Which Minimum Visual Tracking Performance is Needed in a Remote Tower Optical System?

Abstract

To maintain or even increase the ATCO’s situation awareness at a remote tower working position, augmentation features are introduced, such as automated tracking of objects. Moving objects (aircraft, vehicles, persons, etc.) the ATCO is interested in are tracked and augmented by this function. However, a tracking function is never reliable by 100% and nuisance tracking information occur, information, which is not of operational relevance and is disturbing to the ATCO. Not only human performance gains by “wanted” tracking information but also losses due to “nuisance” tracking information are to be expected. This paper investigates the effect of visual tracking function in a Remote Tower Optical System on ATCO’s acceptance and effects on situation awareness and workload. Erroneous tracking performance will be discussed within the framework of a response matrix (e.g. Wickens, Elementary Signal Detection Theory, Oxford University Press (2002) and Appendix B). It collects the correct and false (system) responses as conditional probabilities for two alternative situations (object to be tracked/object not to be tracked), or for signal and noise. In a human-in-the-loop real time simulation, seven ATCOs performed a realistic traffic scenario. The study was conducted at Remote Tower laboratory at DLR in Braunschweig. As an experimental condition the performance of the visual tracking was operationalized by the number of nuisance tracking indication: (1) none (no visual tracking (baseline)), (2) low, (3) medium, and (4) large number. The results show that ATCOs very much appreciate visual tracking information. ATCOs can more easily detect critical traffic situation, which increases their situation awareness and safety. Further on, acceptance is rather high and workload on a moderate level, and both parameters behave rather robustly, even when the number of nuisance tracking information increases.