For the complex multi-ship encounter scenarios, this article proposes a dynamic collision avoidance path planning algorithm based on the A-star algorithm and ship navigation rules, namely Dynamic Anti-collision A-star (DAA-star) algorithm. A dynamic search mechanism of the DAA-star algorithm considering time factors is designed to enable the collision avoidance for situations with known moving obstacles. A quaternion ship domain is generated based on Automatic Identification System (AIS) data, and the navigation risk cost is calculated with the combination of the quaternion ship domain and potential field. The searching constraints conforming with the Regulations for Preventing Collision at Sea (COLREGS) rules are set for the DAA-star algorithm to guarantee the safety of collision avoidance. Meanwhile, the individual ship maneuverability constraints and maneuverability differences from ship to ship are both considered in the proposed DAA-star algorithm, which can solve the path planning problem with dynamic obstacles in multi-ship encounter scenarios. The simulation results show that, compared with the traditional A-star algorithm and dynamic A-star algorithm, the DAA-star algorithm can generate more reasonable dynamic and static obstacle avoidance paths in complex navigation scenarios in the trade-off between the navigation risk and economical efficiency.

Since vessel dynamics could vary during maneuvering because of load changes, speed changing, environmental disturbances, aging of mechanism, etc., the performance of model-based path following control may be degraded if the controller uses the same motion model all the time. This article proposes an adaptive path following control method based on least squares support vector machines (LS-SVM) to deal with parameter changes of the motion model. The path following controller consists of two components: the online identification of varying parameters and model predictive control (MPC) using the adaptively identified models. For the online parameter identification, an improved online LS-SVM identification method is proposed based on weighted LS-SVM. Specifically, the objective function of LS-SVM is modified to decrease the errors of parameter estimation, an index is proposed to detect the possible model changes, which speeds up the rate of parameter convergence, and the sliding data window strategy is used to realize the online identification. MPC is combined with the line-of-sight guidance to track straight line reference paths. Finally, case studies are conducted to verify the effectiveness of the proposed path following adaptive controller. Typical parameter varying scenarios, such as rudder aging, current variations and changes of the maneuverability are considered. Simulation results show that the proposed method can handle the above situations effectively.

Real-time collision avoidance with full consideration of ship maneuverability, collision risks and International Regulations for Preventing Collisions at Sea (COLREGs) is difficult in multi-ship encounters. To deal with this problem, a novel method is proposed based on model predictive control (MPC), an improved Q-learning beetle swarm antenna search (I-Q-BSAS) algorithm and neural networks. The main idea of this method is to use a neural network to approximate an inverse model based on decisions made with MPC for collision avoidance. Firstly, the predictive collision avoidance strategy is established following the MPC concept incorporating an I-Q-BSAS algorithm to solve the optimization problem. Meanwhile, the relative collision motion states in typical encounters are collected for training an inverse neural network model, which is used as an approximated optimal policy of MPC. Moreover, to deal with uncertain dynamics, the obtained policy is reinforced by long-term retraining based on an aggregation of on-policy and off-policy data. Ship collision avoidance in multi-ship encounters can be achieved by weighting the outputs of the neural network model with respect to different target ships. Simulation experiments under several typical and multi-ship encounters are carried out using the KVLCC2 ship model to verify the effectiveness of the proposed method.

Underactuated autonomous surface vehicles (ASVs) have stringent requirements on automatically tracking a predefined path. This paper proposes a model predictive control (MPC) approach based on adaptive line-of-sight (LOS) guidance for path following of ASVs. For the controller, a second-order nonlinear Nomoto model with disturbances is proposed as the vessel dynamic motion model after reviewing and comparing different ship motion models applied for path following control. For the guidance system, a novel adaptive LOS guidance with a variable acceptance circle radius is proposed to improve the precision of reference path tracking. Specifically, the acceptance circle radius is adapted with the angle between two adjacent straight segments of a reference path. Simulation experiments illustrate that the LOS guidance system with a variable acceptance circle radius results in smaller tracking errors compared with the fixed acceptance circle radius. The proposed path following method can track reference paths well even in the face of disturbances.

Motion control in absence of human involvement is difficult to realize for autonomous vessels because there usually exist environmental disturbances and unmeasurable states at the same time. A discrete-time model predictive control (MPC) approach based on a state-compensation extended state observer (SCESO) is proposed to achieve more precise control performance with state estimations and disturbance rejections simultaneously. The main idea is that lumped disturbances encompassing nonlinear dynamics and external disturbances are handled as two parts, unlike the standard extended state observer (ESO). Particularly, the nonlinear terms are compensated by estimated states and the external disturbances are considered as extended states and attenuated by the traditional ESO strategy. Assuming that the lumped disturbances are constant over the prediction horizon, the prediction model is linearized to save computational time since after linearization the online MPC optimization problems are solved as quadratic programming problems instead of nonlinear programming problems. The convergence of the proposed SCESO estimation errors to zero is proved even when the disturbances keep variable. Two case studies involving a numerical example and ship heading control have been conducted to verify the effectiveness of the proposed control method.