An obstacle detection system based on amphibious unmanned aerial vehicle (UAV) patrol.

By introducing active airflow disturbance and multimodal data acquisition, combined with dynamic feature analysis and physical interaction verification, the problem of existing obstacle detection systems being unable to distinguish between suspended and root-fixed aquatic plants in complex aquatic environments has been solved, achieving high-precision obstacle classification and path planning.

CN122307589APending Publication Date: 2026-06-30ZHEJIANG TIANYU INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG TIANYU INFORMATION TECH CO LTD
Filing Date
2026-06-02
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing obstacle detection systems struggle to distinguish between suspended obstacles and root-fixed aquatic plants in complex aquatic environments. In particular, under rotor airflow disturbances, they lack observation of the dynamic response of obstacles, resulting in inaccurate assessment of path planning.

Method used

By introducing an active airflow disturbance mechanism and synchronous acquisition of multimodal data, and through obstacle localization, initial screening, active disturbance, dynamic feature analysis and state classification modules, combined with physical interaction verification, the dynamic response features of obstacles can be extracted and classified.

Benefits of technology

It can accurately distinguish between root-fixed aquatic plants and floating obstacles on the water surface, improve classification accuracy, overcome optical interference from complex water surfaces, ensure that dynamic analysis resources are focused on the real target, and optimize task time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122307589A_ABST
    Figure CN122307589A_ABST
Patent Text Reader

Abstract

This invention discloses an obstacle detection system based on amphibious unmanned aerial vehicle (UAV) patrol flights, comprising an obstacle localization module, a preliminary screening module, an active disturbance module, a dynamic feature analysis module, and a state classification module. The localization module identifies and locates obstacles using UAV patrol flight images; the preliminary screening module combines image feature matching and physical interaction verification to filter target areas; the active disturbance module controls the UAV to hover, generating downdrafts and ensuring effective coverage; the dynamic feature analysis module extracts the displacement trajectory and motion pattern features of obstacles; and the classification module, through rule-based judgment and machine learning, accurately classifies obstacles into root-fixed or surface-suspended states. This invention, through active airflow disturbance and multi-dimensional dynamic feature analysis, overcomes the limitations of traditional static identification, accurately distinguishing between root-fixed and surface-suspended obstacles in complex waters, while improving detection efficiency and environmental adaptability, providing reliable support for route planning in amphibious UAV flow measurement missions.
Need to check novelty before this filing date? Find Prior Art