Research On Nonlinear Control Methods For A Micro-quadrotor

The concept of quadrotor has been given since the year of 1907. But due to the technology limitation and lack of control methods, those designed vehicles tended to be unwieldy and had poor stability. Recently with the technological development in sensors, actuators, processors and power storage devices, the research on the quadrotor has allured the researchers again and made a great progress, especially in the realm of micro-quadrotor.Micro-quadrotor is an electric vertical take-off and landing (VTOL) unmanned aerial vehicle, which has broad applications in both military and civilian region. For example, it can be used for the reconnaissance and surveillance in the ground battlefield, or the search and rescue after serious disasters. As a 6 DOF vehicle, micro-quadrotor is a kind of underactuated system with 4 motor drives as the inputs, so its manipulation is very flexible, having its own characteristics in control.This thesis mainly studies the nonlinear control methods of the micro-quadrotor with respect to its unique performance on the basis of the overviews about the current research situation, key technologies and great application prospects. The main research contents are as follows.Firstly, a nonlinear dynamic MIMO model of an underactuated quadrotor is derived based on Newton-Euler equations with the coordinate frames defined to describe the movement of micro-quadrotor. And the system equations are transformed to affine nonlinear form.Secondly, the model is simplified to meet the needs of research and a control scheme is proposed with the underactuated characteristic of the quadrotor analyzed. A flight controller based on the Backstepping is designed for both hovering and tracking, which can set the micro-quadrotor track their desired trajectories and stabilize the pitch and roll angles. Simulations show that the proposed controllers are validity.Thirdly, the robust control problem of attitude tracking is investigated. At part one, a backstepping sliding mode controller, which combined advantages of the backstepping control and sliding mode control, is proposed, and the effectiveness of the controller is proved by the simulation. At part two, to enhance the convergence speed of the attitude and acquire global robustness, a terminal sliding mode controller is designed. This controller can guarantee the tracking error converges to zero in finite time which can be set arbitrarily. In addition, the reaching phase is totally eliminated wherever an initial state is located in the phase space, so the overall system can show invariance to the disturbance and uncertainty. The final simulation results show that the controller can meet the design requirements.Concluding remarks are presented at the end of this dissertation. Some open problems are also pointed out, which deserve further study.