Development of an Open-Source Flight Controller for Rotary Wing UAVs

Due to the growing applications involving Unmanned Aerial Vehicles (UAVs), Flight Controllers for rotary wing UAVs are playing an important role in guaranteeing proper flight performances and high stability. Many companies are investing to improve drones’ reliability and extend their use to several applications ten years from now. Flight performance of drones needs to be improved to meet this increasing demand, and researches about best-performing control strategies have been carrying on. To increase flight performance and control strategies efficiency, an "open" platform is needed, so that the Flight Controller software can be fully accessible and easily modifiable. In the following thesis, an Open-Source Flight Controller for rotary wing UAVs is developed and tested in a real environment, while a Linear Quadratic Regulator (LQR) controller is implemented to perform the auto-stabilizing function and maneuvres. Arduino STM-32 board is chosen as the micro-controller, and it is programmed by using Arduino Integrated Development Environment (IDE). An Inertial Measurement Unit (IMU) is developed to perform sensor fusion between accelerometer and gyroscope sensors. For this purpose, a complementary filter is implemented in the discrete-time domain. Particular attention is given to the software working frequency and the Pulse Width Modulated (PWM) signals generated by the micro-controller: STM-32 interrupt logic and timers documentation are analyzed in detail to avoid delays and provide a fast response to disturbances by motors. The developed software is also enabled to communicate with a GNSS/GPS module, providing the Flight Controller with the longitude and latitude for position control and autonomous flight. Before implementing the control algorithm, a simulation model of the quadcopter is developed in a Simulink environment to preliminary tune the LQR controller. Finally, flight tests are performed, and satisfactory results about pitch, roll, and yaw response to commands demonstrate the capability of the developed Flight Controller to perform maneuvres properly. The LQR controller is found to be a valid alternative to the most common Proportional-Integrative-Derivative (PID) controller, and an efficient open-source platform to improve flight performances of quadcopters is provided.