Abstract
Information assistance helps in many application domains to structure, guide and control human work processes. However, it lacks a formalisation and automated processing of background knowledge which vice versa is required to provide ad-hoc assistance. In this paper, we describe our conceptual and technical work to provide information assistance for smart assembly stations. Our contribution comprises a conceptual architecture which was implemented in an industrial prototype, the Plant@Hand smart assembly trolley. We describe three major aspects of our approach, the recognition of assembly tasks based on probabilistic models, decision making using a cognitive architecture, and strategies for dealing with sensor-based and knowledge-based errors. Finally, we present the industrial prototype setup.