ROS-Based Robust and Recursive Optimal Control of Commercial Quadrotors

Abstract

The rapid growth in the Unmanned Aerial Vehicles (UAVs) industry made access to quadrotors easier due to reduced costs and miniaturization. Although it is important to consider parametric uncertainties in the design of robust controllers, this task is seemingly complicated when considering commercial drones, where the model is usually unknown. Therefore, this paper develops a platform based on the Robot Operating System (ROS) where a designed robust and recursive LQR-based controller is implemented. Furthermore, details on the identification of a simplified model are given and fundamental parts of the system are discussed. Finally, experimental results obtained with the help of a Vicon motion capture system successfully validate both the platform and the robust approach in a trajectory tracking task.