Sound and Complete Reconfiguration for a Class of Hybrid Systems

Reconfiguration is the automated recovery from a fault in hybrid systems such as tank systems from process engineering. Until today, many heuristics for this purpose have been developed. However, the reliability of these heuristics is not discussed or approximated using artificial systems. In this article, soundness and completeness properties are defined in the context of reconfiguration. In addition, an algorithm based on the transformation of the reconfiguration problem to search and a subsequent solution using Best-First Search is presented. It is shown that this algorithm is sound and complete for a class of hybrid systems.