Optimal PID Tuning Methodology for Quad-Copter Unmanned Aerial Systems

Abstract

Quadcopters are the most widely used multirotors Unmanned Arial Vehicles (UAVs). In addition, the control of UAVs is very crucial as it faces several challenges such as: difficulties in obtaining aerodynamic parameters and dynamic coupling. Furthermore, Proportional, Integral and Derivative (PID) controller is widely used in industry due to its practicality and ease of use. However, tuning its parameters is very time consuming since it depends on trial and error methods. The focus of this research is enhancing the tuning process of PID controller by using full state feedback Linear Quadratic Regulator (LQR) to obtain the PID gains. optimal PID tuning methods combine the ease of PID controller and the robtuness of LQR with energy and time considerations. In this thesis, the linearized dynamic model was adopted from literature. Different linear controllers, such as: LQR, Linear Quadratic Gaussian (LQG) and PID, were implemented on the hovering, yawing and the combined (hovering and yawing) modes of a quadcopter in plus configuration. Moreover, two optimal PID tuning methods are presented. The first method is an improvement on a pre-existing optimal PID tuning method. While the second method is a novel analytical tuning methodology for PID gains. The simulation results show the effectiveness of both methods. However, the second method outperforms the first method in terms of application, performance and computational time. In addition, different matrix inverse methods and newly developed Riccati solution were adopted and compared to pre-existing methods in terms of computational time. The optimal PID tuning methods were validated by comparison to experimental data available in the literature. Indexing Terms: PID, LQR, Optimal PID, LQG, Quadcopter, Riccati, Hovering, Yawing, Combined.

Date of Award	2018
Original language	American English
Supervisor	Yahya Zweiri (Supervisor)