一种基于视觉识别的挖掘机防倾翻方法、系统及挖掘机
By combining visual recognition technology with multi-level protection strategies and using the YOLOv8 model to identify the excavator's operating mode and hydraulic support status, the problem of single stability control and high false alarm rate of traditional wheeled excavator anti-tipping systems has been solved, achieving precise anti-tipping protection in complex environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QINGDAO LOVOL EXCAVATOR
- Filing Date
- 2025-12-22
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional anti-tipping systems for wheeled excavators rely on sensors for judgment, have a single stability control strategy, a high false alarm rate, cannot judge the status of hydraulic supports in real time, and are easily affected by vibration and mud cover.
It adopts visual recognition technology combined with multi-level protection strategy. By acquiring key point images of the excavator in real time and using YOLOv8-OBB and YOLOv8-Pose models to identify the operation mode and hydraulic support status, the microcontroller unit executes the anti-tipping protection strategy according to priority. It integrates dual AI cameras and infrared supplementary light module.
It enables accurate identification of operating modes and hydraulic support status in complex environments, reduces false alarm rates, provides flexible safety protection, ensures that excavators work efficiently under extreme conditions and avoid the risk of tipping over, and adapts to all-weather operation in harsh working conditions.
Smart Images
Figure CN121429062B_ABST