A MODEL-BASED DESIGN OF A DYNAMIC APPROACH STABILIZATION ADVISORY SYSTEM

This paper describes the model-based approach taken by the manned aeronautics vehicle (MAV) group of the EU-ARTEMIS JU project D3CoS (Designing Dynamic Distributed Cooperative Human-Machine Systems) to design an approach stabilization advisory system (AStA) with a strong cooperative, multi-agent perspective. Two experimental cycles were conducted on a Boeing 737 full- flight simulator. During these experiments we were able to induce unstabilized approaches and collect data about human-machine interaction, psycho-physiological and behavioural data. Outputs from the first experiment served as an input for development of AStA as well as for the definition of normative standard tasks, which have been modelled with a formal procedure editor (PED). The latter experiment demonstrated the usability of the developed approach stabilization system and was used as a comparative baseline for outputs of cognitive architecture utilizing modelled pilot behaviour. Based on this, we present the latest development towards an approach stabilization system, which monitors aircraft performance, human-machine-interaction and, subsequently, pilot behaviour or cognitive state.