Model-Based Mitigation of Biodynamic Feedthrough for Touchscreen Dragging Tasks in Turbulence

Abstract

The anticipated arrival of touchscreens on the commercial flight deck will make pilots vulnerable to erroneous screen inputs under vibration, creating a potential safety hazard. Biodynamic feedthrough (BDFT) has shown to be a key obstacle in continuous touchscreen dragging tasks under simulated turbulence, disrupting task performance. This research therefore focuses on the implementation of a model-based approach to mitigate the adverse effects of BDFT. A human-in-the-loop experiment with 18 participants was performed. The experiment consisted of two simulator sessions, the first with the goal of collecting data used for identifying two BDFT models: a subject-average (SA) and a one-size-fits-all (OSFA) model. In the second session these two models were tested for their ability to cancel BDFT in the same two-dimensional pursuit task they were identified from, and in an additional point-to-point dragging task emulating a waypoint modification. The model-based BDFT cancellation approach was tested for two combinations of motion disturbance axis and touchscreen input direction: sway (side-to-side motion) with horizontal screen inputs and heave (up-down motion) with vertical screen inputs. The results showed it is possible to cancel between 80-90% of the BDFT in both cases despite poor cancellation for certain specific participants. The point-to-point dragging task showed much less BDFT than the continuous task used for BDFT model development, however, making the cancellation ineffective. Overall, the results show that while model-based BDFT cancellation is possible, it is crucial to account for individual variability and the specific touchscreen task.