A path planning strategy unified with a COLREGS collision avoidance function based on deep reinforcement learning and artificial potential field

Unmanned ship navigates on the water in an autonomous or semiautonomous way, which can be widely used in maritime transportation, intelligence collection, maritime training and testing, reconnaissance, and evidence collection. In this paper, we use deep reinforcement learning to solve the optimization problem in the path planning and management of unmanned ships. Specifically, we take the waiting time (phase and duration) at the corner of the path as the optimization goal to minimize the total travel time of unmanned ships passing through the path. We propose a new reward function, which considers the environment and control delay of unmanned ships at the same time, which can reduce the coordination time between unmanned ships at the same time. In the simulation experiment, through the quantitative and qualitative results of deep reinforcement learning of unmanned ship navigation and path angle waiting, the effectiveness of our solution is verified.

/pdf/deep-reinforcement-learning-based-path-control-and-1rquzz53.pdf

Deep Reinforcement Learning-Based Path Control and Optimization for Unmanned Ships

Dynamic anti-collision A-star algorithm for multi-ship encounter situations

Uninterrupted path planning system for Multi-USV sampling mission in a cluttered ocean environment

Ship path planning based on safety potential field in inland rivers

Design of UDE-based dynamic surface control for dynamic positioning of vessels with complex disturbances and input constraints

The invention relates to a visual unmanned ship and a visual navigation method thereof. The visual unmanned ship comprises a shipborne GPS_Beidou navigation module, an attitude detection module, a binocular camera module, a laser radar module, a satellite communication module and a core processor module, wherein the binocular camera module is used for detecting obstacles of a ship bow and a stern;the laser radar module is used for assisting the binocular camera so as to realize automatic obstacle avoidance; the satellite communication module is used for realizing communication between the unmanned ship and a shore-based system; and the core processor module collects and processes data information of the GPS_Beidou navigation module, the attitude detection module, the binocular camera module and the laser radar module in real time, and makes a ship power driving module track and set a route to travel in real time according to a control instruction. In the invention, rapid high-precision target detection and distance measurement, target tracking or target avoidance can be realized, the unmanned ship system does not have a superstructure, construction cost is saved, a resistance in anavigation process is reduced, a shaking amplitude and overturning danger in heavy stormy weather are decreased, and a cargo loading space is increased.

Youqiang Huang

Papers

A path planning strategy unified with a COLREGS collision avoidance function based on deep reinforcement learning and artificial potential field

Design of UDE-based dynamic surface control for dynamic positioning of vessels with complex disturbances and input constraints

Visual unmanned ship and visual navigation method thereof