Stochastic Longitudinal Autopilot Tuning for Best Autonomous Flight Performance of a Morphing Decacopter

Abstract

In this conference paper autonomous flight performance maximization of a morphing decacopter is considered by using stochastic optimization approach. For flight controller a PID based hierarchical control system is chosen. In this paper PID controller which is used for pitch angle is considered. In this application only longitudinal flight and longitudinal autopilot is considered where the pitch motion is in primary interest and the used controller is the pitch control. For optimization technique simultaneous perturbation stochastic approximation (i.e., SPSA) is selected. It is fast and safe in stochastic optimization problems when it is not possible to evaluate gradient analytically. At the end a cost function consisting terms that settling time, rise time and overshoot is minimized. A detailed graphical analysis is done in order to better present effect of morphing on longitudinal flight of decacopter flight. Moreover, the cost function consist of rise time, settling time, and overshoot during trajectory tracking.