Abstract
Path following of underactuated marine vessels is a fundamental marine practice in shipping industry. However, the collision avoidance, which is frequently encountered during the process of path following of ships sailing in seaways, is neglected in traditional studies of path following. In this paper, a novel control design for path following with auxiliary system for collision avoidance is presented. Taking advantage of the capability of dealing with multi-variable system with the constraints, the model predictive control (MPC) method is employed to deal with the input saturation (rudder) and underactuated problem. Furthermore, the parallel computational nature of projection neural network (PNN) is included to reduce the computational burden of traditional MPC technique and make the control design more efficient. Simulations are conducted to validate the effectiveness and efficiency of the proposed control design.
Projection neural network for a class of sparse regression problems with cardinality penalty
