Visual environment for designing sequence control program with AI planning

A visual environment for designing sequence control program with AI planning is described. In order to increase productivity and improve the quality and reliability of the control specifications, the authors have been developing a planning-based Visual Programming System for Sequence Control Program Specification Design (VIPS/S-SFCD). VIPS/S-SFCD features an example-based programming technique in which designers specify the initial states and goal states of machines using visual representations, and generates control specification by AI planning. To overcome the difficulty of dealing with complex data/knowledge of the planner, VIPS/S-SFCD analyzes and confirms user intention using context-dependent valence theory and case grammar. They used VIPS/S-SFCD to experimentally generate control program specifications for actually existing steel plants. The experimental results showed that visual representation with dynamic, intuitional icons, the coverage of the grammar rule, and generating time of control specification are practical