STREAMLINING INFORMATION FLOW IN CIVIL INFRASTRUCTURE INSPECTIONS THROUGH AUTOMATED REGISTRATION PROCEDURES

Abstract

Expert-driven inspections on civil infrastructure assets remain fundamental for accurately capturing and assessing asset conditions. However, the manual geometric and semantic alignment of inspection data with existing digital models (field-to-model data registration) and on-site access to relevant information (model-to-field data registration) represent onerous procedures that cause significant information loss and data fragmentation, slowing down maintenance processes and emergency responses. This paper proposes digital information flows to automate both the on-site access to and the storage of infrastructure asset data, leveraging automated bidirectional data registration procedures. A field experiment is presented to demonstrate the technical feasibility of the proposed bidirectional information flow, including real-time performance of both data registration directions and qualitative assessment of registration outcomes through visual overlay in condition assessment operations. This work paves the way for the automated feeding of expert agents, both human and virtual, enabling a new paradigm of human-machine cognitive collaboration in inspection and maintenance activities.