Flight Testing an Optionally Piloted Aircraft: a Case Study on Trust Dynamics in Human-Autonomy Teaming

TL;DR

This study addresses how trust in human-autonomy teaming evolves during flight testing of an optionally piloted aircraft. By reanalyzing ~200 hours of flight data through DSL, 3P, and IMPACTS-H trust models, it reveals phase-dependent trust thresholds and homeostatic dynamics across taxi, takeoff, enroute, and landing. The findings demonstrate that static trust factors cannot fully predict real-time trust behavior, highlighting the need for dynamic trust models and quantitative measures to guide design-for-trust in future HAT systems. The work offers practical insights for improving safety-critical aviation autonomy and outlines concrete directions for future controlled studies and model development.

Abstract

This paper examines how trust is formed, maintained, or diminished over time in the context of human-autonomy teaming with an optionally piloted aircraft. Whereas traditional factor-based trust models offer a static representation of human confidence in technology, here we discuss how variations in the underlying factors lead to variations in trust, trust thresholds, and human behaviours. Over 200 hours of flight test data collected over a multi-year test campaign from 2021 to 2023 were reviewed. The dispositional-situational-learned, process-performance-purpose, and IMPACTS homeostasis trust models are applied to illuminate trust trends during nominal autonomous flight operations. The results offer promising directions for future studies on trust dynamics and design-for-trust in human-autonomy teaming.

Figures (7)

• Figure 1: The OPA used in this study, shown during remotely supervised autonomous flight tests near Canadian Forces Base Cold Lake.
• Figure 2: The GCS used by remote pilots and flight test engineers to monitor and interact with the OPA from a ground-based setting.
• Figure 3: Test pilot trust dynamics during a sample taxi sequence, highlighting a relatively stable trust curve and a visualization of test pilot impatience impacting the trust threshold. When the threshold exceeds the trust, human intervention occurs.
• Figure 4: Crew member trust dynamics during a sample takeoff sequence. The threshold steadily rises during the takeoff roll owing to increased perceived risk and workload, before eventually re-establishing along a new baseline for flight.
• Figure 5: Comparison of pilot hand positions during periods of increased perceived risk (top, such as during takeoff or landing) versus decreased perceived risk (bottom, such as during taxi or enroute).