The path and location planning of workpieces by genetic algorithms

NCU Institutional Repository > 工學院 > 機械工程研究所 > 期刊論文 > Item 987654321/27058

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/27058

題名:	The path and location planning of workpieces by genetic algorithms
作者:	Chen,CJ;Tseng,CS
貢獻者:	機械工程學系
關鍵詞:	N/A
日期:	1996
上傳時間:	2010-06-29 18:06:40 (UTC+8)
出版者:	中央大學
摘要:	This paper presents the planning of a near-optimum path and location of a workpiece by genetic algorithms. The purpose of this planning is to minimize the processing time required for a robot to complete its work on a workpiece. The location of the workpiece can be anywhere by translating it along any direction and by rotating it about the fixed z-axis of the robot coordinate system. Owing to the changeable location of the workpiece and the alterable motion time required for a robot to move between two workpoints, the path and location planning problem is much more complicated than the travelling salesman problem. It is definitely impossible to obtain an optimum path and location within an acceptable time. In this paper, genetic algorithms are applied to solve this problem. The location of the workpiece is defined by three position parameters and one angular parameter, and the path is determined based on the values of the parameters for all workpoints. All the path and location parameters are encoded into a binary string. They are modified simultaneously by genetic algorithms to search for a global solution. As the workpiece can be anywhere, a penalty function is used to prevent the selection of illegal paths. Two experiments are given to show the performance of genetic algorithms: one has 30 workpoints and the other has 50 workpoints. Compared with four human-generated plannings, planning by genetic algorithms has much better performance in minimizing the processing time.
關聯:	JOURNAL OF INTELLIGENT MANUFACTURING
顯示於類別:	[機械工程研究所] 期刊論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	494	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....