Design and Real-Time Implementation of a Cascaded Model Predictive Control Architecture for Unmanned Aerial Vehicles

Abstract

Modeling and control are challenging in unmanned aerial vehicles, especially in quadrotors where there exists high coupling between the position and the orientation dynamics. In simulations, conventional control strategies such as the use of a proportional–integral–derivative (PID) controller under different configurations are typically employed due to their simplicity and ease of design. However, linear assumptions have to be made, which turns into poor performance for practical applications on unmanned aerial vehicles (UAVs). This paper designs and implements a hierarchical cascaded model predictive control (MPC) for three-dimensional trajectory tracking using a quadrotor platform. The overall system consists of two stages: the mission server and the commander stabilizer. Different from existing works, the heavy computational burden is managed by decomposing the overall MPC strategy into two different schemes. The first scheme controls the translational displacements while the second scheme regulates the rotational movements of the quadrotor. For validation, the performance of the proposed controller is compared against that of a proportional–integral–velocity (PIV) controller taken from the literature. Here, real-world experiments for tracking helicoidal and lemniscate trajectories are implemented, while for regulation, an extreme wind disturbance is applied. The experimental results show that the proposed controller outperforms the PIV controller, presenting less signal effort fluctuations, especially in terms of rejecting external wind disturbances.