Mazari UAV Adaptive Autopilot System Design and Implementation

Abstract

Autonomous unmanned aerial vehicle (UAV) research started at AUS in 2005. Linear and nonlinear aircraft models and system identification of the aircraft system were developed to design flight control laws. Proportional-integral-derivative control laws as well as fuzzy logic supervisory control for fixed wing autopilots in cruise mode were developed. Hardware in the loop for UAV flight simulation was also developed. The current work focuses on designing control laws for takeoff, cruise, and landing, and conducting flight test verification of these law. Implementation of control laws for different flight modes need further study in switching between flight modes. A hybrid control scheme, which acts as a supervisory control for coordinating the conventional control laws for each flight mode is proposed. This involves the usage of a state-machine block to coordinate the transition of each of the flight modes. A way of using fuzzy logic control to map the gain values for different flight equilibrium states is developed and simulated in this work. The contributions of this thesis are the development of digital flight control system software and hardware. The autopilot hardware is designed around an MPC 555 single-board computer. Modem configuration and data consistency checks are performed inside ground station and autopilot hardware to ensure there is no delay of communications, and the software is designed inside the SIMULINK environment. Data reception, data processing, command generation, and data transmission are the main functions inside the software. PI control loops and fuzzy logic scheduling control are developed, simulated, and tested as part of the implementation of all phases of the flight in a real flight test.