Abstract:Automatic collision avoidance decision making for vessels is a critical challenge in the development of autonomous ships and has become a central point of research in the maritime safety domain. Effective and systematic collision avoidance strategies significantly reduce the risk of vessel collisions, ensuring safe navigation. This study develops a multi-vessel automatic collision avoidance decision-making method based on deep reinforcement learning (DRL) and establishes a vessel behavior decision model. When designing the reward function for continuous action spaces, the criteria of the "Convention on the International Regulations for Preventing Collisions at Sea" (COLREGs) were adhered to, taking into account the vessel's collision risk under various encounter situations, real-world navigation practices, and navigational complexities. Furthermore, to enable the algorithm to precisely differentiate between collision avoidance and the navigation resumption phase in varied vessel encounter situations, this paper incorporated "collision avoidance decision making" and "course recovery decision making" as state parameters in the state set design, from which the respective objective functions were defined. To further enhance the algorithm's performance, techniques such as behavior cloning, residual networks, and CPU-GPU dual-core parallel processing modules were integrated. Through simulation experiments in the enhanced Imazu training environment, the practicality of the method, taking into account the effects of wind and ocean currents, was corroborated. The results demonstrate that the proposed algorithm can perform effective collision avoidance decision making in a range of vessel encounter situations, indicating its efficiency and robust generalization capabilities.

COLREGs-Adaptive trajectory planning and decision-making in maritime autonomous surface ships

COLREG-Compliant Optimal Path Planning for Real-Time Guidance and Control of Autonomous Ships

Multi-Ship Collision Avoidance Decision-Making Based on Collision Risk Index

Parallel Trajectory Planning for Shipborne Autonomous Collision Avoidance System

Intelligent collision avoidance decision-making method for unmanned ships based on driving practice

Vessel Behavior Decision-making and Planning for Autonomous Surface Vessels in Inland Waterway

COLREGS-based Path Planning for Ships at Sea Using Velocity Obstacles.

Multi-functional and Practical Adaptive Collision Avoidance Decision-making System for Autonomous Ships

A COLREGs-Compliant Deep Reinforcement Learning Approach

Multi-ship collision avoidance decision-making and coordination mechanism in Mixed Navigation Scenarios

Autonomous ship navigation with an enhanced safety collision avoidance technique

Collision Avoidance and Path Point Tracking Control for Underactuated Unmanned Surface Vehicles with Unknown Model Nonlinearity

A COLREGs-Compliant Motion Planner for Autonomous Maneuvering of Marine Vessels in Complex Environments

Unmanned Surface Vehicle Collision Avoidance Path Planning in Restricted Waters Using Multi-Objective Optimisation Complying with COLREGs

An integrated decision support system for ship collision avoidance

A Coordination System between Decision Making and Controlling for Autonomous Collision Avoidance of Large Intelligent Ships

Long-Range Risk-Aware Path Planning for Autonomous Ships in Complex and Dynamic Environments

A real-time multi-ship collision avoidance decision-making system for autonomous ships considering ship motion uncertainty

Autonomous Collision Avoidance at Sea: A Survey

A Real-Time Collision Avoidance Framework of MASS Based on B-Spline and Optimal Decoupling Control

Knowledge transfer enabled reinforcement learning for efficient and safe autonomous ship collision avoidance