Maritime Autonomous Surface Ships

The maritime industry faces many pressing challenges due to increasing environmental and safety regulations and crew safety concerns. In light of these challenges, autonomous ships can provide potential solutions for addressing smart shipping, fuel efficiency, and safety issues. The development of marine autonomy technology will significantly improve the situation and is expected to become a cost-efficient alternative to conventional ships. Currently, automated shipping technology is rapidly transitioning from theoretical to practical applications as the number and scope of autonomous ship prototypes increase around the globe. These prototypes are widely used in both navy and commercial applications, such as ocean observation, coast patrol, underwater monitoring, and underwater production system operation, to name just a few.

The main goal of this book is to address key challenges, thereby promoting research on Maritime Autonomous Surface Ships (MASS). Firstly, one review paper on developing Digital Twin (DT) technology in the maritime domain is provided [1]. The following topics on autonomous surface ships are included in this book: methods of ship control [2,3,4,5], collision avoidance [6,7], ship detection methods [8,9], and manoeuvring models [10].

Madusanka et al. [1] reviewed the development of Digital Twin (DT) technology and its applications within the maritime domain, mainly surface ships. In [2], a finite-time, active fault-tolerant control (AFTC) method was proposed for autonomous surface vehicles, and the framework was based on an integrated design of fault detection (FD), fault estimation (FE), and controller reconfiguration. Simulation tests using the CyberShip II were carried out to validate the proposed AFTC method. In [3], the authors studied collision avoidance in the formation-containment tracking control of multi-USVs with constrained velocity and propulsion forces. A multi-USV formation-containment tracking control strategy was designed based on a dual-layer control framework, and stability was validated using the Lyapunov method. In [4], the cooperative formation trajectory tracking problem for heterogeneous unmanned aerial vehicles (UAV) and multiple unmanned surface vessels (USV) was invested in this paper, and a simulation study is provided to show the efficacy of the proposed strategy. To provide a higher lift force and improve the seakeeping performance of a ship, a control method for the T-foil’s swinging angle is established and optimized based on model testing [5]. The results obtained by model testing show that T-foils with pitch angular velocity control can decrease the vertical motion response in the resonance region of a ship’s encounter frequency and increase the anti-bow acceleration effect.

Another important topic for autonomous surface ships is collision avoidance, since the ship must be able to avoid unexpected obstacles. Niu et al. [6] proposed a multi-ship autonomous collision avoidance decision-making algorithm using a data-driven method and adopted the Multi-agent Deep Reinforcement Learning (MADRL) framework. The 40 encounter scenarios were designed to verify the proposed algorithm, and the results show that this algorithm can efficiently make a ship collision avoidance decision in compliance with COLREGs. Hwang et al. [7] proposed a method for analyzing collision risk situations extracted from AIS data through graph-based modeling and establishing validation scenarios. Yasir et al. [8] presented a survey of AI- and ML-based techniques for ship detection in SAR images that provide a more effective and reliable way to detect and classify ships in various weather conditions, both onshore and offshore. In [9], a new attitude-estimation framework was proposed to extract the geometric features using point clouds from shipborne LiDAR and compute the attitude of the target ships. The experimental results demonstrated the filtering ability and practical applicability of the proposed method in real water-pool experiments under real environmental noises. In [10], a data-driven method, the truncated LS-SVM, was proposed for estimating the nondimensional hydrodynamic coefficients of a maneuvering model. The results demonstrate that the truncated LS-SVM method effectively models the hydrodynamic force prediction problems with an extensive training set, reducing parameter uncertainty and yielding more convincing results.