DBGRIP - A LEARNING EXPERT SYSTEM FOR DETAILED SELECTION OF ROBOT GRIPPERS


PHAM D., TACGIN E.

INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH, cilt.29, sa.8, ss.1549-1563, 1991 (SCI-Expanded) identifier identifier

Özet

A practical expert system for selecting commercially available robot grippers is presented. The system is implemented using a powerful hybrid expert system development tool coupled with a database management program. The knowledge base employed by the expert system in the selection processes is outlined. Details of the main features of the system are given, including its ability to learn by positive reinforcement of past experience.