Model-Based Optimal H_∞ Controller on the Stability of a 2-DoF Quadrotor

In this work the synthesis of a MIMO (Multiple Input Multiple Output) robust optimal model-based H_∞ controller is proposed. The whole process takes into account the dynamic equations of a 2-DoF quadrotor Mini Aerial Vehicle (MAV) attached to a steel stand. We consider the gamma-iteration algorithm to find the controller. Our analysis focuses on the control of roll and pitch axes, thereby neglecting the yaw axis control. As our goal is, a priori, to observe the behavior of the H_∞ controller while it is controlling the four motors individually in order to stabilize our MAV, this set up provides us with the possibility of a very close overview of the aircraft. Indeed, it allows the easy insertion of disturbances in both axes (individually and simultaneously) and then closely observe the behavior of the platform. Besides, and most important at any laboratory environment, it is an extremely safe mode to run indoor tests, avoiding the quadrotor from causing harm to the crew if any technical problem occurs. The optimal H_∞ robust controller presents a high capability of rejecting noises and disturbances. The controller can also suppress the uncertainties of our model. Besides presenting the dynamical model of our MAV, we present the experimental results of both roll and pitch control using the dSpace™ 1103 high performance controller board to embed the designed H_∞ MIMO controller.