Synchronized Digital Twin Prediction and Decision Support for Six-Degree-of-Freedom Nonlinear Aircraft Landing

Abstract

A synchronized digital-twin is used to provide decision support for nonlinear six-degree-of-freedom aircraft landing under wind-induced plant–twin mismatch. A nominal landing trajectory is first generated offline by solving a constrained nonlinear trajectory optimization problem with an imperfect physics-based aircraft model. The reference is then tracked on a higher-fidelity plant proxy using a closed loop tracking controller. During execution, wind disturbances, actuator lag, and initial-condition errors cause the plant trajectory to deviate from the nominal prediction. An Extended Kalman Filter is used as the data assimilation method to estimate a reduced local pose-and-wind state from the measured aircraft response. At each update time, the digital twin uses the current synchronized state to initialize a set of candidate terminal control modes that modify the tracking controller near touchdown, and evaluates each one on how it improves landing metrics. Simulation results shows that digital twin support can provide real time prediction of future aircraft states by propagating the nonlinear aircraft dynamics forward, and intervene ahead of time to reduce harsh terminal conditions.