Frequent bad weather at sea severely damages the quality of visual images captured by imaging equipment. Ship instance segmentation in adverse weather conditions remains a major challenge because of poor visibility at sea. Existing approaches for instance segmentation are primarily designed for clear days and rarely consider the aforementioned severe weather. Blurred ship objects can easily cause missed ship detection and decrease the instance segmentation performance on ship images, especially in the case of frequent fog at sea. To this end, we propose a ship instance segmentation framework (IRDCLNet) based on Interference Reduction and Dynamic Contour Learning in foggy scenes. The Interference Reduction Module is proposed to reduce the interference caused by fog and solves the problem of missed ship detection. Meanwhile, we present Dynamic Contour Learning to predict the overall contour of the blurred ships to assist in mask prediction. To handle the scarcity of ocean data in foggy weather, we build the Foggy ShipInsseg dataset, which contains 5,739 real and simulated foggy ship images with 10,900 fine instance mask annotations. Experiments on the Foggy ShipInsseg dataset show that our IRDCLNet outperforms the Mask R-CNN and CondInst baselines and achieves the state-of-the-art performance.

In maritime scenes, instance segmentation of small object ships is of vital importance. Small ship objects in images have the characteristics of smaller size, lower image cover rate and fewer appearance features. However, existing instance segmentation methods fail to recognize and segment them and can cause missed ship segmentation. To this end, we propose a dual-branch activation network (DANet) for small object instance segmentation of ship images. DANet consists of a Feature Encoding, a Dual Mask Branch, and a Dual Activation Branch. The Feature Encoding adopts feature refinement and a pyramid structure to obtain more fine-grained features. The proposed Dual Mask Branch extracts dual-path mask features for encoding small object information. We propose a Dual Activation Branch to activate more small object regions and generate instance features. Furthermore, we build the Small ShipInsSeg dataset from a total of 5,256 images and 11,612 instances. The experiments show that DANet outperforms the SparseNet baseline and achieves state-of-the-art performance. Additionally, our method achieves a good trade-off between accuracy and speed.

This paper solves the path following problem of unmanned surface vessels, where model uncertainties, external environmental disturbances, and input saturation are considered. Specifically, the fixed-time line of sight (FXTLOS) based on fixed-time theory is used to design a guidance subsystem where the convergence time is independent of multiple initial states. Given the unavailability of velocity information, novel cascaded extended state observers have been proposed to estimate velocities and unknown disturbances to provide feedback to the controller for better performance, where only position and heading information is required. Furthermore, fast finite-time sliding mode control is used to solve the motion control problem for the USV, where a suitable auxiliary system is designed to deal with the adverse effects of input saturation and to ensure the stability of the system. The stability of the whole system is demonstrated, and the stability of the kinematic and dynamic cascade system is analyzed. Finally, a large number of comparative simulation cases are given to demonstrate the effectiveness and superior performance of the proposed scheme.

Path following and obstacle avoidances are used in heading control and path replanning for unmanned surface vessels (USVs), which have attracted the interest of many researchers over the years. This paper investigates the path-following and obstacle avoidance problems for USVs. First, on the basis of the current position and desired path, an adaptive line-of-sight (ALOS) algorithm is used to obtain the desired heading angle, and the sideslip angle compensation is considered. Then, to ensure that the USV follows the desired path. Model predictive control (MPC) is used to reduce the lateral error. The event-triggered mechanism (ETM) strategy is utilized to reduce the computational cost of MPC. Moreover, to obtain accurate state quantities in real time, a linear extended state observer (LESO) is used to counteract the effects of external disturbances and the nonlinear term of the model. Furthermore, an improved obstacle avoidance algorithm based on the geometric relationship is proposed. This method can better ensure USV navigation safety and reduce consumption and computation. Lastly, multiple simulation experiments illustrate that the algorithm improves the path-following capability and security and ensures smooth input changes by setting input constraints. Therefore, the designed controller has better feasibility and robustness.