起重机远程控制图像延时检测方法及系统
By combining edge computing and multimodal deep learning, image latency in crane remote control systems can be accurately detected and predicted, solving the problem of difficult-to-estimate video stream transmission latency and enabling efficient and safe remote operation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NINGBO SPECIAL EQUIP INSPECTION & RES INST
- Filing Date
- 2025-05-21
- Publication Date
- 2026-07-17
AI Technical Summary
In existing remote control systems for cranes, the transmission delay of video streams is difficult to estimate accurately, especially under complex and ever-changing network conditions, resulting in poor delay compensation and affecting operational accuracy and safety.
Edge computing servers are used for dual-channel parallel preprocessing. Adaptive Kalman filtering and multimodal deep learning models are combined to extract video stream features through temporal graph convolutional networks, generate delay prediction curves, and combine model prediction control algorithms to optimize video coding parameters and intelligent routing mechanisms to achieve adaptive time compensation.
Accurate detection and prediction of image latency improves the real-time performance and accuracy of remote control, enhances system robustness, reduces the impact of operation delays, and improves work efficiency and safety.
Smart Images
Figure CN120378603B_ABST