Landing drones in harsh winds and turbulent open waters poses its own unique challenges. Landing them on an unmanned surface vehicle can cause damage due to the severe roll and pitch angle of the USV. Parakh M. Gupta, Eric Pairet, Tiago Nascimento, and Martin Saska have come up with a novel model predictive control approach to enable UAVs to land autonomously on a USV in harsh conditions. As the researchers explain:
The MPC employs a novel objective function and an online decomposition of the oscillatory motion of the vessel in order to predict, attempt, and accomplish the landing during near-zero tilt of the landing platform. The nonlinear prediction of the motion of the vessel is performed using visual data from an onboard camera.