Autonomous zero-visibility quadrotor landings towards persistent ship-based UAV ocean observations

Unmanned aerial vehicles are increasing their role in ocean observations. They provide complementary measurements to traditional in-water sensors resulting in richer information extraction and better insight into ocean processes (e.g. monitoring of ocean temperatures). An enabling capability to enhance the role of unmanned aerial vehicles in ocean observations, based off ships, is autonomous landing in poor (or zero) visibility and sea states. The intended application is lower atmosphere ocean observation. The authors' earlier work presented a solution to zero visibility landing in lower sea states (3 or less). This paper reports on the development, implementation and validation of an algorithm to predict when to land on the ship given high sea states. The accuracy of the predictor is quite good even in high sea states. Future work involves testing the algorithm at sea in a sea state.