Search for: All records

Objective and scalable pilot assessment is crucial due to rising training demands and subjective evaluation limitations. Existing data-driven methods often require extensive data labelling and focus on maneuver classification over execution quality. Addressing the need for interpretable, label-independent proficiency metrics, this paper introduces an unsupervised pipeline to quantify roll control smoothness during turns from X-Plane logs. The methodology features adaptive trend slicing for segmentation, zero-phase Butterworth filtering (0.6-4.5 Hz) to isolate roll oscillation during turns, and extraction of time domain descriptors (RMS, Peak to Peak, and crest factor). After aggregation and outlier removal, via isolation forest, K-means clustering identified distinct “smooth” vs “oscillatory” performance groups with high statistical validity. Principal Component Analysis (PCA) further derived a continuous “smoothness index,” capturing 79.6% of the variance. This study confirms unsupervised oscillation metrics can differentiate control finesse, offering a quantitative foundation for augmenting training assessment and enabling future calibration against expert evaluations. 

Poor synchronization of physiological and flight data hinders human factors research by obscuring cause-effect relationships and requiring laborious manual alignment. We developed a custom X-Plane plugin which uses Lab Streaming Layer (LSL) for real-time synchronization to integrate flight data with simultaneously streamed ECG and eye-tracking signals. Our setup simplifies data collection and ensures precise synchronization without manual intervention, reducing error and preprocessing time. This paper describes the system configuration, plugin, and advantages over conventional methods. Our approach enables accurate correlation between physiological responses and in-flight events, offering deeper insights into pilot performance and workload, resulting in a practical, reproducible method of simplifying multimodal data collection for accessible and efficient research in aviation human factors.