Investigating Traffic Avoidance Maneuver Decisions of Unmanned Aircraft Pilots

For unmanned aircraft to share airspace with manned aircraft, extensive testing is first required to ensure that such vehicles can fly safely with manned traffic. Safe operation includes not only avoiding collisions with other traffic but also complying with the Federal Aviation Regulations to remain “well clear” of other traffic. One method for investigating the safety of unmanned aircraft operations is fast-time Monte Carlo simulation of encounters between unmanned and manned aircraft. As part of that simulation, one must model how the pilots of unmanned aircraft react to the encounters. To that end, a stochastic model of realistic responses of unmanned aircraft pilots is being built. A preliminary model was formulated based on a review of existing literature on pilot decision-making, and Human-in-the-Loop experiments are being used to improve the model’s representation of unmanned pilot responses and parameterize its stochastics elements. This paper summarizes the first of those experiments, conducted in July 2015, and highlights key results that inform the pilot model.