Modeling and PID controller design of a four rotor unmanned aerial vehicle

In our era, unmanned aerial vehicles have a widespread usage area. Unmanned aerial vehicles are used in search and rescue, fire fighting, cargo transportation, agricultural spraying, mapping, scientific research and defense industries. Due to unmanned aerial vehicles do not have a pilot on board, they could perform maneuvers that may be harsh and dangerous for manned aircraft. Moreover, the lower production costs of unmanned aircraft makes them more advantageous than manned aircraft. Unmanned aerial vehicles with four rotors have become popular thanks to their vertical landing and take-off, hanging in the air, hard maneuverability, flying in open and closed areas and their small size. The dynamics of the quadrotor makes possible to move in 6 DOF. It consists of 3 translational and 3 rotational DOF. Euler angles, which also are known as roll (φ), pitch (θ), yaw (ψ) angles, and x, y and z axes, which represent linear movements in 3-dimensional space, constitute 6 degrees of freedom. The rotors of quadrotor could be controlled separately, allowing these movements to be performed. In this thesis, first of all, nonlinear mathematical model of a quadrotor has been extracted. In the next step, this mathematical model was implemented in Simulink. The real physical parameters of a quadrotor used in system. Different controllers are designed for the height, roll angle, pitch angle and yaw angle of the quadrotor. In this work, PID controller design is used. Designed controllers are added to the system in Simulink. In the PID controller, the signal from the output is compared with the reference signal and an error occurs from its difference. PID controller tries to minimize the error. The error is detected by continuous feedback from output to input until the error is minimized and the error is decreased by sending the controller effect to the output. The PID controller has three gain parameters: proportional constant Kp, integral constant Ki and derivative constant Kd. Proportional constant decreases the rise time but does not remove the steady state error. The integral gain reduces the steady state error, but makes the transient response poor. Derivative constant reduces the overshoot and increases the stability of system. In this work, different PID controllers are designed for different reference signals.Controllers designed for altitude and attitude have been found to follow different reference values successfully. In addition, PID Tuning method was used to design the controllers. In this method, PID coefficients are determined automatically by Matlab. However, it has been found that the controllers identified by this method are incompatible with the physical parameters of the quadrotor because the maximum torque that the rotors could produce is exceeded. In this thesis, the performance of controllers under parameter uncertainty was also examined. The disturbances includes parameter uncertainties in the quadrotor mass (m) and moments of inertia (Ix, Iy and Iz). It was observed that designed PID controllers successfully tracked unit step and sinusoidal reference values under parameter uncertainty. The noise resistance of the designed controllers is also examined. The performance of the controllers was examined by giving noise to the inputs and outputs of the system separately. As a result, the designed controllers were found to be noise resistant. Parameter uncertainty and noise were also applied to controllers designed with PID Tuning. However, the controllers we designed were more resistant to parameter uncertainty and noise than the controllers designed with PID Tuning.