Related references
Note: Only part of the references are listed.
Article
Engineering, Marine
Chao Pan et al.
Summary: This paper proposes a fully data-driven distributed control approach based on model-based deep reinforcement learning for multiple under-actuated unmanned surface vehicles (USVs) with fully unknown models, in order to achieve a desired formation and collision avoidance. The dynamic models of each USV are approximated by training a deep neural network with recorded input and output data. Model predictive formation controllers are then proposed to achieve safe formation control while considering collision avoidance. Simulation results demonstrate the feasibility and efficacy of the proposed method.
Article
Engineering, Electrical & Electronic
Zhouhua Peng et al.
Summary: This article proposes a safety-aware constrained anti-disturbance control method for the automatic berthing of maritime autonomous surface ships (MASSs) in a constrained water region. The method incorporates a line-of-sight guidance scheme and an extended state observer to achieve position-heading stabilization and compensate for ocean disturbances. Simulation results demonstrate the effectiveness of the proposed control law in ensuring the input-to-state safety of MASSs.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Automation & Control Systems
Hao Wang et al.
Summary: This article introduces a new cooperative path following control scheme for a team of AUVs, where each AUV guarantees individual path following and evenly disperses on a curve through a containment control approach. The proposed control strategy eliminates the assumption of second-order derivative of the reference path, offers a globally uniformly ultimately bounded path following control structure, and achieves coordination between multiple AUVs.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Engineering, Marine
Jun Ning et al.
Summary: This paper addresses the distributed formation control problem for multiple under-actuated unmanned surface vehicles (USVs) in the presence of input quantization, external disturbances, and model uncertainties. A two-level distributed guidance and neuro-adaptive quantized control architecture is proposed to achieve a time-varying formation. Simulation results demonstrate the effectiveness of the proposed method.
Article
Automation & Control Systems
Zhouhua Peng et al.
Summary: Autonomous surface vehicles (ASVs) are marine vessels capable of operating without a crew in various water/ocean environments, and coordinating multiple ASVs for complex missions offers enhanced capability and efficacy. Challenges in coordinated control of ASVs include their diversity, intravehicle interactions, collision avoidance requirements, and limited communication bandwidth in sea environments.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Engineering, Civil
Yong Ma et al.
Summary: This paper proposes a co-design method for fault detection filter and controller for a networked-based unmanned surface vehicle system. It utilizes an event-triggering communication scheme and the piecewise Lyapunov functional approach for stability analysis. Simulation results demonstrate the effectiveness of the method in ensuring safe and stable operation of the system while reducing data transmissions.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Engineering, Marine
Hongqiang Sang et al.
Summary: To address the issues of USVs, deploying multiple USVs as a formation fleet is a trend to enhance system autonomy and fault-tolerant resilience. A novel multiple sub-target artificial potential field (MTAPF) algorithm is proposed to optimize trajectory generation and assist USVs in avoiding local minimums by improving APF.
Article
Engineering, Marine
Xun Yan et al.
Summary: This paper presents a formation generation algorithm and formation obstacle avoidance strategy for multiple unmanned surface vehicles (USVs) that combines a virtual structure and artificial potential field to achieve high accuracy in formation shape maintenance and flexibility in formation shape change. The improved dynamic window approach is utilized to address obstacle avoidance for the multi-USV system, ensuring that the USV formation can navigate around obstacles while maintaining its shape. The combination of virtual structure and artificial potential field results in fewer calculations, allowing for real-time performance and ease of deployment on actual USVs.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Zhouhua Peng et al.
Summary: This article introduces reduced- and full-order data-driven adaptive disturbance observers (DADOs) for estimating unknown input gains and total disturbance of maritime autonomous surface ships. The proposed DADOs offer simultaneous estimation of total disturbance and input gains with guaranteed convergence through data-driven adaption. Simulation results validate the efficacy of the proposed DADO approach for model-free trajectory tracking control of autonomous surface ships without prior knowledge of their dynamics.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yujiao Zhao et al.
Summary: This study addresses the problem of path following for underactuated unmanned surface vessels formation using a modified deep reinforcement learning with random braking approach. A formation control model based on deep reinforcement learning is developed, along with a novel random braking mechanism to prevent training from getting stuck in local optima. A virtual leader-based path-following system is proposed to automatically adjust formation and maintain flexibility even when some vessels deviate, with simulation results verifying the effectiveness and superiority of the control strategy.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Engineering, Marine
Guoqing Zhang et al.
Summary: This paper introduces an adaptive neural event-triggered formation control strategy for underactuated surface vehicles (USVs) in marine environments. The strategy uses a logic virtual ship (LVS) guidance principle and neural networks (NNs) based observer to generate smooth reference paths and estimate velocities for followers, ensuring formation stability. The proposed fault-tolerant control algorithm compensates for uncertainties and actuator failures, reducing communication burden and achieving semi-global uniform ultimate bounded (SGUUB) stability. Simulation experiments confirm the effectiveness of the strategy.
Article
Computer Science, Artificial Intelligence
Zezhi Sui et al.
Summary: The article proposes a method based on deep reinforcement learning to solve the collision avoidance problem in the leader-follower formation control structure, consisting of two stages: imitation learning and reinforcement learning. A compound reward function and formation-oriented network structure are designed to guide the training process effectively.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Engineering, Marine
Caoyang Yu et al.
Article
Engineering, Marine
Zehua Jia et al.
Article
Engineering, Marine
Joohyun Woo et al.
Article
Automation & Control Systems
Zhouhua Peng et al.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2018)
Article
Automation & Control Systems
Zhijian Sun et al.
Article
Automation & Control Systems
Shen Yin et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2017)
Article
Engineering, Ocean
Yuanchang Liu et al.
APPLIED OCEAN RESEARCH
(2016)
Article
Engineering, Marine
Yuanchang Liu et al.
