Investigation of Linear Quadratic Control and Guidance System Design for a small UAV

Autonomous air vehicles that are effcient and robust require reliable control systems that enable them fly a mission. Modern ight controllers such as the linear quadratic regulators provide optimal solutions for obtaining control gains, however this approach requires the selection of weighting matrices which often entail time consuming trial and error process. A multi-objective particle swarm optimization scheme is developed to select weighted matrices based on the output performance specification. Simulation results of the application of this method for an UAV lateral-axis model provided an output response that was within the specification limits. Furthermore the rapid design and development of the autonomous air vehicle requires a fairly accurate simulation model which in this case for the first time was obtained for the GULMA air vehicle using the Athena Vortex Lattice method that gives aerodynamic coeficients and stability derivatives as well as a linear model. Model-based controller synthesization was conducted using this linear model to test for robust stability and performance within specified percentage of uncertainty. The results were within stable margins for u high order controllers. Finally a guidance system based on the Lyapunov Vector Field was applied to the airborne test platform for tracking of straight line and circular paths in the presence of wind disturbance. The simulated results provided satisfactory results that would be considered for real-time flight tests.