Linked Data, Augmented Reality (AR), and technical advancements in mobile information technology lead to an increasing desire to exploit Linked Data for the integration and visualization in mobile AR applications. However, current approaches are either bound to existing client-server-based infrastructures or use closed data sources and proprietary data formats. Moreover, a number of related approaches are built upon content-based recognition algorithms that are both memory and processing-intensive, require a permanent connection to a host, and thus are inappropriate for a direct deployment onto mobile devices. In this work, we present a computational model that builds on a sensor-based tracking approach and maps proactively replicated Linked Data sets to a virtual representation of the user’s vicinity computed by a mathematical model. We demonstrate the applicability of our approach through a proof-of-concept AR application that retrieves and aggregates mountain-specific data from a set of different sources and displays such data in a live-view interface. In consequence, our approach is resource-efficient, does not require a permanent network connection, is independent from existing server-based infrastructures, and allows to process Linked Data directly on a mobile device.

Top