Design of a Non-linear Controller for a Multi-Rotor UAV

Abstract

Multi-rotor unmanned aerial vehicle (MUAV) has gathered a lot of interest from the research community owing to its simple structure, easy maneuverability and ability to take-off and land vertically. These characteristics have immensely increased the chances for MUAV to become a potential candidate for various applications for which traditional UAVs proved ineffective. Major challenges highlighted by literature survey are un-modeled dynamics estimation and exploration of robust techniques for controlingthe flight. This work is aimed to develop an adaptive sliding mode controller for a quadrotor. In the formulation of a control problem there are inevitable diffrences between real world plant and the mathematical equations used for design of controller. Adaptive sliding mode controller tries to overcome this mismatch. The proposed technique is demonstrated by applying it to a simulink model of a commercially available mini drone i.e. Parrot mambo. The effectiveness of the aforementioned technique in the presence of wind disturbance is evaluated using simulations. Developed algorithm is also demonstrated by deployment on actual hardware. The proposed techniques provide excellent guidance for the controller design of other MUAV systems and also for other nonlinear under actuated systems.