Adaptive repetitive learning control for an offshore boom crane

This paper proposes an efficient nonlinear controller for an offshore boom crane, which is a combination of a learning strategy and an adaptive robust control method. An offshore boom crane is a kind of typical underactuated system which has less number of actuators than its degrees of freedom (DOFs), and it is also a sophisticated nonlinear system with strong-coupling characteristics, therefore, controller design for this kind of system becomes an extremely challenging task. Moreover, different from an overhead crane fixed on land, an offshore boom crane fixed on a vessel suffers from some peculiar disturbances of the attached ship's multi-dimensional movement induced by waves and ocean currents, which implies that the motion of the ship can cause a tremendous effect on this system. Considering the periodic property of sea waves, this paper proposes an adaptive repetitive learning control strategy containing a learning law to deal with the aforementioned practical problems, as well as an adaptive law to handle the unknown system parameters. Specifically, different from traditional learning control strategy, the proposed control algorithm successfully addresses unknown periods of the periodic disturbances by introducing a period identifier into the control scheme. Meanwhile, the developed control strategy presents good robustness against ever-lasting disturbances and unknown parameters. The stability of the designed closed-loop system is guaranteed in the Lyapunov sense. Furthermore, some comparative experimental results are presented to demonstrate the efficiency of the proposed control method.