Development of Pitch Control Subsystem of Autopilot for a Fixed Wing Unmanned Aerial Vehicle

Autopilot is an augmented flight control system that is used to reduce the workload of the pilot during long endurance missions. Development of autopilots for Unmanned Aerial Vehicles (UAVs) also needs such a control system before building it for pilotless operation. Presently, the autopilot for fixed wing aircraft is dominated by classical PID controller. However, state space methods provide systematic and more robust control system design. Hence, state space method is considered for an UAV, named as Casper, designed and ground tested at authors University. This paper presents modeling and simulation aspects of UAV and development of pitch control subsystem of autopilot with state space methods. Simulink environment is used with inbuilt aerospace block- sets for flight dynamics and Digital DATCOM to obtain aerodynamic characteristics of UAV. FlightGear simulator is interfaced with Simulink model for visualization of UAV response. Linear Quadratic Regulator (LQR) approach is used in state space method to determine state feedback gain matrix. Comparison of state feedback and PID for pitch control autopilot shows, state feedback controller has better "command following performance", as it has lesser over shoot of 75%, lower settling time by 0.5 s and 55% lesser controller effort. Response to gust loads of state feedback and PID autopilot are comparatively same. However, transient performance of PID controller is smoother and settling time is 75 s lesser than that is achieved by state feedback controller.