Optimal Tuning of Altitude Controller Parameters of Unmanned Aerial Vehicle Using Iterative Learning Approach

Abstract

Dynamics and flight stabilization of a multirotor unmanned aerial vehicle (UAV) can be shaped by appropriate mechanisms of tuning parameters of its position and orientation controllers. In the article, the attention is focused on a fixed-parameters altitude controller. Its gains can be tuned optimally and automatically according to the expected criterion, and the search process takes place during the UAV short-time flight. For this purpose, it is proposed to use the auto-tuning method based on the bootstrapping technique and zero-order optimization using Fibonacci-search algorithm. The theoretical basis of the proposed method and discussion of the results from conducted simulation experiments for the exemplary quadrotor model, are presented in the paper.