Search

With the continuous increase in maritime traffic and navigation tasks, the demand for mission execution in complex maritime environments, such as islands and reefs, has grown. Path planning for Unmanned Surface Vehicles (USVs) in these areas requires overcoming challenges posed by sea wind interference, island topography, and other complex geographical factors, ensuring safe and efficient navigation. This paper aims to delve into the impact of factors such as water currents and wind speed on ship navigation paths and proposes a comprehensive path planning compensation strategy that considers these factors. Firstly, regarding the planned potential field space, directional and dilation factors are introduced to alter the distribution of potential risks, enhancing the safety and reducing potential risks in the planned routes of USV. Secondly, a detailed analysis of different environmental factors is conducted, exploring their impact mechanisms on the performance of USV navigation and planned routes. Based on the analysis, this paper establishes a mathematical model for environmental interference factors, considering the parameters of USV, laying the foundation for modeling environmental issues in global path planning. In terms of path evaluation metrics, this paper defines path length and time, as well as safety indicators for avoiding dangerous areas in the environment, as evaluation criteria. To address the interference of waves and currents in complex water areas with the planned navigation route, this paper proposes a navigation point compensation optimization strategy based on the Fast Sweeping Method (FSM). Combining the calculated estimated drift distance with the planned route comprehensively, a corrective strategy model is proposed by integrating it with the B-spline method. Finally, simulation and emulation were conducted. • An improved path planning algorithm is proposed in this study. • In response to complex navigational water environments and narrow navigation channels, this paper proposes a directional factor. • A novel heading point path compensation is introduced for enhancing the path planning of USVs under the influence of real-world conditions and disturbances like wind and currents. [ABSTRACT FROM AUTHOR]

This paper proposes a novel memory-based event-triggered path-following controller for unmanned surface vehicles (USVs) in the presence of denial of service (DoS) attacks and limited communication resources. In the guidance part, a double distance logical virtual ship (DDLVS) guidance law is developed to eliminate the accumulation error of the yaw angle velocity in the traditional LVS algorithm, and the USV will keep a smoother tracking performance for the curve reference path. In the control part, the memory-based event-triggered fusion control algorithm proposed in this paper incorporates the signal non-transmission phenomenon caused by DoS attacks into the event-triggered interval through a set of adaptive parameters. Consequently, the adverse affect caused by DoS attacks is solved in the process of designing the event-triggered control law. By combining the neural networks (NNs) and the dynamic surface control (DSC) techniques, the updating weights are significantly compressed. The proposed algorithm is with advantages of concise controller structure, low energy cost, and easy engineering implementation. All the error signals of the closed-loop system are proved to be semi-global uniform ultimate bound (SGUUB) by the Lyapunov analyses. Finally, through two numerical examples, including the comparative simulation and path following maneuver in Shuang island (an island distribute in Liaoning Dalian), are provided to verify the effectiveness of the proposed strategy. • A memory-based event-triggered mechanism is proposed to reduce the wear of steering gear. • A fusion control algorithm is proposed in the presence of DoS attacks. • An improved guidance method is put forward in this paper. [ABSTRACT FROM AUTHOR]

This paper explores the disturbance observer based adaptive trajectory tracking control strategy of Unmanned Surface Vehicle with input and state quantization. To devise the controller without prior knowledge of quantization parameters, a linear analytical model of input quantization is introduced, facilitating adaptive parameter estimation. Estimations of external disturbances and quantized states are obtained through the application of a disturbance observer. Additionally, the control law is constructed using a combination of backstepping control and dynamic surface techniques. By leveraging Lyapunov stability theory, this paper demonstrates the stability of the trajectory tracking control system for Unmanned Surface Vehicle with input and state quantization, as well as the boundedness of errors between quantized and non-quantized variables within the closed-loop control system. Finally, the effectiveness of the proposed tracking approach is validated through comprehensive simulation experiments. • This paper incorporates signal quantization into the USV trajectory tracking method. • Disturbance observer is used for estimating quantized states and external disturbances. • This paper introduces a linear analytical model of the input quantization. [ABSTRACT FROM AUTHOR]

In maritime search and rescue (SAR), the boat swarm mode can greatly improve the success rate, but because of the influence of the curvature of the earth, the communication distance on the ground is short. Relay communication using unmanned surface vehicles (USVs) is the main means of maritime SAR, but communication speed and reliability are the core difficulties. To solve the problem, this paper proposes a trajectory planning aided unmanned surface vehicle optimization communication method with hierarchical reinforcement learning (TP–OC–HRL), which introduces relay USVs into the maritime SAR system and uses trajectory planning and the adaptive orthogonal frequency-division multiplexing (OFDM) technology as variables to optimize the communication rate of the SAR system. The paper proposes a communication optimization scheme based on hierarchical reinforcement learning, which divides the modulation coding scheme and the trajectory planning of the USVs into two hierarchical subproblems to improve the communication rate of the maritime SAR system through reinforcement learning. The experimental verification in the island area of the South China Sea shows that the TP–OC–HRL algorithm proposed in this paper has a faster convergence speed compared with that of the existing relay communication optimization methods, and can improve the communication rate of the maritime SAR system under the condition of meeting the reliability. • We established a high-speed maritime search and rescue communication method utilizing USV relays. • We established a surface USV communication optimization scheme based on hierarchical reinforcement learning. • Our method can adjust the MCS in real time based on the motion status of the USV, effectively improving the communication speed. [ABSTRACT FROM AUTHOR]

This paper proposes a novel fixed-time output feedback trajectory tracking control method suitable for autonomous surface vehicles (ASVs). Firstly, various complex uncertainty terms, such as unknown external disturbances, unmodeled errors, and actuator faults are considered in this paper. And these terms are uniformly converted into the total uncertainty terms through model transformation for convenience. Then, a fixed-time extended state observer (FTESO) with the newly expanded state is designed to observe and estimate the total uncertainty terms. Next, the saturation dynamic filter (SDF) is adopted to address the input saturation, and the saturation factors of SDF are incorporated into the trajectory error calculation to reduce the torque mutation caused by an excessive error in the control input. Finally, a fixed-time control law is proposed by taking into account the ASV system error and FTESO estimation state. Simulation experiments show that this method has excellent control performance, and this conclusion is validated through stability proof and case comparison. • A FTESO is designed to estimate the unknown states of ASVs. • A SDF is employed to address input saturation. • A novel FTOFC method is proposed for the stable and efficient operation of ASVs. [ABSTRACT FROM AUTHOR]

This paper delves into the distributed formation control of unmanned surface vehicles (USVs) under conditions of simultaneous input and state quantization. The primary objective is to ensure that USVs in the formation can effectively follow the specified trajectory in a bandwidth-limited communication environment. A distributed extended state observer (ESO) is proposed to observe the state information of the virtual leader in the USV formation. Furthermore, another ESO is introduced to address the impact of uncertainty terms in the ship model and quantized state variables during communication on the formation control system. Subsequently, the USVs formation guidance laws are formulated in the kinematics subsystem. In addition, the quantized control laws are described linearly in the kinetics subsystem so that the USVs distributed formation controller does not need to anticipate specific information about the quantization parameters. Meanwhile, the USVs formation tracking underlying control laws are designed in the quantized environment to achieve the tracking of the desired kinematic guidance signal. Then, the stability of the designed distributed formation tracking control system for USVs is proved using Lyapunov stability theory. Finally, the effectiveness of the proposed strategy is validated through two sets of simulation experiments, providing empirical evidence of its practical applicability. • Compared to the existing formation tracking control methods for USVs, this paper investigates distributed formation tracking control of USVs with input and state quantization in complex marine communication environments, which reduces the actuator execution frequency without significantly sacrificing the control quality. • Compared with the prior research addressing state quantization in control systems, this paper introduces a distributed ESO to indirectly address the effect of quantized state variables on the designed control system during the transmission of communication signals between USVs. • Unlike current quantized control methods, which setting parameters in quantizers as constant or adjustable within specific constraints, this paper utilizes a linear model to describe the input quantization process, eliminating the need for a priori parameter information of the quantizer. [ABSTRACT FROM AUTHOR]

In response to the intricate marine environment and the need for diverse, intelligent, and large-scale operations, clustered Autonomous Underwater Vehicles (AUVs) have emerged as a pivotal trend in AUV oceanic applications. This paper delves into three key aspects of clustered AUV control: optimal formation assembly, dynamic cooperative control, and autonomous obstacle avoidance. The formation assembly is approached as a target point allocation problem, employing an F-Norm and Dubins Path (FNDP) based method for efficient path planning and optimal allocation. A novel hybrid control strategy, LFVS, combines leader–follower (LF) and virtual structure (VS) approaches, facilitating cohesive formation maintenance with minimal communication. Additionally, the paper introduces an intelligent adaptive artificial potential field (IA2PF) method. This technique, enhancing energy efficiency and obstacle avoidance, dynamically adjusts to current conditions, thereby overcoming limitations of traditional APF methods like unreachable targets , oscillation, and dead zones. The efficacy and applicability of these methods are corroborated through extensive simulations and practical experiments using three 260 AUVs, and compared with formation control based on the LF strategy, proving their utility and stability in real-world formation assembly and maintenance tasks. • This study developed LFVS-IA2PF, a new multi-AUV formation control strategy. • FNDP is introduced for effective target point allocation in AUV operations. • This paper proposed an adaptive algorithm for AUV formation and obstacle avoidance. • The proposed algorithms are validated with both simulations and actual experiments. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of automatic berthing of unmanned surface vehicles (USVs) in complex scenarios and proposes a multitask optimal control berthing method (MT-OCBM). The berthing process is divided into stages, each with distinct goal constraints and path restrictions. Through a combination of topology and computational geometry, a sparse Delaunay triangulation map segmentation (SDTMS) is introduced. SDTMS reduces map complexity while preserving accuracy by extracting map feature points and dividing the map into multiple triangular regions. The application of the traditional hp-Radau pseudospectral method (hp-RPM) to solve optimal control problems (OCPs) in complex marine environments is restricted by problems such as slow convergence and susceptibility that arise because of greatly different magnitudes of path constraints. To overcome these problems, we propose the hp-normalized Radau pseudospectral method (hp-NRPM). Normalizing the path constraints equalizes the magnitudes of path constraints, thereby significantly reducing the computational complexity of solving the OCPs. In contrast to traditional approaches that separate control from path planning, the optimal control algorithm employed in this paper integrates the control variables of unmanned vessels into the planning process of berthing paths. This integration allows the planned berthing paths to reflect the motion characteristics and constraints experienced by the unmanned vessel. It uses the KCS ship model to construct the Maneuvering Modeling Group ship motion model, and using the power spectrum given in the NORSOK standard, it establishes a slow, time-varying wind field model. The hp-NRPM algorithm is used to adjust the control inputs within the constraint region, thereby deriving a berthing route that optimizes the performance function. For validating the method, Qingdao Port was used as a simulation environment, and multiple sets of experiments were performed. The experimental results confirmed that MT-OCBM is robust and effective against disturbances and various obstacles encountered during berthing. • Pseudospectral method can effectively solve the optimal control berthing problem. • Using Delaunay triangulation for map segmentation can mitigate problem complexity. • Multitasking method allows algorithm focusing on different goals at various stages. • Normalizing path constraints can expedite the convergence speed of berthing problem. [ABSTRACT FROM AUTHOR]

In order to improve the autonomous navigation capability of autonomous underwater vehicles (AUV), this paper applies Deep Reinforcement Learning (DRL) to realize autonomous path planning. In response to the limited detection range and difficult tracking DRL policy in three-dimensional unknown environments, this paper proposes an improved deep reinforcement learning method. Firstly, performing AUV modeling and processing obstacle detection information. Second, under a framework of reinforcement learning with the control system, an improved environmental interaction method was designed by analyzing the motion characteristics and controller performance of the AUV. Finally, based on this method, a policy network that can generate paths that satisfy obstacle avoidance and navigation is trained in combination with a twin delayed deterministic policy gradient (TD3). The simulation results show that the proposed method can achieve AUV path planning in a three-dimensional unknown environment. The proposed framework can ensure that the planned path meets the dynamics constraints and is trackable. The improved algorithm can enhance the convergence of training and the attitude stability of AUV. • This article applies forward-looking sonar to achieve three-dimensional path planning in unknown environments based on deep reinforcement learning, relying on the limited obstacle detection range of AUV itself. • The proposed path planning algorithm considers the characteristics of the model and controller in the process of deep reinforcement learning. • In other algorithm details, the reciprocal obstacle distance, and dense reward function are designed. These methods can significantly improve the efficiency of training. [ABSTRACT FROM AUTHOR]

The use of autonomous vehicles for marine and submarine work has risen considerably in the last decade. Developing new monitoring systems, navigation and communications technologies allows a wide range of operational possibilities. Autonomous Underwater Vehicles (AUVs) are being used in offshore missions and applications with some innovative purposes by using sustainable and green energy sources. This paper considers an AUV that uses a hydrogen fuel cell, achieving zero emissions. This paper analyses the life cycle cost of the UAV and compares it with a UAV powered by conventional energy. The EN 60300-3-3 guidelines have been employed to develop the cost models. The output results show estimations for the net present value under different scenarios and financial strategies. The study has been completed with the discount rate sensibility analysis in terms of financial viability. • This paper considers an AUV that uses a hydrogen fuel cell, achieving zero emissions. • This paper analyses the life cycle cost of the UAV and it is compared with a UAV powered by conventional energy. • The EN 60300-3-3 guidelines have been employed to develop the cost models. • The output results show estimations for the net present value under different scenarios and financial strategies. • The study has been completed with the discount rate sensibility analysis in terms of financial viability. [ABSTRACT FROM AUTHOR]

This paper introduces a prescribed-time extended state observer (PTESO) guidance law specifically developed for the path tracking of underactuated unmanned surface vehicles (USVs) navigating under time-varying ocean current conditions. The PTESO is utilized for accurate determination of the sideslip angle of the USV, particular during maneuvering along curved trajectories or in the presence of time-varying ocean disturbances. This process guarantees a prompt and accurate calculation of the sideslip angle, thus enabling more accurate path tracking in the presence of external disturbances caused by wind, waves, and currents. The stability of the closed-loop system's input-state is established using cascade theory. Furthermore, in order to improve the accuracy of USV positioning, the paper suggests utilizing ultra-wideband (UWB) technology for USV localization. A layout scheme has been developed for the placement of UWB devices on the vessel, and an equation has been derived for calculating the heading angle. The proposed approach's effectiveness is substantiated through the validation of simulation results, while the feasibility of the method is confirmed by experimental results. • The prescribed-time extended state observer is employed for the estimation of sideslip angle. • It effectively suppresses the initial peak phenomenon of the extended state observer (ESO). • Ultra-Wideband (UWB) technology is applied to ship positioning. [ABSTRACT FROM AUTHOR]

This paper investigates the path-guided coordinated control (PACC) of unmanned surface vehicles (USVs) under communication interruption. All USVs follow the single-path-guided control method and only certain USVs have access to the virtual leader. In order to maintain the USVs formation to the maximum extent and avoid individual losses when avoiding the communication shielding area, a fixed-time formation control method based on exponential smoothing prediction is proposed in this paper. Firstly, a sigmoid function based artificial potential field is designed to optimize the kinematics control output of USVs during obstacle avoidance. Secondly, a fixed-time distributed kinematics control law based on Holt-winters exponential smoothing (HWES) prediction is developed to maintain the USVs formation to the maximum extent under communication interruption. Finally, a fixed-time cascade control system is proposed to achieve fast convergence of kinematic and kinetics errors. Numerical simulations and comparisons are conducted to demonstrate the effectiveness and advantages of the proposed method. • A sigmoid function based artificial potential function is proposed to ensure USVs avoid obstacles smoothly and avoid kinematic input saturation. • A data-driven predictor based on Holt-winters exponential smoothing is designed to maintain USVs formation under communication interruption. • A fixed-time cascade control system is proposed. In kinematic layer, a fixed-time distributed kinematic control law is developed by combining a fixed-time sideslip angle observer. In kinetic layer, a CONN based fixed-time sliding mode controller is developed. [ABSTRACT FROM AUTHOR]

This study investigates an optimal fuzzy output feedback tracking control problem with a Q-learning algorithm on unmanned surface vehicles (USVs) with immeasurable velocities information. Firstly, the USVs are modeled by a Takagi–Sugeno (T–S) fuzzy system. Based on the fuzzy system, an ideal optimal fuzzy output feedback tracking controller is proposed by utilizing reconstruction states, measured output and input data. Theoretically, the analytical solution of the designed optimal controller is reduced to solving the algebraic Riccati equations (AREs). Due to the difficulty of solving AREs directly, a Q-learning value iteration algorithm is proposed to obtain its approximation solution. Furthermore, it is rigorously mathematically proved that the proposed Q-learning algorithm can approximate the ideal solution and guarantee the USVs output to track the desired reference signal. Ultimately, the simulation and comparison results with the existing control method verify the efficacy and advantage of the presented fuzzy optimal output feedback tracking control approach with a Q-learning algorithm. • This paper presents an optimal output feedback tracking control method. The developed controller can make the USVs system be stable and the tracking errors converge to zero. Compared with (Wang et al. (2018);Zhang et al. (2021)), they did not solve the optimal control problem considered like this paper. • An output feedback Q-learning algorithm is proposed to search the analytical optimal solution. Although authors in (Zhang et al. (2020);Zhang et al. (2019)) also proposed optimal algorithm based on RL for fuzzy systems, they require states of systems must be measurable. Hence, they cannot be applied to USVs systems with unmeasured states. [ABSTRACT FROM AUTHOR]

Unmanned boat clusters formed by unmanned boats may not only enhance efficiency but also improve the accuracy and stability of the overall positioning of the cluster, which is of significant research interest. The challenging offshore island environment makes GNSS susceptible to disruption from occlusion effects and signal reflection. Existing cooperative positioning methods are unable to achieve the accuracy and stability required for large-scale cluster positioning of unmanned boats. To address the aforementioned issues, this paper proposes a Geomagnetic Matching Cooperative Positioning Method for Unmanned Boat Cluster Based on Factor Graph (GUCP). The method applies geomagnetic matching technology by first using cooperative positioning based on factor graph to correct the positioning of unmanned boat clusters. It then uses the corrected positioning as the reference position for geomagnetic matching, and feeds the position estimate back to the unmanned boat clusters for the next iteration positioning. This approach eliminates the reliance on inertial navigation and enables long-term autonomous navigation and positioning of unmanned boat clusters. The results indicate that the GUCP proposed in this paper compared to existing cooperative positioning methods can enable prolonged use of unmanned boat clusters and effectively enhance the positioning accuracy and stability. • We established a co-localization model for UVC(Unmanned Vehicle Cluster). • We established an autonomous geomagnetic co-localization method for UVC. • Our positioning accuracy is higher without satellite navigation. • Our stabilization is better without satellite navigation. • When there is a sudden change in the geomagnetic field, we can implement corrections faster. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of formation control and obstacle avoidance for a swarm of unmanned surface vehicles operating in complex environments with unknown disturbances and obstacles. We propose a novel model-free sliding mode predictive adaptive control scheme (MFSPAC) that combines model predictive control and backstepping structure. The proposed MFSPAC only requires input and output data to achieve effective performance. The value of this paper lies in the design of a novel model-free sliding mode predictive control law, which enables formation control without relying on the system model and requiring no prior knowledge. By combining model predictive control with backstepping, this paper not only facilitates the application of model-free control schemes to rotating robotic systems but also enables obstacle avoidance for such systems using MFSPAC schemes. Subsequently, the model-free extended state observer method is employed to observe unknown lumped disturbances and pseudo-Jacobian matrix estimation errors without requiring a system model. Application of the Gerschgorin disk theory demonstrates that the proposed method is bounded-input, bounded-output stable upon completion of obstacle-avoidance actions. Numerical simulations confirm the superiority of the designed controller. • The method and design comprehensively consider all disturbance, noise, and limitations. • A model-free control method is obtained that relies only on input and output data. • Model-free extended observer methods estimate system state without a priori knowledge. • The control method is superior to legacy methods in complex marine environments. • Control is bounded-input bounded-output stable after obstacle-avoidance activity. [ABSTRACT FROM AUTHOR]

In this paper, a novel fixed-time fault-tolerant control (FTC) scheme is proposed for unmanned marine vehicles (UMVs) with dynamics uncertainties and ocean disturbances. By introducing a high-order integral term, a fixed-time integral terminal sliding mode (ITSM) with faster convergence rate is proposed, which mitigates the effect of the initial system state on the convergence time successfully. Furthermore, based on the online estimation information from the adaptive mechanism, the effects of thruster faults, dynamics uncertainties, and ocean disturbances can be compensated by designing an adaptive fixed-time fault-tolerant controller. Compared to the existing results, the proposed FTC strategy ensures stability in a fixed-time framework. Finally, the comparison simulations of a floating production ship have demonstrated the efficacy of the proposed scheme. • A new fixed-time integral terminal sliding mode (ITSM) control strategy is designed for UMVs in this paper. Unlike our previous work in asymptotic convergence over an AISMC (Yu et al., 2022), the proposed FTC method can ensure fixed-time convergence. • A novel adaptive FTC strategy in the framework of fixed-time ITSMC is presented. Compared with existing result that neglected actuator faults (Qiao et al., 2018), the fault effect are taken into account. [ABSTRACT FROM AUTHOR]

This paper investigates a parameterized path-guided following/maneuvering control problem of an autonomous surface vehicle (ASV) subject to actuator faults via a noncooperative game strategy. Unlike the existing path maneuvering control methods, the noncooperative game of this paper includes two sub-games. On one hand, the path update is considered to be against the effort of kinematic control. On the other hand, the kinetic control is considered to be against the total disturbances consisting of actuator faults and external disturbances. A robust adaptive fault-tolerant path maneuvering controller is designed based on a noncooperative game approach. Specifically, we design a kinematic control law using an improved dynamic surface control approach to achieve the geometric objective. An improved neural predictor is constructed, in which a concurrent learning-based adaptation is designed to identify unknown fault coefficients. A path update law and a kinetic control law are calculated to cover noncooperative games by using an adaptive dynamic programming approach. Theoretical analysis shows that the closed-loop system is input-to-state stable, and the proposed method can meet the geometric objective, the dynamic objective, and the fault-tolerant objective. Finally, simulation results demonstrate the efficacy of the proposed robust adaptive fault-tolerant control method for path maneuvering. • A noncooperative game-based variable tuning law is proposed for path update. • A fault-tolerant control method is developed using a noncooperative game strategy. • An improved neural predictor is designed for identifying fault coefficients. [ABSTRACT FROM AUTHOR]

Path planning is a key factor for the unmanned surface vehicle (USV) to achieve efficient navigation. In this paper, to solve the global path planning and obstacle avoidance problems for the USV, an improved Q-Learning algorithm called neural network smoothing and fast Q-Learning (NSFQ) is proposed. Three main improvement parts are composed of the proposed algorithm. Firstly, the radial basis function (RBF) neural network is combined with the Q-Learning algorithm to approximate the action value function Q, which improves the convergence speed of the Q-Learning algorithm. Secondly, to ensure that the planned path conforms to the maneuvering characteristics of the USV, the heading angle, motion characteristics, ship length, and safety of the USV are taken into account by the proposed algorithm. Based on these factors, the action space and reward function are optimized, the state space is reconstructed, and the safety threshold is proposed. Finally, a third-order Bezier curve is used to smooth the initial path, so that the USV can maintain its heading stability during navigation. Based on simulation results, the proposed NSFQ algorithm outperforms the A* and RRT algorithms in terms of evaluation indicators such as heading angle, angular velocity, path length, sailing time, and path smoothness. • An improved Q-Learning algorithm called neural network smoothing and fast Q-Learning (NSFQ) is proposed in this paper. • The heading angle, angular velocity, and other indicators of the USV are taken into account by this proposed algorithm. • A safety threshold is proposed to ensure the safety of the USV. • A third-order Bezier curve is used to smooth the initial path. [ABSTRACT FROM AUTHOR]

In this paper, we propose and demonstrate a method for risk-based path planning, generating risk-acceptable trajectories for an autonomous surface ship. This objective is achieved in two steps. The first step uses a dynamic probabilistic risk assessment method called K-shortest-paths Probabilistic Risk Assessment (KPRA), alongside risk models and dynamic system simulation, whose output is a list of paths ranked with respect to risk. Multiple risk influencing factors and accident types may be considered in this step. In this work, collision and powered grounding accidents, and their risk-influencing factors and indicators, are considered. The second step is to perform risk evaluation on the ranked scenarios, comparing them to predefined risk acceptance criteria and defining which scenarios have acceptable risk. Evaluating multiple paths allows operators to choose how strict the autonomous ship's decision-making should be regarding risk, avoiding the problem of minimization algorithms being too conservative in terms of risk. One of the risk-acceptable paths is then chosen as a local plan to be acted by the autonomous ship. This paper has two main scientific contributions. First, an online Bayesian belief network risk model for navigation accident probability is designed and integrated to improve KPRA. Second, we propose a modified guidance, navigation, and control framework for autonomous ships, to include risk-based planning. We implement the modified framework into a real autonomous surface vehicle (ASV) named Grethe and demonstrate the method's capabilities in field trials, where the ASV uses the method to generate local plans that avoid possible collision and grounding accidents. • Proposal of a method for risk-based path planning for maritime autonomous surface ships. • The method uses a risk model and simulation to generate a risk-ranked list of paths. • Risk acceptance criteria define a set of acceptable paths which the ship can follow. • The method is demonstrated in field trials with an autonomous surface vehicle. • The autonomous surface vehicle successfully avoided collisions and groundings. [ABSTRACT FROM AUTHOR]

In this article, the fixed-time Nash equilibrium (FTNE) seeking problem of multiple autonomous surface vehicles (ASVs) is considered. As individuals fight for an allocation of specific shared but restricted resources, two novel FTNE seeking algorithms are designed for this topic through an error decomposition method. First, a model-based control algorithm with only one convergence time function in a single-layer structure is designed to achieve Nash equilibrium within a fixed time. Then, a FTNE algorithm with a three-layer structure is presented for multiple ASVs subject to system uncertainties. In particular, this paper proposes a hybrid fixed-time event-triggered approach for reducing controller update frequencies and communication frequencies of two algorithms simultaneously, which is capable of decreasing resource consumption in a constrained resource environment. Furthermore, rigorous sufficient criteria are established for fixed-time convergence through Lyapunov stability analysis, and the upper bound on settling time without the requirement of initial conditions is derived. Finally, simulation examples are given to demonstrate the viability of two algorithms. • This paper solves noncooperative game problem of multiple ASVs. • Two FTNE seeking algorithms are firstly proposed. • An error decomposition approach is firstly presented. • A hybrid fixed-time event-triggered approach is presented. • A three-layer control framework is established. [ABSTRACT FROM AUTHOR]

This paper considers the course tracking control of an unmanned surface vehicle with event-triggered mechanism and input quantization, aiming to reduce the influence of external interference and saving communication resources. The fuzzy logic system is used to approximate the model uncertainty and external interference. A linear analytical model is used to describe the input quantization process. The adaptive fuzzy quantization controller does not need the prior information of quantization parameters. Then an event-triggered adaptive fuzzy quantization control law is proposed. In this way, the frequency and amplitude of execution are reduced and the communication burden is relaxed. On the basis of the Lyapunov stability theory, the stability of the proposed control structure is proved, and all the signals in the closed-loop system are ultimately bounded. Finally, the effectiveness and feasibility of the proposed algorithm are verified by simulation experiments. • Compared with the existing ship course tracking control methods, the course tracking control of USV with input quantization and event-triggered mechanism is considered in this paper for the first time. • Compared with the existing quantitative control methods which regard quantized variables as perturbations of unquantized variables, a linear analytic model is used to describe the input quantization process. The designed controller does not need the prior information of quantization parameters. [ABSTRACT FROM AUTHOR]

In this paper, a reinforcement Q-learning fuzzy optimal tracking control method is investigated for nonlinear underactuated unmanned surface vehicles (USVs) with external disturbances. Firstly, the motion dynamics of USVs are described by the Takagi–Sugeno (T–S) fuzzy discrete-time systems. Secondly, by applying parallel distributed compensation (PDC) method and constructing cost function, the existence condition of optimal solutions is derived and reduced to the algebraic Riccati equations (AREs). To solve the solutions of the GAREs, a Q-learning value iteration (VI) algorithm is further developed, which can be implemented by eliminating the demand of initial allowable control policy and system information. The developed strategy has the advantage that the optimal control policy can ensure the USV to be stable and the desired reference signal can also be well tracked by the USV output. Finally, the proposed fuzzy optimal control algorithm is applied to control the USV, the simulation results verify the feasibility of the proposed optimal control algorithm. • A fuzzy optimal control strategy for underactuated USVs is designed in this study. The designed fuzzy optimal controller can make the output of underactuated USV track the desired reference trajectory. Though the authors in (Wang et al. (2018); Hao et al. (2021); Ma et al. (2020); Zhang et al. (2021); Song et al. (2023)) also investigated the fuzzy control design problems of the USVs, they did not study optimal control design problem and thus the optimal control performances of USVs cannot be achieved like this paper. • This paper proposes a fuzzy optimal tracking control algorithm via convergent Q-learning VI which can obtain the optimal control input policy of underactuated USVs. Although the existing results (Zhang et al. (2019b); Zhang et al. (2019a); Zhang et al. (2020); Wang et al. (2022b)) also investigated the fuzzy optimal tracking control design problems, they ignored the ocean disturbances existing in USVs systems, thus they have no robustness against the ocean disturbances. In addition, though the authors in (Wang et al. (2022c)) proposed fuzzy optimal controller, it did not investigate the fuzzy optimal tracking control design problem of USVs. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of time-varying formation control for heterogeneous unmanned swarm systems (USSs) consisting of multiple unmanned aerial vehicles (UASs) and unmanned surface vehicles (USVs) based on a directional switching topology. Firstly, a novel USSs model is established, involving the effect of multiple network attacks, network bandwidth constraints and input saturation. In particular, a scaling attack model is proposed to account for denial-of-service (DoS) and deception attacks. Secondly, a distributed controller based on a dynamic event triggering mechanism is designed to reduce the communication load by adjusting the communication frequency between individuals, and the Zeno phenomenon is effectively avoided. Finally, the Lyapunov function is used to obtain sufficient conditions for system stability and bounds on the frequency and duration of network attacks. The experiments and comparison results verify the proposed strategy's effectiveness and superiority. • A novel heterogeneous USS model consisting of UAVs and USVs is constructed. • A new scaling attack model is established that integrates the DoS attacks and deception attacks. • A novel controller based on a dynamic event triggering scheme is designed. [ABSTRACT FROM AUTHOR]

This paper presents a novel feedback motion planning and control framework tailored for unmanned surface vehicles (USVs), validated through experiments on a physical USV platform. The framework utilizes sparse random neighborhood trees and a velocity command generator based on partial feedback linearization. Compared to existing approaches, the methodology introduces a sequential composition of elliptical regions and associated nonlinear feedback control algorithms (funnels) to enhance sparsity and computational efficiency. The approach comprises two primary stages: neighborhood tree generation and feedback motion control. In the tree generation stage, a sparse, collision-free, connected tree structure is constructed within the environment using elliptical regions (funnels). Subsequently, each funnel is assigned a dedicated control policy to ensure safe vehicle operation and guide it toward its goal position. To validate the effectiveness of the proposed approach, comprehensive experiments were conducted involving both simulation studies and physical implementations on an autonomous surface vehicle platform. A significant contribution of this work is the development and hardware validation of a partial feedback linearization-based controller, which ensures robust navigation of the vehicle within elliptical regions. The proposed trajectory-free, sampling-based feedback motion planning scheme is compatible with 2D polygonal map representations, enabling practical deployment in real-world environments for USVs. Moreover, the use of elliptical funnels enhances tree sparsity compared to circular funnels, thereby improving computational efficiency, reducing mode changes, and decreasing the total operation duration. • Global path planning solution via local planning on the sparse elliptical funnel trees. • A novel motion control strategy based on this local funnels. • Proof of stability of our control strategy. • Simulation and the implementation of this control strategy on this tree based path. [ABSTRACT FROM AUTHOR]

This paper investigates the problem of trajectory tracking control for unmanned surface vehicles (USVs) with uncertain dynamics, external disturbances, in consisting with actuator fault, dead-zone and saturation. First, a Sigmoid function-based equivalent transformation method is proposed, with which the actuator constraints can be handled under the same framework in a smooth way. To improve the converge rate of the system, a new type of sliding mode surface is given, based on which, an adaptive finite-time trajectory tracking control scheme is further developed to guarantee the surface vehicles track the given trajectory with finite time, while dealing with the physical limitations of the actuator. At last, both simulation and comparative results are provided to illustrates the effectiveness of the proposed control scheme. • A Sigmoid function-based equivalent transformation strategy is proposed, based on which multiple actuator constraints can be dealt with simultaneously. • An improved non-singular sliding mode control strategy is proposed to achieve accurate tracking with finite time convergence. • Under the given scheme, faster convergence rate, smaller overshoot and smaller steady-state error can be obtained. [ABSTRACT FROM AUTHOR]

Considering a trajectory reproducibility mission of an autonomous surface vehicle (ASV), this paper investigates an unsupervised path following control framework that uses a fusion mechanism of the Automatic Identification System (AIS) data and an improved iterative learning control (ILC) strategy. The proposed framework consists of two modules: the guidance term and the control term. For the former, a dynamic virtual ship (DVS) guidance principle is developed by combining the data of AIS, leads to a smooth desired path. Linked with the guidance principle, an improved ILC algorithm is designed that fuses an explicitly fact that an unpredictable actuator fault may be encountered due to the marine disturbance. Note that the tracking accuracy can be increased while the number of iterations is adding. Besides, the dynamic surface control (DSC) and robust neural damping technique are introduced for approximating the model uncertainties and the derivative of the virtual control law. Through the Lyapunov theorem, the semi-global uniform ultimate bounded (SGUUB) properties of the proposed control algorithm are proved. Finally, two numerical examples, including a comparative simulation and a simulated path following operation, are carried out for illustrating the superiority and effectiveness of the proposed framework. • A DVS guidance principle is developed by coupling the AIS data for an unsupervised path of ASV. • An ILC strategy is presented to enhance the control accuracy with the increasing number of iterations. • An actuator failure compensated mechanism is employed for improving the fault-tolerant performance of the control system. • The DSC and robust neural damping technique are introduced for ensuring a concise form of controller with lower conpensation load. [ABSTRACT FROM AUTHOR]

The Foldable-Wing Unmanned Aerial-Underwater Vehicle (F-UAUV) demands swift movement over water surfaces at significant pitch angles during the water-exit procedure, a phase highly influenced by strong, non-linear, time-varying hydrodynamic effects. To tackle these complexities, we devised a model experiment scheme mirroring the actual water-exit process. Crafted through structural drainage comparisons, high-fidelity models were developed, complemented by a test platform featuring a double track for accurate evaluations. Subsequent water-exit tests, conducted at various speeds and angles, unveiled a consistent increase in water-exit resistance with speed, peaking at the moment of exit. Concurrently, resultant torque escalated inversely with both speed and angle. Notably, the heightened resistance due to structural drainage exhibited a tendency to stabilize during the latter half of the water-exit process. Utilizing the foundational formula for resistance in a single medium, we proposed an enhanced formula for predicting maximum water-exit resistance, a tool instrumental in facilitating F-UAUV design and powertrain selection. This paper offers an exhaustive analysis of hydrodynamic influences during the water-exit process, serving as a pivotal reference for Unmanned Aerial-Underwater Vehicles development. • Hydrodynamic experiments analyzed the structure drainage and water-exit resistance of F-UAUV. • An improved formula for predicting maximum water-exit resistance is proposed by quantifying structural drainage resistance. • A water-exit experiment scheme suitable for F-UAUV confinement is designed based on the real water-exit process. [ABSTRACT FROM AUTHOR]

This paper proposes a distributed dynamic event-triggered control that addresses the flocking control problem with a virtual leader for multiple unmanned surface vehicle (USV) systems. Unlike existing flocking algorithms, which often result in the formation of quasi α-lattice structures, the proposed approach combines artificial potential functions (APF), directionally consistent potential functions (DCPF), and a navigation feedback term to form an α-lattice. A novel artificial potential function is developed to facilitate flocking and maintain a safe distance between USVs to avoid collisions. Additionally, a dynamic event-triggered mechanism, initially employed in flocking control, is introduced to reduce communication among USVs. Furthermore, it is demonstrated that Zeno behavior does not occur. The algorithm employs Radial Basis Function Neural Networks (RBFNNs) and adaptive algorithms to dynamically estimate and compensate for parameter uncertainties and external disturbances. Finally, the effectiveness and robustness of the proposed algorithm are verified through numerical simulation. • A distributed dynamic event-triggered control is proposed for the first time to study the flocking problem in multi-USV systems. • A smooth artificial potential function is designed to take into account the collision avoidance problem among non-pointed agents. • A smooth artificial potential function is designed to guarantee the function with a bound cut-off. [ABSTRACT FROM AUTHOR]

Unmanned surface vehicles (USVs) are gaining significant interest, particularly in support of complex maritime operations. To ensure efficient autonomous operation of the USV swarm in the case of multi-type missions, a task allocation system that combines off-line and on-line task allocation is constructed in this paper. The task set of each USV is obtained by improved genetic algorithm (GA). A distance matrix of these tasks is calculated by fast marching method (FMM) and integrated into the self-attention mechanism (SAM). The algorithm can shorten swarm's total voyage and make USVs' task load balanced. Furthermore, to accommodate the situations of task addition, deletion, and failure of USVs, an on-line task reallocation algorithm called the improved contract net protocol and vacancy chain (ICNP-VC) is developed. It unites the contract net protocol (CNP) and vacancy chain (VC) to enhance the intelligence of USVs. The ICNP-VC modifies the content of the bid and distinguishes the type of the cost. Meanwhile, the algorithm can balance the task load of the cluster and reduces the waste of resources. Through extensive simulations, demonstrate the proposed algorithms' capability to optimize task allocation and plan safe paths, leading to improved performance in complex marine scenarios. • A new off-line task allocation method for USVs swarm is proposed, which has a shorter total voyage of USVs and smaller relative time deviation and is applicable to complex environments with multiple static obstacles. • A new method is proposed to optimize the task sequence of USVs so that higher priority or time-bound tasks are processed first and is suitable for multi-type task situations. • A new on-line task reallocation algorithm is proposed to improve the performance of load balancing and reduce the waste of resources when meeting burst situations in complex scenarios. [ABSTRACT FROM AUTHOR]

The Unmanned Surface Vehicles (USVs) cluster shows significant potential in executing complex maritime tasks. However, existing task decision-making approaches often overlook inherent constraints of tasks such as executable time and operational area, leading to inefficiency during execution. This paper addresses the task decision-making problem for USV clusters by a bi-layer framework and introduces a novel Multi-Constrained Task Point Adaptive Self-Organizing Map (MCTP-ASOM) algorithm to improve the task execution efficiency within these clusters. In the task planning layer, the MCTP-ASOM algorithm incorporates the dynamic time window and perceptual distance constraints of task points. It adapts neuron movement by incorporating task waiting time and perceptual distance into the SOM algorithm's WTA and Kohonen's rules, enabling the planning of task execution sequences and optimization of execution positions, effectively reducing average waiting time and navigation costs during task completion. In the multi-task allocation layer, the K-means algorithm is utilized to minimize the inter-task distance assigned to each USV, thereby optimizing the cluster efficiency. Extensive simulations across various scenarios confirm the effectiveness of this approach, with an average reduction in task waiting time by 18.34%, per USV waiting time by 18.86%, and navigation cost by 16.3%, particularly beneficial in multi-USV operations with various constraints. • A novel MCTP-ASOM algorithm is proposed for task planning of multi-constrained tasks. • Time window constraint of tasks is considered by MCTP-ASOM in task sequence planning. • Execution position is adaptively decided by MCTP-ASOM with reduced navigation costs. • The solution complexity of task decision-making is reduced by a bi-layer framework. • Cluster performance is optimized by efficient multi-task allocation using K-means. [ABSTRACT FROM AUTHOR]

In the presence of global positioning system signal rejection, the low-cost inertial navigation system exhibits a significant cumulative error, leading to poor positioning accuracy of slave unmanned surface vehicles (SUSVs) in the cooperative navigation (CN) system for leader-slave. To improve the positioning accuracy of SUSVs, the extended Kalman filter (EKF) is augmented with distance information from external distance sensors. However, in the CN system of single leader unmanned surface vehicle, there may be a weaker observability, resulting in suboptimal estimated effects of the EKF. For the first time, the CN algorithm with observability and trilateral positioning method (TPM) is proposed to compensate the estimated position errors for EKF. An event trigger constructed from observability is used to assign the base stations and tags of the TPM. Then, the estimated positions of the tags are obtained by EKF when observability becomes weaker and the estimated positions are corrected by the measurements from TPM. Finally, more accurate estimated positions are obtained by weighted mean and the weights of compensation are automatically adjusted with observability. The results of the validation experiments and simulations show that the estimated positional accuracy of the proposed CN algorithm has improved by 40–50% compared to EKF. • Trilateral positioning method and observability are first offered to compensate the estimated position errors for EKF when observability becomes weaker in the cooperative navigation system of unmanned surface vehicles. • In order to improve the operational efficiency of the proposed collaborative navigation algorithm, event trigger is constructed based on observability to selectively compensate the estimation errors from the EKF. Finally, estimated positions with higher accuracy are achieved by obtaining automatically varying compensation weights through observability. • The experimental results and simulations show that the positional accuracy of the algorithm proposed in this paper is improved by 40–50% compared to EKF. [ABSTRACT FROM AUTHOR]

A novel control design based on Generalized Udwadia–Kalaba (GUK) theory is proposed to achieve the trajectory tracking under comprehensive constraints in this paper. To describe the complex task requirements, a constraint framework including equality and inequality constraints is constructed. Thus, the complex trajectory tracking problem is converted into a constraint-following problem. Based on the framework above, GUK equation is introduced to consider the equality and inequality constraints individually. Specifically, the equality constraints are transformed into second-order matrix form, while the inequality constraints are addressed via a diffeomorphism. Then, a novel control method including servo control term and high-order control term is proposed. Through these two control terms, the constraint forces for equality and inequality can be solved separately. The effectiveness of this control method is verified by the stability analysis based on Lyapunov minimax approach and illustrative application of unmanned surface vehicle (USV). Meanwhile, the superiority of this method is demonstrated through a comparative analysis with Udwadia–Kalaba (UK) method, Linear Quadratic Regulator (LQR) method and sliding mode control (SMC) method. • A comprehensive constraint framework is considered based on the real situation. • The generalized Udwadia–Kalaba method is proposed to establish the dynamic model. • The controller is designed for trajectory tracking and uncertainty compensation. [ABSTRACT FROM AUTHOR]

This paper is concerned with a coordinated search and rescue problem of heterogeneous ocean vehicle cluster composed of autonomous surface vehicles (ASVs) and autonomous underwater vehicles (AUVs) subject to model uncertainties and ocean disturbances. Firstly, a three dimensions (3D) cross domain kinematic controller with auto-stop function at the rescue points is designed for the coordinated ASV-AUV system by using a line-of-sight guidance law. Secondly, an event-triggered based communication mechanism is constructed to ensure that the ASV-AUV system can broadcast information at its trigger moments except to the auto-stop points. Finally, an adaptive motion control law based on a low-frequency learning fuzzy predictor is designed, which is able to filter out the high-frequency components of uncertainties and disturbances. The stability of the closed-loop system is proven by employing input-to-state stability and cascade stability analysis. Simulation results and numerical analysis show the effectiveness of the proposed 3D heterogeneous cross domain coordinated path following controller with auto-stop function for marine search and rescue. • In contrast to the existing coordinated path following controllers which applicable to a two-dimensional or three-dimensional coordinated control in the same domain, a three-dimensional cross domain coordinated controller with auto-stop function applicable to sea surface and underwater scenarios are constructed. • In contrast to the existing communication that cannot stop during the whole control period, a predictor-based event-triggered communication mechanism is proposed such that the narrow network band could be effectively alleviated, and the communication will be suspended at the rescue points. • In contrast to the existing predictor design methods, a low-frequency learning fuzzy predictors are presented not only to filter out high-frequency components of non-linearities uncertainties and unmodeled random external disturbances, but also to achieve fast adaptation. [ABSTRACT FROM AUTHOR]

Abstract Collision avoidance in 3D environments is important to the problem of planning safe trajectories for an autonomous vehicle. Existing literature on collision avoidance assumed that obstacle shapes are known a priori and modeled obstacles as spheres or bounding boxes. However, in 3D environments, an obstacle shape is unknown to the autonomous vehicle, and the vehicle detects an obstacle boundary using 3D sensors, such as 3D sonar. In this paper, we introduce control laws for collision avoidance, considering scenarios where a vehicle detects arbitrarily shaped and non-convex obstacles using sensors. Moreover, our control laws are designed considering motion constraints, such as the maximum turn rate and the maximum speed rate of the vehicle. The effectiveness of our control laws is verified using MATLAB simulations. Highlights • Existing literature on collision avoidance assumed that obstacle shapes are known a priori and modeled obstacles as spheres or bounding boxes. • However, in 3D environments, an obstacle shape is unknown to the autonomous vehicle, and the vehicle detects an obstacle boundary using 3D sensors, such as 3D sonar. • In this paper, we introduce control laws for collision avoidance, considering scenarios where a vehicle detects arbitrarily shaped and non-convex obstacles using sensors. [ABSTRACT FROM AUTHOR]

The paper focuses on developing a method that enables the unmanned surface vehicle (USV) to effectively track dynamic objects on the water surface. The integral line-of-sight (LOS) guidance law is improved by introducing the concept of vector field for curved paths and the time-varying lookahead distance with a smoothing control parameter for straight paths. These enhancements facilitate faster convergence to the desired path with reduced oscillations and allow for mitigating the impact of constant external disturbances. The research demonstrates that the proposed guidance law exhibits κ -exponential stability when converging towards a desired path composed of both straight lines and curves. The results presented in the paper demonstrate that the proposed method effectively improves the accuracy of tracking dynamic targets for unmanned surface vehicles while ensuring the safety of docking operations. • A method for unmanned surface vehicle to track dynamic objects on water is designed. • The vector field improves integral line-of-sight guidance for curved path tracking. • The effectiveness of the method is demonstrated, ensuring safe docking operations. [ABSTRACT FROM AUTHOR]

The Cross-Domain Vehicle (CDV) is a sophisticated marine equipment capable of performing tasks across different domains. This paper aims to develop a data-driven digital twin system for the CDV to achieve precise motion control and data monitoring during its operation. Leveraging the kinematic physical characteristics of the CDV, the digital twin system constructs a virtual counterpart of the vehicle, establishing a mapping relationship between the physical and virtual entities. The use of a Long Short-Term Memory (LSTM) network model is proposed for forecasting the motion state of the CDV within the digital twin framework, supporting the vehicle's autonomous motion decision-making process. Additionally, an innovative model predictive control strategy is employed to achieve surface speed tracking control for the CDV, leveraging predictive data within the digital twin system. Utilizing the digital twin system to optimize control parameters, better underwater depth control effects can be achieved for the CDV. Finally, through the system implementation and experiment verification, the effectiveness of the digital twin system is successfully demonstrated, providing reliable results for both surface speed tracking control and underwater depth control. • A digital twin system framework for cross-domain vehicle (CDV) is developed. • The motion attitude and velocity of CDV can be predicted by the digital twin system. • The CDV's digital twin system effectively achieves speed tracking control by the model predictive control algorithm (MPC) algorithm. • In terms of underwater motion control, optimizing the parameters of underwater depth control through the digital twin system can improve the depth adjustment effect. [ABSTRACT FROM AUTHOR]

A wave-driven unmanned surface vehicle (WUSV) can sail on the sea simply by utilizing wave energy as its driving power. Recently, many studies have been performed on practical applications of oceanic scientific exploration by a WUSV, where the estimation criterion of the navigation performance of the WUSV is always its sailing velocity, although this method is very intuitive but not effective in analyzing the shortcomings in the design of the WUSV. On this basis, a new estimation method, which focuses on the efficiency of utilizing wave energy by the WUSV, is proposed in this paper. The theoretical contribution of this paper is that a generalized solution function for the efficiency of utilizing wave energy is obtained and is expressed in the form of the product of several non-dimensional parameters. In detail, algorithms to calculate input power, i.e., wave energy harvested by the surface float, and usable power, i.e., the product of sailing velocity and thrust force, have been developed. The other development presented in this paper is the proposal of some effective ways to improve the ability of the WUSV to utilize wave energy. Then, numerical simulations are carried out to validate the proposed method, and corresponding experiments are employed to investigate the performance of the method. The results show that the four major factors to impact the ability of the WUSV utilize wave energy are the thrust force of the propulsor, the sailing velocity of the WUSV, the principal dimensions of structures and the wave environment parameters. Summarizing the analysis in numerical validations and experimental studies, the WUSV can generally achieve higher efficiency in utilizing wave energy when it sails in a specific wave environment with certain parameters. Meanwhile, the utilization efficiency of the WUSV is not always much higher, and the reasons for this phenomenon mainly include two aspects: (1) most of the wave energy harvested by the surface float is used to work against the calm water resistance of the propulsor and the umbilical cable; and (2) based on the current design, the motions of the surface float cannot fully respond to the waves. Therefore, an integrated optimizing design for the propulsive and resistance performance of the propulsor, as well as the motion response and resistance performance of the float, will be an effective way to improve the ability of the WUSV to utilize wave energy. In addition, the differences in the sailing velocity and heaving amplitude of foils cause different curve trends for the utilizing efficiency between numerical simulations and experiments. [ABSTRACT FROM AUTHOR]

This paper investigates the moving-target enclosing problem in multi-obstacle environments for an unmanned surface vehicle (USV) with complex unknown factors, including target dynamics, vehicle dynamics, and disturbances. A reinforcement learning-based moving-target enclosing control scheme is proposed to ensure collision-free behavior and bolster the enclosing capability. Specifically, an extended state observer is deployed to estimate the target dynamics. Leveraging the estimated data and control obstacle functions, a virtual safety control law is formulated to dynamically harmonize obstacle avoidance and target enclosing control. Then, a novel controller is constructed to track this control law utilizing a Nussbaum-type function in conjunction with actor–critic neural networks (NNs). The actor and critic NNs are employed to approximate unknown dynamics encapsulating vehicle dynamics and disturbances, and value function, respectively. The Nussbaum-type function is embedded to adaptively identify unknown inertia mass, regulating the control input of the USV with the actor NN online. The proposed scheme effectively decouples obstacle avoidance and target enclosing control and only relies on the measurable variables. A rigorous theoretical analysis is further employed to ensure the closed-loop stability of the USV system. Eventually, Simulations are demonstrated to validate the effectiveness and superiority of the proposed scheme for the USV in multi-obstacle environments. • The RLTEC scheme achieves safe enclosing control of USV with complex unknown factors. • The virtual safety control law is designed to ensure real-time avoidance performance. • The NNs-based online RL controller boosts the learning and adaptive abilities of USV. • The modular design efficiently decouples obstacle avoidance from enclosing control. [ABSTRACT FROM AUTHOR]

This paper investigates the defense penetration problem for unmanned surface vehicle (USV) with limited detection range under unknown defenders. A novel end-to-end defense penetration strategy is proposed for the USV based on modified soft actor–critic (MSAC). Specifically, the state–action space of the attacker is designed by combining the maneuvering model of USV and detection information. To guide the USV to reach the target area under the interception of defenders, the penetration reward is constructed. Then, the long short-term memory network (LSTM) is applied in the actor–critic network design. A dynamic training mechanism is proposed by combining curriculum learning and multi-memory pools, which aims to improve the success rate of defense penetration. Finally, simulation results validate the effectiveness and superiority of the proposed defense penetration method for the USV suffering from unknown defenders. The higher success rate can be guaranteed by using MSAC compared to the previous methods. • The defense penetration strategy for the USV under unknown defenders is proposed. • The proposed strategy has strong adaptability due to the long short-term memory network. • A higher defense penetration success rate can be guaranteed due to the dynamic training mechanism. [ABSTRACT FROM AUTHOR]

System identification (SI) is a research focus in the field of autonomous underwater vehicles (AUV) which can be considered as the estimation to the hydrodynamic coefficients (HCs) in nonlinear dynamic equations of the AUV. This paper presents an SI method based on deep reinforcement learning (DRL) to estimate the HCs of AUVs only through the utilization of position and altitude information of the AUV. Compared with most existing methods, fewer design parameters are required in the proposed SI method, thereby making the SI method simpler to perform. First, a sensitivity analysis of the HCs is performed, and then the action and reward mechanism in the origin deep deterministic policy gradient (DDPG) algorithm is improved according to the results of sensitivity analysis. The simulation results show that the HCs can be estimated with a low error level and greater accuracy than the other methods. Therefore, it can be concluded that the DRL SI method possesses effective performance in HC estimation and the improved-DDPG algorithm is more suitable for SI than the original one. • The AUV velocity information is unnecessary so that the method is simpler to perform. • In Improved-DDPG algorithm, system identification accuracy is improved. • The training of neural networks is set to be segmented. • The training of NNs is terminated upon the first attainment of the reward standard. [ABSTRACT FROM AUTHOR]

Existing Global Path Planning (GPP) algorithms predominantly presume planning in static environments. This assumption immensely limits their applications to Unmanned Surface Vehicles (USVs) that typically navigate in dynamic environments. To address this limitation, we present OkayPlan, a GPP algorithm capable of generating safe and short paths in dynamic scenarios at a real-time executing speed (125 Hz on a desktop-class computer). Specifically, we approach the challenge of dynamic obstacle avoidance by formulating the path planning problem as an Obstacle Kinematics Augmented Optimization Problem (OKAOP), which can be efficiently resolved through a PSO-based optimizer at a real-time speed. Meanwhile, a Dynamic Prioritized Initialization (DPI) mechanism that adaptively initializes potential solutions for the optimization problem is established to further ameliorate the solution quality. Additionally, a relaxation strategy that facilitates the autonomous tuning of OkayPlan's hyperparameters in dynamic environments is devised. Comprehensive experiments, including comparative evaluations, ablation studies, and applications to 3D physical simulation platforms, have been conducted to substantiate the efficacy of our approach. Results indicate that OkayPlan outstrips existing methods in terms of path safety, length optimality, and computational efficiency, establishing it as a potent GPP technique for dynamic environments. The video and code associated with this paper are accessible at https://github.com/XinJingHao/OkayPlan. • An obstacle kinematics augmented path planning problem is formulated. • A dynamic prioritized initialization mechanism is introduced to improve planning. • A relaxation strategy is established to facilitate hyperparameter tuning. • A real-time path planning algorithm for dynamic environments is established. [ABSTRACT FROM AUTHOR]

To solve the problem of path discontinuity and tracking difficulty in unmanned surface vehicle path planning, this paper introduces the Minimum Snap algorithm to optimize the trajectory, and makes some improvements to the Minimum Snap algorithm to address the problem of imperfect paths caused by uneven time allocation. The concept of a safe corridor is introduced to optimize the Minimum Snap algorithm. In response to the problem of excessive deviation from the reference value in some areas of the optimized path, a guidance function is added to make the path closer to the original trajectory. In the simulation, the proposed improved Minimum Snap is compared with Bezier curve and B-spline curve, the results show that the improved Minimum Snap can enable unmanned surface vehicle to complete turns and other actions with minimal energy consumption while driving safely, without experiencing severe acceleration and Jerk, laying the foundation for subsequent path tracking control. • The Minimum Snap algorithm is introduced to optimize the trajectory of USV to ensure path continuity. • A secure corridor is introduced into Minimum Snap algorithm to solve uneven time allocation. • A guiding function is added to make the path closer to the original trajectory and ensure smaller path deviation. [ABSTRACT FROM AUTHOR]

Maritime Autonomous Surface Ships (MASS) face regulatory challenges, with some suggesting that the existing Collision Regulations (COLREG) present linguistic barriers for autonomous vessels' development and implementation. While academic research has focused on developing autonomous collision avoidance (CA), it is producing inconsistent results compared to conventional navigation practices. This study aims to identify trends and weaknesses in recent studies on CA for MASS by conducting a systematic review and analysis of the most relevant literature. The Conventional-Collision-Avoidance-Process (CCAP), which benchmarks manned modern ships' capacity for CA compliance under COLREG and industry requirements, is used to break down a ship's collision avoidance process into 53 CA functions under eight main categories. A total of 32 papers were chosen through filtering based on keywords, publication period, language, and relevance. The content of the recent academic literature was then grouped under appropriate CCAP codes. Statistical and graphical interpretations were generated using the collected literature content data and evaluated statistics of the existing digital contribution of CCAP. The study uncovers significant trends, inconsistencies, and weaknesses that could guide future academic research towards comprehensive CA solutions for MASS. • A systematic literature review in the context of Autonomous Collision Avoidance. • Identification of 53 Conventional-Collision-Avoidance-Process (CCAP) functions under eight main categories. • Illustration of the flow map of all the CCAP functions for collision avoidance assurance. • Analysis of the trends and gaps of collision avoidance for maritime autonomous surface ships (MASS). [ABSTRACT FROM AUTHOR]

This paper discusses the use of unmanned surface vehicles (USVs) in the ocean for traversing multiple targets and proposes a novel multiple targets traversing algorithm that takes into account the multi-level priority of target assignment and path optimization. The algorithm includes a bundled genetic optimization target assignment method and a fast marching-Q-Learning path planning method. Comparative studies with existing methods show that the proposed algorithm have efficient performance, where the calculation time for target assignment has been reduced by 45.64%, and the path length has been further optimized by 15.3%. Considering the velocities of USVs navigation, path turning motion and multi-level priorities constraints are satisfied, and collision risk is avoided in dynamic ocean environment. Hence, the proposed multiple targets traversing algorithm plays an important role in the ocean engineering field of USVs. • Bundled-genetic target assignment method: The core of this method is to use bundling mechanism to solve target assignment efficiency problem, including improving the speed of target assignment, satisfying USV velocity and multi-level priority constraints. • FM-Q-Learning path planning method: This method utilizes FM's reward function to solve path optimization problem, including improving path convergence speed, decreasing path length, and smoothing the paths through navigation restriction. • Simulation of remote sensing satellite background: This strategy can reflect the experimental results in a real environment, which helps to verify the applicability of theoretical algorithms in practical situations. [ABSTRACT FROM AUTHOR]

In this paper, a practical prescribed time tracking control scheme is proposed to drive the unmanned surface vehicle (USV) under input saturation and unknown disturbances to track the desired trajectory with guaranteed transient and steady-state performances. First, by employing the logarithmic type barrier Lyapunov function and prescribed time function, the virtual velocity law is designed at the kinematic level. Then, to deal with the unknown disturbances, a prescribed time observer is constructed to achieve perturbation attenuation in a user-defined time. Together with the linear anti-windup compensator, a robust trajectory tracking controller is constructed, which is capable of guaranteeing that all signals in the closed-loop system are uniformly ultimately bounded (UUB) and the tracking error converges to a user-defined compact set in predetermined time with prescribed transient performance. Finally, the effectiveness of the proposed control scheme is validated through Lyapunov theory and numerical studies. [ABSTRACT FROM AUTHOR]

This paper investigates fixed-time adaptive neural quantized formation control of unmanned surface vehicles (USVs) with tunnel prescribed performance in the presence of leader's information unknown, model dynamic uncertainties and input quantization. Initially, a fixed-time observer is constructed to rapidly estimate the leader's unknown information. Following this, a control strategy is designed employing a fixed-time sliding mode differentiator-based backstepping technique combined with a neural network-based minimum parameter learning approach (MLP-NN). This strategy addresses the "complexity explosion" issue arising from the differentiation of the virtual control law and model dynamic uncertainties separately. Moreover, a specialized prescribed performance function is incorporated to ensure that performance specifications of the formation tracking errors meet specific performance criteria. The integration of hysteresis quantizer aids in conserving save communication resources and significantly reduces the detrimental effects caused by quantization errors through the proposed control methodology. It is demonstrated that all closed-loop signals are practical fixed-time stable and formation tracking errors consistently adhere to the predetermined performance envelopes. The efficacy of proposed control strategy is validated through comprehensive numerical simulations. • A novel fixed-time tunnel prescribed performance is designed for the formation controller with its initial boundary distributed on the same side, which effectively reduces the overshooting of the system and facilitates the implementation of the control. • The finite precision control signals are employed in the proposed controller to achieve closed-loop system stability with limited communication data size. Which extends the practicability of the control scheme in practical applications. • Fixed-time observer is constructed to estimate the leader's information unknown, replacing the traditional transmission from leader to followers over the network. The scheme can decrease the bandwidth of the network to some extent. • The design scheme based on the fixed-time sliding mode differentiator-backstepping solves the problem of "complexity explosion" caused by the derivation of the virtual control law, and the differentiator error is compensated by the fractional power error compensation mechanism. [ABSTRACT FROM AUTHOR]

This paper presents a dynamic formation optimisation strategy that focuses on constructing an energy-saving formation for a fleet of unmanned surface vehicles (USVs) by considering two optimisation objectives: minimum energy consumption and minimum switching cost. The dynamic formation optimisation strategy based on robust optimisation over time (ROOT) is used to determine the energy-saving formation with the optimal position layout. The formation switching cost is first proposed to define the energy consumption cost of switching from the original position to the new position. Dynamic environments are established using the modified moving peak benchmark (mMPB) to simulate the movement of waves. A new swarm intelligence algorithm for solving ROOT is designed using an adaptive gradient training predictor and a solver embedded the dingo optimisation algorithm. The ROOT results are compared with the tank test results, which confirms that the dynamic formation optimisation strategy has a high credibility. The energy-saving formation optimised by dynamic formation optimisation based on ROOT not only ensures the minimum energy consumption, but also ensures the minimum switching cost. The survival time of the spacing configuration is prolonged, indicating that the acquired energy-saving formation can be consistently utilized over an extended period without necessitating positional adjustments. This mitigates the escalating energy consumption resulting from frequent formation switching when a fleet encounters dynamic marine conditions. • Dynamic optimise an energy-saving formation for a fleet of USVs. • Robustness over time in terms of survival time. • Energy-saving formation has a longer maintenance time. • Formation reconstruction considers switching cost. • Solver embeds with bioinspired swarm intelligence algorithm. [ABSTRACT FROM AUTHOR]

Due to the complexity and variability of the ocean environment, it is difficult for high-speed unmanned surface vehicles (USVs) to construct a unified mathematical model for nonlinear motion under different Froude numbers. In this paper, a nonparametric modeling method for high-speed USVs at three speed regions is presented based on Gaussian process regression with a hybrid kernel function (HKF-GPR). First, the hybrid kernel functions are designed based on the forecasting performance of Gaussian process regression with different single kernel functions (SKF-GPR) using sea trial speed data from the "Jiuhang 750" USV. Second, to better capture the dynamic performance of the USVs, the sea trial speed data are clustered into three speed regions by an improved k-means++ algorithm; these regions are the displacement, semiplanning, and planning regions. Finally, USV modeling and forecasting are performed in the three speed regions using HKF-GPR. The trained model shows good modeling accuracy and forecasting, verifying the effectiveness of the proposed method. • The data set is the real sea trial data collected by "Jiuhang 750" High-Speed Unmanned surface vehicle. • The design of hybrid kernel functions based on the Gaussian process regression prediction performance of different single kernel functions. • The sea trial data are clustered into three speed regions by an improved k-means++ algorithm, which are displacement, semi-planning, planning regions. • The USV modeling and forecasting are performed in the three speed regions using HKF-GPR. • The modeling of High-speed USV at the three regions can improve the modeling and forecasting accuracy of SVR, BP neural network and GPR algorithms, where HKF-GPR modeling algorithm exhibits optimal modeling forecasting accuracy. [ABSTRACT FROM AUTHOR]

This paper deals with the tracking control problem of small autonomous tethered underwater vehicles. It proposes a new extended robust integral of the sign of the error (RISE) feedback control. The proposed RISE-based extension benefits from a fuzzy inference system to automatically and online tune the parameters of the RISE controller. The resulting intelligent control scheme is named Fuzzy RISE (FRISE) feedback control. Several real-time experimental scenarios, in different operating conditions, were conducted on Leonard underwater vehicle to demonstrate the efficiency and robustness of the proposed control scheme. It was also compared with some existing controllers from the literature to show its performances. • A new intelligent RISE feedback control of autonomous tethered underwater vehicles is proposed. • A nominal RISE feedback control is enhanced based on a fuzzy logic inference system. • The resulting intelligent RISE is validated in real-time experiments on Leonard AUV. • Real-time experiments demonstrate the effectiveness and robustness of the proposed controller w.r.t existing controllers, namely a PID and a conventional RISE controller. [ABSTRACT FROM AUTHOR]

This paper delves into the challenge of heading control for unmanned surface vehicles (USVs) in conditions with unknown dynamics and environmental disturbances. Adaptive neural synergetic controllers are introduced for the purpose of effectively steering the USV, ensuring precise heading control to track and maintain the desired angle, even in the presence of environmental disturbances such as sea winds and waves. Firstly, different forms of macro variables and evaluation constraints are introduced to design the synergetic controllers. Then, to estimate the unknown dynamics and environmental disturbances, a single input single output RBF neural network is developed. Besides, to ensure tracking at the desired angle, an adaptive neural synergetic law is derived from the Lyapunov stability theorem. Finally, the simulation results compare the control effect of heading controllers with different macro variables and evaluation constraints and verify the effectiveness of the proposed control strategy. The feasibility of the adaptive neural synergetic controller is verified through the USV experiment. • A comprehensive comparison of USV heading controllers from various macro variables and evolution constraints. • A novel adaptive neural synergetic controller for USV heading system with unknown dynamics and external disturbances. • The experiment was conducted on a USV with controller implemented on Raspberry Pi to directly control the motors. [ABSTRACT FROM AUTHOR]