Comparing and developing pid and sliding mode controllers for quadcopter

Unmanned Aerial Vehicles are aircrafts without a pilot on board. Due to not being restricted with the limitations of human pilots, unmanned aerial vehicles can enter environments which are hazardous to human life. UAVs have also become increasingly popular because of their wide range of applications. Search and rescue, transportation, surveillance, aerial imagery, fire monitoring, law and enforcement, commercial and journalism, military and defense, surveying and crop spraying, reconnaissance, scientific research are some of the areas which UAVs can be used. Recently an aerial vehicle named quadcopter (or quadrotor) became popular in unmanned aerial vehicle design. Due to their small size and agile maneuverability, quadcopters can be flown indoors as well as outdoors. Their four rotors can be controlled separately giving them capability to move rotational and translational. In controller designs disturbances, dynamics, plant parameters causes discrepancies between the actual system and mathematical model. To solve this problem, robust control methods are developed. Sliding mode controller is the one of the robust control design approaches. The purpose of the sliding mode controller is keeping plant at desired predefined sliding surface. Another controller approach proportional integral derivative compares the desired value with the present value and calculates the difference as error. When the variation of error is considered; proportional control depends on present error, integral control depends on past errors and derivative controller predicts the future errors. This thesis focuses on developing sliding mode controller and proportional integral derivative controller and comparing these two controller approaches. Controller design is made for two common types of configuration (plus configuration and cross configuration). Fundamentally, altitude and the angles caused by quadcopters' rotational movement (roll, pitch and yaw angles) are controlled. Coefficients are adjusted by iteration method. Additionally chattering problem caused by the fast dynamics which are neglected by model causing fast switching of sliding mode controllers is solved. For translational movement on x and y axis proportional integral derivative controller is used. As a result, controllers are capable of tracking the desired values. It can be said that the sliding mode controllers can track the desired values with faster response and less oscillation than the proportional integral derivative controllers.