基于图像数据的自动搬运设备动态避障控制系统及方法
By constructing a forward load mask and extracting pseudo-foot points, the problems of misjudgment of forward load and mirrored obstacles in visual obstacle avoidance algorithms are solved, enabling precise obstacle avoidance control of automated transport equipment in narrow alleys and reducing the risk of collision between equipment and facilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHAOYANG LIAOXI CIVIL AIR DEFENSE ENG PROTECTION FACILITIES CO LTD
- Filing Date
- 2026-04-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing visual obstacle avoidance algorithms struggle to distinguish between the extended load structure of automated transport equipment and external obstacles, leading to misjudgments and mirrored obstacles. This affects the accuracy of obstacle avoidance control and fails to effectively handle the risk of lateral outward scanning, making it difficult to meet the dynamic obstacle avoidance requirements in narrow alleyways.
By constructing a forward load structure mask, eliminating interference from the equipment's own structure, extracting pseudo-foot points and calculating lateral and longitudinal control coefficients, and generating appropriate steering and speed commands, accurate conversion from the image plane to the ground plane and risk assessment are achieved.
It improves the accuracy of obstacle detection, reduces the risk of collisions between equipment and facilities, adapts to the dynamic obstacle avoidance control requirements in narrow alleyways, and enables efficient obstacle avoidance operation of equipment.
Smart Images
Figure CN122044175B_ABST