Title :
Fuzzy seam-tracking controller
Author_Institution :
Toshiba Corp., Tokyo, Japan
Abstract :
The welding spot of an automatic laser welding system is very small, thus requiring real-time accurate control of the welding position. The author has developed a seam-tracking controller that uses fuzzy logic. This controller incorporates an operator´s seam recognition ability and uses a method for auto-tuning the fuzzy rules. The major difficulty of an optical sensing seam tracking controller is to estimate the seam location from an image which may be graded by changes in the workpiece´s surface condition. The inference rule for seam detection using fuzzy logic can incorporate ambiguous descriptions. It can be used to infer the appropriate seam location from a graded image which does not exactly match the rules
Keywords :
computer vision; fuzzy control; fuzzy logic; image processing; inference mechanisms; laser beam welding; position control; process computer control; real-time systems; auto-tuning; automatic laser welding system; computer vision; fuzzy control; fuzzy logic; fuzzy rules; image processing; inference rule; position control; process computer control; real-time; seam-tracking controller; surface condition; welding spot; Automatic control; Control systems; Feature extraction; Fuzzy control; Fuzzy logic; Laser beams; Laser tuning; Optical control; Real time systems; Welding;
Conference_Titel :
Industrial Electronics, Control, Instrumentation, and Automation, 1992. Power Electronics and Motion Control., Proceedings of the 1992 International Conference on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-0582-5
DOI :
10.1109/IECON.1992.254453