Trajectory Prediction and Alerting for Aircraft Mode and Energy State Awareness

Abstract

This paper describes the implementation and evaluation of technologies that predict and assess the future aircraft energy state and autoflight configuration, and provide appropriate alerting to better inform pilots of the effect of problematic autoflight inputs or conditions. Prediction algorithms are used to extrapolate the current state of the aircraft based on flight management, autopilot and autothrottle system control laws and knowledge of mode transition logic. The resulting predicted trajectory represents the future four-dimensional flight path of the aircraft if the current course of action is continued. Probabilistic methods are used to estimate the trim envelope through high-fidelity model-based computation of attainable equilibrium sets. The corresponding maneuverability limitations of the aircraft are determined through a robust reachability analysis (relative to the trim envelope) through an optimal control formulation. The combination of prediction and assessment technologies are used to trigger timely alerts to avoid loss of control situations. The maneuvering envelope limits are also indicated on the primary flight display, and the predicted trajectory is displayed on navigation and vertical situation displays. The display features and alerts were evaluated in the Advanced Concepts Flight Simulator at NASA Ames Research Center, where commercial airline crews flew multiple problematic approach and landings scenarios to investigate the impact on current and future aircraft energy state awareness. Results show that the display features and alerts have the potential to improve situational awareness of what the automation is doing now and what it will do in the future.