two proximity sensors to only one sensor for th said task. In order to implement one sensor fo normally dual sensor approach in general contou tracking application need creativity in roboti programming. The creativity aspect need to b tapped in order to plan sensing movement an application movement simultaneously with onl one sensor since the objective of contour trackin work is to do online tracking. Online trackin meaning the robot tracks unknown contour an performing application work such welding o painting simultaneously. Offline tracking is quit forgiving since the robot can do any movement t capture contour information in the searching phas while filtering out the noise points in th subsequent execution phase such as doing weldin or sealing jobs. The result from this work show that the adapting gradient method reduced th overall accumulation of tracking error even thoug employing only single sensor for detection purpose |
Description: | Conference Venue : UniKL MFI |
URI: | http://ir.unikl.edu.my/jspui/handle/123456789/4613 |
Appears in Collections: | Conference Paper |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Autonomous Edge Detection Improvement using Adapting Gradient Methodology for the Automation of Robotic Teaching Process.pdf | 201.6 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.