Initial Study of Controller/Automation Integration for NextGen Separation Assurance

Abstract

This paper describes an initial human-in-the-loop part-task study on service providerbased automated separation assurance for the Next Generation Air Transportation System (NextGen). The study varied levels of trajectory-based separation assurance automation from manual trial planning to fully automated conflict resolution over 1x, 2x and 3x traffic densities. The experiment was conducted in the Airspace Operations Laboratory at NASA Ames Research Center in August 2007. In this study envisioned trajectory based operations for the NextGen were simulated. The ground-side automation was responsible for conflict detection. Four recently retired air traffic controllers and five aviation knowledgeable students participated and were responsible for resolving potential conflicts between aircraft with varying levels of automation. Two combined sectors engaged participants in solving conflicts between an average of 30 (1x), 60 (2x), and 90 (3x) aircraft during the 30 minute long simulation runs, resulting in approximately 6 (1x), 42 (2x) and 90 (3x) conflicts per run. The within subjects design gave operators access to a fast manual trial planning function in half the human-in-the-loop runs. During the other half of the runs participants could also interact with the advanced airspace concepts’ conflict resolution algorithm developed at NASA Ames Research Center. The controllers could use the algorithm to request a conflict resolution trajectory and uplink it unchanged, modify the resolution trajectory using the trial planner and then uplink it, or cancel the modification. The analysis shows that in the simulated part-task setting the 2x density was very manageable and adequate for the interactive mode and was somewhat manageable with a manual trial planner. 3x density was somewhat manageable interactively, but would not pass the “spill your coffee” test. 3x density was unmanageable using the trial planning function alone and resulted in many separation violations. Objective and subjective data indicate that the conflict resolution automation was highly acceptable across traffic levels and provided efficient trajectory changes at high traffic densities. The conflict resolution automation reduced the average and the peak workload significantly. In this paper the problem, method, and key findings will be presented. This research is an initial step towards addressing the challenges associated with human/system integration for ground-based NextGen separation assurance and was funded by the Separation Assurance element of NASA’s Next Generation Air Transportation – Airspace Project.