Ship autonomous navigation decision method and device, storage medium and computer equipment

CN122239725APending Publication Date: 2026-06-19XIAMEN UNIV OF TECH

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: XIAMEN UNIV OF TECH
Filing Date: 2026-05-22
Publication Date: 2026-06-19

AI Technical Summary

Technical Problem

Existing autonomous navigation technologies for amphibious vessels have significant shortcomings in areas such as cross-domain multimodal data fusion, complex environment perception, water-land mode switching control, multi-constraint path planning, and perception-decision-control link coordination. These shortcomings make it difficult to meet the autonomous and intelligent navigation requirements of high-safety scenarios such as port inspection, emergency rescue, and inland waterway operations.

Method used

By acquiring multimodal datasets and performing data augmentation, a knowledge graph and unified embedded features for the navigation domain are constructed, a risk cost function is generated, path decision-making and local optimization are performed, and navigation control is carried out by combining a water-land coupled motion model and a digital twin world model, thereby improving the accuracy of cross-domain environmental perception and path optimization.

Benefits of technology

It improves the accuracy of cross-domain environmental perception, ensures the safety and efficiency of navigation routes, dynamically adapts to changes in dynamic characteristics during the water-land transition, and enhances the reliability and safety of decision-making.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122239725A_ABST

Patent Text Reader

Abstract

The ship autonomous navigation decision-making method, apparatus, storage medium, and computer equipment provided in this application first perform cross-domain feature fusion on the ship's multimodal dataset and navigation domain knowledge graph within the waterway area during autonomous navigation decision-making, thereby effectively improving the accuracy of cross-domain environmental perception. Then, a risk cost function constructed from the fused features is used to calculate the risk cost of the waterway area, resulting in a navigation risk map. Subsequently, a globally optimal path is generated on this map based on navigation constraints to ensure the safety and efficiency of the path from a macroscopic perspective. Furthermore, the path is locally optimized using a water-land coupled motion model, enabling it to dynamically adapt to changes in the ship's dynamic characteristics during the transition between water and land areas. After each optimization, this application can perform navigation risk simulation verification on the optimized path, executing only the locally optimized paths that pass the verification, further improving the reliability of ship autonomous navigation decisions.

Need to check novelty before this filing date? Find Prior Art