一种基于多传感器融合的智能助盲方法及系统
By using multi-sensor fusion technology, LiDAR, IMU and GNSS are used for 3D environment modeling and trajectory planning, which solves the problem that existing white cane systems cannot express the 3D environment and have coarse navigation. This enables fine-grained navigation and dynamic obstacle avoidance for visually impaired people, with small errors and high real-time performance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- UNIV OF SCI & TECH BEIJING
- Filing Date
- 2024-02-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing white cane systems cannot fully express 3D environmental information, and navigation methods cannot provide fine-grained guidance, which poses challenges for visually impaired people when traveling.
Employing multi-sensor fusion technology, point cloud data is acquired through LiDAR, combined with IMU and GNSS, and voxels are used as a scene representation method for stereo modeling. Semantic occupancy information and occupancy flow are used for environmental perception and trajectory planning, and global navigation information and local environmental information are fused to provide fine-grained navigation guidance.
It achieves full representation of the 3D environment and prediction of dynamic obstacles, can dynamically avoid obstacles during navigation, provides clear point-level path navigation with small errors and high real-time performance, and is suitable for safe travel for visually impaired people.
Smart Images
Figure CN118038143B_ABST