... for Offshore Intelligent Navigational Aids using Deep Reinforcement Learning Ruolan Zhang, Chenhui Zhao and Mingyang Pan Navigation college, Dalian Maritime University, Dalian, China ABSTRACT Maritime facilities are increasingly electrified and digitalized as smart technology is widely applied. Offshore...

... ships. The main control methods for trajectory tracking include: backstepping method; Optimal control; Robust control; Model predictive control; Control based on reinforcement learning, etc. Dong et al obtained a universal tracking controller for both straight and curved trajectories of unmanned ships...

... ABSTRACT With the increasing demand for autonomous navigation of ships, ship trajectory tracking has gained significant attention. This paper aims to use deep reinforcement learning (DRL) to select adaptive Proportional Integral Derivative (PID) parameters and develop a trajectory tracking...

... ABSTRACT In recent years, some control designs have been studied using reinforcement learning (RL) and succeeded in each control task. In the framework of RL, the design of reward function influences the resultant control policy. In this study, two control policies are designed using different...

... ABSTRACT This paper proposes model-based reinforcement learning for trajectory tracking of surface ships. To avoid the exploration of reinforcement learning in the physical environment, we construct a maneuvering simulation environment consisting of a ship maneuvering model, an actuator...

.... reinforcement learning artificial intelligence ship avoidance risk management obstacle planning & scheduling machine learning decision-making algorithm r-ddpg experiment coefficient ocean engineering avoidance scenario obstacle avoidance apf trajectory scenario architecture Proceedings...

... real-time control strategies for Wave Energy Converters (WECs), this study proposes a real-time model for latching control of the WEC by coupling Deep Reinforcement Learning (DRL) and Computational Fluid Dynamics (CFD). An agent-based optimization framework for the model is established to achieve end...

... performance of the robotic fish, we establish a threedimensional motion model that incorporates both the soft fishtail part and rigid fish body part. We develop a lightweight control strategy based on reinforcement learning, which is successfully implemented onboard and trained in a physical flow field...

... ABSTRACT Aiming at the problems of low degree of automation, large individual differences in manual operation and low average efficiency of dredging vessels, an intelligent control method of cutter suction dredgers based on reinforcement learning is proposed. First, control variables...

... Deep Deterministic Policy Gradient (MTD3) algorithm is adapted to train a controller for obstacle avoidance. A Markov decision process model including state, action, and reward functions is designed, and a suitable reinforcement-learning training environment is designed. The obstacle-avoidance ability...

... ABSTRACT This paper proposes a deep reinforcement learning algorithm for the thorough evaluation of the overall performance of ships in order to appropriately assess the performance of surface ships. The computer can quickly adjust the indication weights in accordance with the experts...

... robot information actuation trajectory neural network machine learning ray deflection reinforcement learning curet complexity artificial intelligence fin algorithm amplitude operation international ocean control policy application patankar tracking controller kalman filter...

... model and thus results in the inefficiency or even infeasibility of the conventional methods. In this paper, we propose a novel deep reinforcement learning based method to solve the AUV path planning problem. Specifically, we focus on the AUV path planning problem where the environment map is unknown...

... (Deep reinforcement learning) technology coupled with traditional numerical simulation models, a free wave-making model is trained. The original case can only generate regular waves, but the observed waves are usually irregular waves to solve practical problems. Qu (2016) studied the comparison...

... reinforcement learning breakwater amplitude artificial intelligence free surface elevation surface elevation wave condition interaction upstream oil & gas machine learning flat plate wave flume plate breakwater wave dissipation performance dissipation wave dissipation water...

... artificial intelligence controller simulation metals & mining algorithm reinforcement learning dsmv mining vehicle upstream oil & gas validation reinforcement learning-based controller reward function neural network coordinate system history machine learning training step...

... are identified using a reinforcement learning technique. The proposed method would be suitable in identifying the direct and cross coupling damping and restoring parameters for a ship heaving and pitching under the action of unknown excitations effected by a realistic sea state. Published data, results...

..., the accumulative error of each model can be large. To address this issue, this paper presents an end-to-end USV tracking control method via deep reinforcement learning, where a modern Reinforcement learning algorithm Actor-Critic is adopted. Given no prior knowledge of the dynamical system, the proposed method...

... Deterministic Policy Gradient (DDPG) which is one of Reinforcement Learning(RL) algorithms is proposed. The advantage of using DDPG is that it does not require any prior knowledge about dynamics of a ship and environmental disturbances. Instead, the DDPG algorithms may learn them on their own by interacting...