Application of Nonlinear Model Predictive Control to Quadcopter Equipped with Internal Control System

The object used in this study is an unmanned aerial vehicle (a quadcopter) equipped with an internal control system that has a target position and yaw angle as input commands. Simple internal control systems in quadcopters often show limited performance, especially when quadcopters attempt to track trajectories with large velocities. This study applies a nonlinear model predictive control (NMPC) method to improve the tracking performance of quadcopters equipped with simple internal control systems. The NMPC is introduced because it can take the motion of the quadcopter arising from the internal control system into consideration. This study also handles the control of the yaw angle in addition to the tracking of the position, which is required in some applications, such as aerial photography and surveillance. We evaluated the proposed method through a simulation with the task of tracking a sinusoidal trajectory and facing its direction of movement. In the simulation, the proposed method showed a performance that could not be accomplished using only the internal control system. This result indicates that determining the command with NMPC improves the tracking performance of the position.