A Bayesian Network Model of Pilot Response to TCAS Resolution Advisories

The effectiveness of an airborne collision avoidance system (CAS) is influenced by the manner in which pilots respond to the system’s advisories. Current pilot response models used in CAS modeling and simulation are agnostic to parameters affecting pilot response in individual encounters and therefore treat all encounters equally. Simulations using these models can potentially underestimate collision risk in encounters where pilot response probability is low. This paper proposes a parametric pilot response model built from operational data using Bayesian networks. A network was constructed from radar recordings of TCAS encounters and the encounter parameters with the strongest influence on pilot response were identified. These parameters can be used to predict the probability of pilot response for individual encounters. The model was employed in simulation of safety-critical encounters. Results showed that standard pilot response models may underestimate collision risk. These results have implications for the design and performance evaluation of separation advisory systems, including collision avoidance and detect and avoid systems. Key Words—Traffic Alert and Collision Avoidance System (TCAS), Airborne Collision Avoidance System (ACAS), detect and avoid (DAA), aviation safety, Bayesian networks, pilot response, aircraft separation