一种移动机器人视觉轨线识别与偏离修正方法
By fusing visual and inertial data and using adaptive control, the problem of a single visual sensor being susceptible to environmental influences in mobile robot trajectory recognition was solved, thereby improving the stability and accuracy of robot trajectory tracking and enabling trajectory correction in complex scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI HENGZE FUHUI INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2026-04-16
- Publication Date
- 2026-07-17
AI Technical Summary
Current mobile robot trajectory recognition relies on a single vision sensor, which is susceptible to changes in lighting and environmental factors, leading to the failure of feature point detection. Furthermore, it lacks multi-source information fusion and degradation processing mechanisms, making it difficult to achieve continuous deviation prediction and correction, resulting in insufficient trajectory tracking accuracy and robustness.
The system employs visual and inertial data fusion, extracts and assigns confidence scores through dual-path feature extraction, calculates deviations using inverse perspective transformation, and directly inputs the data into the adaptive coupling controller when visual information is normal. When degraded, the system predicts the deviation through extended Kalman filtering, outputs the drive wheel correction amount, and periodically updates the sensor parameters.
It improves the stability and accuracy of robot trajectory tracking in complex scenarios, can adaptively correct trajectory deviations, reduce errors caused by environmental factors, and improve the reliability of robot operation in complex environments.
Smart Images
Figure CN122015833B_ABST