Ship path planning method based on deep reinforcement learning

Through the improved ID3QN-PER deep reinforcement learning structure and artificial potential field force, combined with the B-spline curve, the inefficiency and safety problems of traditional methods in path planning in complex marine environments are solved, and a smooth path planning that conforms to the maritime collision avoidance rules is achieved.

CN120406424APending Publication Date: 2025-08-01JIANGSU OCEAN UNIV
3 Cites 8 Cited by

Patent Information

Application Number
CN202510346924.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-24
Publication Date
2025-08-01

AI Technical Summary

Technical Problem

Traditional ship path planning methods are difficult to take into account real-time and global optimization when facing complex maritime environments, and have low learning efficiency in high-dimensional state spaces, path redundant and non-smooth, and it is difficult to meet the constraints of maritime collision avoidance rules in dynamic environments.

Method used

The improved ID3QN-PER deep reinforcement learning structure is adopted, combining artificial potential field force and B-spline curves, and a comprehensive reward function is designed. Through a dual network structure and a priority experience replay mechanism, the compliance with maritime collision avoidance rules and path smoothness is achieved.

Benefits of technology

It improves the learning efficiency and stability of ship path planning, ensures the safety and smoothness of paths, can meet maritime collision avoidance rules in a dynamic environment, and improves navigation efficiency and safety.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a ship path planning method based on deep reinforcement learning. The ship path planning method is suitable for a complex sea area environment. Through environment modeling and dynamic obstacle identification, a detailed state space is constructed, and perception of the environment is enhanced by using an artificial potential field. An ID3QN-PER algorithm is adopted to train a model, an adaptive exploration strategy and a network parameter updating mechanism are designed, and navigation safety and efficiency in a dynamic environment are ensured. The method further comprises the steps of discretizing an action space, designing an obstacle avoidance strategy in combination with a COLREGS specification, smoothing a route through a B-spline technology, adjusting a navigation strategy according to real-time environment data, and improving the real-time performance and the global optimization performance of route planning.
Need to check novelty before this filing date? Find Prior Art