An unmanned autonomous detection method for tunnel security risk hidden dangers
By combining global path detection and local cost maps with a dynamic trajectory initial value generator, the problem of real-time, robust, and safe flight of UAVs in tunnels was solved, achieving efficient path planning and autonomous detection in complex environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 应急管理部大数据中心
- Filing Date
- 2026-02-03
- Publication Date
- 2026-06-05
AI Technical Summary
Existing UAV trajectory planning technologies struggle to achieve real-time, robust, safe, and control-friendly path planning in complex and dynamic environments, especially in tunnels where collision risks surge, optimization processes become unstable, and semantic gaps exist between planning modules and control systems.
A safe initial trajectory is generated by combining a global detection path with a local cost map and a dynamic trajectory initial value generator. An optimized trajectory is then generated through constrained trajectory optimization to meet the dynamic feasibility and safe distance constraints of the UAV, and environmental perception and risk identification are updated in real time.
It enables real-time autonomous detection of UAVs in complex environments, improves the robustness of trajectory optimization and system reliability, expands application capabilities in extreme scenarios, and ensures flight safety and smoothness.
Smart Images
Figure CN122149452A_ABST