一种无人艇自主靠泊方法、装置、电子设备及存储介质
By preprocessing historical berthing data of unmanned surface vessels (USVs) and training models, a conditional variational autoencoder based on the Transformer architecture was constructed. This solved the robustness and adaptability issues of the autonomous berthing control method for USVs in complex environments, and achieved high-precision and high-stability autonomous berthing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HARBIN ENG UNIV
- Filing Date
- 2026-05-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing autonomous berthing control methods for unmanned surface vessels lack robustness in complex, unstructured marine environments, have poor environmental adaptability, rely on limited sensing capabilities, exhibit non-smooth action sequences, and have limited generalization ability, making it difficult to achieve high-precision and highly stable autonomous berthing.
By preprocessing historical berthing operation data, extracting state features and difference features, constructing a conditional variational autoencoder model based on the Transformer architecture, training it, generating a berthing action prediction model, and using real-time data for inference to output stable and continuous control commands.
It improves the accuracy and stability of autonomous docking of unmanned surface vessels in complex environments, enhances the integrity and reliability of perception information, improves the coherence and generalization performance of action sequences, and achieves high-precision and high-stability autonomous docking.
Smart Images
Figure CN122172841B_ABST