Global workspace theory inspired architecture for autonomous structural health monitoring

Abstract

Structural health monitoring (SHM) systems provide automated assessments of structural health by processing data from sensors fastened to a structure. Most SHM research is focused on developing quick state assessments (i.e., reflexive techniques) using a form of pattern recognition. However, little attention has been given to developing a more `thoughtful', logic-based method for state assessments to enable operational-level decisions. Reflexive-based assessments are unreliable because they often cannot discriminate between changes due to structural damage and other variants (e.g., temperature). A new architecture is being proposed that combines reflexive and deliberative elements for enhanced state assessments and operational decisions. The architecture is demonstrated in the laboratory using a representative airframe component.