Shaken, Not Missed: Improving Touchscreen Tapping Performance under Turbulence via a Moving Keypad
Y.T. Oei (TU Delft - Aerospace Engineering)
D.M. Pool – Mentor (TU Delft - Control & Simulation)
G.G. McKenzie – Mentor (TU Delft - Control & Simulation)
Max Mulder – Graduation committee member (TU Delft - Control & Simulation)
Joost Ellerbroek – Graduation committee member (TU Delft - Operations & Environment)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Touchscreen usability degrades severely under turbulence due to biodynamic feedthrough (BDFT), which displaces the user’s finger from the intended input location. This study introduces and evaluates a moving keypad concept where the keypad shifts in real-time according to predicted BDFT. To generate these BDFT predictions, individualized-per-participant BDFT models were identified under turbulence for 18 participants in the SIMONA Research Simulator using two pointing tasks: a Relax Task, capturing BDFT while a user points passively to the screen, and a Position Task, where a user actively points to a fixed target on the screen. These models were then used to move the keypad during an evaluation task in which participants entered a four-digit pincode followed by an enter key to submit. The evaluation task was performed under six conditions: two baseline static keypad conditions (with and without turbulence), and four moving keypad conditions where "Full" and "Reduced" model variants of the Position Task and Relax Task were used to move the keypad. The Reduced model variants were created by lowering the model gain and increasing the damping ratio of the Full models to slow the keypad motion. In the static keypad baseline conditions, turbulence significantly degraded keypad performance, increasing misclicks by 621% from 0.28 to 2.02 misclicks per pincode. This confirms that BDFT degrades touchscreen performance. In the moving keypad conditions, the Full models drove the keypad with excessive speed, producing more than 5 misclicks per pincode, making these conditions effectively unusable. The Reduced models produced slower keypad motion and saw better keypad performance with less misclicks than the Full Models. Among them, the Position-Reduced condition was the only configuration that significantly improved tapping performance, lowering misclicks by 32% (to 1.38 misclicks per pincode) compared to the baseline static keypad under turbulence. The results indicate that effective mitigation requires slower and more damped keypad motion than predicted by pointing task models. This work demonstrates the feasibility of transparent, real-time BDFT mitigation for touchscreen interfaces and provide a promising direction toward reliable and effective touchscreen operation under turbulence.
Files
File under embargo until 27-11-2027