ABSTRACT
The modern scanning electron microscope (SEM) is a fully software-controlled instrument with significant automation of individual instrumental parameters. Optimum performance still demands operator expertise. A knowledge-based approach to SEM operation is proposed to embed this expertise within the control software to provide global automation. An analysis of SEM investigations has resulted in a model of the control task with some of the associated knowledge being codified into a formal representation. A standalone prototype knowledge-based system has been designed and implemented. Ongoing and future directions of this research are presented.
