Modelling and Control of Unmanned Aerial Vehicles

Unmanned Aerial Vehicle (UAV) is becoming an important tool for military and civilian operations due to its growing applications. Examples of such applications are delivery, sensing, mapping, surveillance and search and rescue, where the manned mission is hard. The mathematical modelling of the UAV starts with the characterization of its kinematics, dynamics and forces and moments, which generates the 12 nonlinear ordinary differential equations. The aerodynamic flow in the design considers laminar flow in the operating velocity range. Linearization is performed around the equilibrium point, which is also called the trim condition. The trim condition depends on the flight scenario. Linearization generates the state space matrix, where the modelling is carried out with the decoupling of the longitudinal and lateral mode separately. We then proceed with the autopilot design using Linear Quadratic Regulator (LQR) techniques with integral action for both the longitudinal and lateral modes. The core part of the longitudinal is the height and velocity controller and for lateral is the heading controller. Next, we present the guidance for the path following that uses the vector field method with constant for the waypoint and straight line following. Simulation results show the behaviour of the dynamic model. The LQR control with the integral action behaves as expected, which holds the height during the path following at constant altitude. The actuator signal is also shown based on the different values of the path constant. Simulink and MATLAB are the main software tools that were used to illustrate the performance of the resulting closed loop system for different control parameters. This model implementation with the full UAV nonlinear dynamics brings higher accuracy to the performance evaluation and it is suitable to test different controllers and guidance algorithms.