一种用于移动机器人的视觉标记自动选择方法及系统
By adaptively selecting visual markers for observation and fusing odometry data for optimization, the problem of insufficient positioning accuracy of mobile robots in complex environments is solved, achieving high-precision robot pose estimation and meeting the autonomous navigation needs of smart agricultural environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI HENGZE FUHUI INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-17
AI Technical Summary
Under complex working conditions, existing technologies struggle to achieve high-precision estimation of mobile robot pose while ensuring real-time performance. This is especially true in long-distance and complex environments, where high noise levels in single-frame visual tag decoding and severe cumulative odometry errors result in insufficient positioning accuracy to meet the precision requirements of high-density planting environments.
By acquiring continuous monocular image frame sequences and odometry data, an uncertainty covariance matrix is adaptively allocated, visual marker observations that meet the accuracy threshold are selected, and a weighted optimization objective function is constructed. Odometry data is then fused for nonlinear optimization, and representative observations with the least uncertainty are selected to improve positioning accuracy.
It significantly improves positioning accuracy and robustness in long-distance and complex scenarios, ensuring high-precision autonomous navigation of mobile robots in smart agricultural environments.
Smart Images
Figure CN122115546B_ABST