This paper investigates Dynamic Positioning (DP) of an vessel next to a larger vessel subject to unknown wind and wave shielding effects in time-varying ocean environments. For safe operations, the vessels need to maintain a distance from each other which can be formulated as constrained control. Under certain conditions, the larger vessel shields the smaller vessel, and when the conditions change or due to the change in heading, the shielding reduces. For the DP system, in order to detect the shielding effect, and monitor the variation of environmental forcing on the vessel, an environment condition detector is designed based on low-frequency force learning and adaptive neural network. In the detector, estimates of wind drag coefficients are used to determine whether or not the shielding effect is present when upper thresholds are crossed. In this case, the neural network is activated in the detector and control when the small vessel moves out from the shadow of the large vessel. Prescribed performance control is proposed to meet the side-by-side safe constraint leveraging the H infinity approach to guarantee the vessel can keep its position against the exogenous disturbance. Numerical simulations demonstrate the effectiveness of the proposed design.(c) 2022 European Control Association. Published by Elsevier Ltd. All rights reserved.
展开▼
