Visual servoing control method and system driven by oil outlet pose perception
By employing a method of view frustum point cloud generation, semantically enhanced instance segmentation, and hybrid visual servo driving, the accuracy and adaptability issues in oil port pose estimation and servo control were resolved, enabling efficient and reliable automated oil extraction operations.
CN122131810BActive Publication Date: 2026-07-14STATE GRID INTELLIGENCE TECHNOLOGY CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- STATE GRID INTELLIGENCE TECHNOLOGY CO LTD
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-14
Smart Images
Figure CN122131810B_ABST
Abstract
The application belongs to the technical field of electric power robots. A visual servo control method and system driven by oil taking port pose perception are provided, a frustum region is generated through target detection frame mapping to filter point cloud noise; a network embedded with a self-attention mechanism is used for semantic enhancement instance segmentation to obtain a target mask; residual compensation and three-dimensional bounding box prediction are combined to convert the point cloud to a real coordinate system to solve the accurate pose; finally, a hybrid visual servo strategy is adopted: based on pose error control at a long distance, and switched to image feature control at a short distance, efficient and accurate automatic docking of the oil taking robot is realized.
Need to check novelty before this filing date? Find Prior Art