Real-Time Operator-in-the-Loop Anti-Sway Control for Shipboard Cranes

Shipboard cranes play an important role in many maritime operations, however unexpected payload sway due to the motion of the ship or external disturbances creates a hazardous environment for deck personnel. To improve deck safety, the development of anti-sway control systems to reduce undesired payload motion has been of great interest to researchers; however, existing literature has focused primarily on complex nonlinear control strategies that require highly simplified dynamic systems, and the consideration of anti-sway control on high-fidelity, high-degree-of-freedom (DOF) crane/ship dynamic systems is lacking. Additionally, anti-sway control systems in the literature are typically developed with little consideration of real-time operator-in-the-loop interaction, and rarely are the human factors of shipboard crane anti-sway control seriously considered. Therefore, to address the deficiencies in the literature this thesis presents the development of a real-time anti-sway control system that can be applied to high-DOF shipboard cranes, developed with consideration of an operator-in-the-loop and tested in a virtual reality (VR) human factors study. To provide complete motion compensation, the anti-sway system can switch smoothly between deck-frame compensation, where the payload is held stationary with respect to the ship deck, and world-frame compensation, where the payload is held stationary with respect to the ocean's surface. To provide intuitive control, the operator is only required to provide the desired Cartesian position of the payload; as part of the anti-sway system, a nonlinear trajectory optimizer maps the Cartesian position to the crane actuators. To demonstrate its performance, the anti-sway system is validated in high-fidelity simulations with a five-DOF shipboard gantry crane, as well as a six-DOF, seven-DOF and nine-DOF shipboard knuckle boom crane, each with dynamics of increasingly complexity, including a double-pendulum between the sheave, hook and payload. The anti-sway system is enhanced iteratively with each crane, showing a reduction in payload tracking error of 73%, 84%, 92%, and 93%, respectively. Finally, the anti-sway system is tested in a real-time VR simulator, where it is demonstrated through a human factors study that the anti-sway system is both effective at reducing payload tracking error and intuitive for operators to use.