Domain Spanning Search for the Identification of Solution Patterns for the Conceptual Design of Self-Optimizing Systems

The design of mechatronic systems is an interdisciplinary task. With the introduction of the new paradigm of self-optimization the number of domains involved in the early design phases is even increased. Mechanical, electrical, control, and software engineers are involved as well as experts from mathematical optimization and artificial intelligence. The domain spanning modelling of solution patterns for self-optimizing systems and the definition of appropriate means to identify these patterns are only two of the challenges arising from this new paradigm. This paper introduces the active pattern for self-optimization as a new type of solution pattern and defines the challenges arising from the various domains involved in the design process. To meet these challenges a concept for a domain spanning search is introduced and examplified.