In this project, an MPC algorithm based on feedback linearization is devised to facilitate trajectory tracking for a quadcopter. The application of feedback linearization to the nonlinear model of the quadcopter results in a transformation, rendering the model linear. This linearized system serves as the basis for subsequent controller design. Leveraging feedback linearization allows the utilization of linear control schemes for quadcopter control. Subsequently, a model predictive controller is formulated for the linearized quadcopter model, initially overlooking external disturbances. To address external disturbances, a disturbance observer is developed to estimate these disturbances, ensuring the estimation error is bounded-input, bounded-output (BIBO) stable, thereby compensating for the impact of external disturbances.
Research Paper referenced: https://ieeexplore.ieee.org/document/9642982