Performance criterion sensitivity analysis of ship‐steering models with respect to shaping filter design parameters

Abstract

AbstractThe design of linear‐quadratic controllers for the steering control of ships requires the design of sway force and yaw moment disturbance shaping filters to adequately represent the statistical properties of the sea‐way. The estimated disturbance states are employed in the formulation of the optimal control law to improve the steering quality to minimize propulsion losses. In this paper a sensitivity analysis of the optimal cost criteria with respect to the shaping filter parameters is made to determine the relative importance of the filters and parameters with differing ships, cost criteria and sea‐way conditions. For simplicity of the present analysis the Kalman filter is excluded from the dynamics, i.e. all the states are assumed to be perfectly measurable. This allows the analysis to be done at the upper bound of optimal performance. The percentage deviation in the resulting optimal propulsion loss from a known deviation in the parameters is calculated. Sensitivities with respect to two different controllers, each minimizing a different performance criterion are analysed for two ship types. The results indicate that the yaw shaping filters are relatively more important for both of the ship types. It is also shown, however, that the sway shaping filter, although having no significant effect on the optimal performance of ship types such as the large tanker examined, does affect that of fine‐lined high‐speed vessels such as the container ship analysed.