@inproceedingsmultidimensional2015, Author = {Suck, Stefan and Fortmann, Florian}, Title = {Multi-dimensional Pilot Crew State Inference for Improved Pilot Crew-Automation Partnership}, Year = {2015}, Month = {3}, Publisher = {IARIA}, Booktitle = {Proc. of COGNITIVE'15}, type = {inproceedings}, note = {Automation is a substantial technology of modern aircraft. Even though automation has significantly improved aviation safety, insufficient partnership between the pilot crew and the automation, and confusion over the status of the automation is still a pr}, Abstract = {Automation is a substantial technology of modern aircraft. Even though automation has significantly improved aviation safety, insufficient partnership between the pilot crew and the automation, and confusion over the status of the automation is still a problem. The European project A-PiMod addresses these problems by developing a virtual crew member, which takes the position of classical aircraft automation. As part of the crew, the virtual crew member must be able to anticipate the internal states of the human crew members. This ability helps, e.g., to improve the task share in the cockpit by means of dynamic adaptations of task distributions. In this paper, we present the concept of the A-PiMod pilot model, which will be used for inferring the internal state of the human crew members. The internal state is composed of different sub-states, which have been defined during the initial phase of the project. The addressed sub-states are situation awareness, workload, and intentions. The target states will be inferred based on real-time data about the mission, tasks, and pilot behaviors, including what they say, where they look at, and how they act. } @COMMENTBibtex file generated on