A pile position automatic correction system based on Beidou and AI vision

The automatic pile position correction system, which combines BeiDou inertial navigation with AI vision, solves the positioning accuracy and safety issues of pile position alignment technology in complex environments, and achieves high-precision and safe pile position correction.

CN122411067APending Publication Date: 2026-07-17CCCC SECOND HARBOR ENGINEERING CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CCCC SECOND HARBOR ENGINEERING CO LTD
Filing Date
2026-06-11
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing pile alignment technology suffers from insufficient positioning accuracy in complex construction environments, poor adaptability of control models to dynamic working conditions, and lack of active obstacle avoidance capabilities, resulting in low safety.

Method used

The system uses a BeiDou inertial navigation combined positioning unit to obtain global pose, combines an active optical reference field generation unit to project optical reference graphics, utilizes an AI visual perception unit to accurately and quantitatively measure pose deviation, and uses a predictive control unit to generate correction control commands based on the pile driver dynamics model, while performing local environmental perception and safety constraints to achieve dynamic correction.

Benefits of technology

It achieves pile position correction with centimeter-level accuracy, improves the system's adaptability and safety in complex environments, and ensures the high efficiency, stability, and safety of the pile driver in automated operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122411067A_ABST
    Figure CN122411067A_ABST
Patent Text Reader

Abstract

本发明涉及工程机械自动化控制技术领域,公开了一种基于北斗与AI视觉的桩位自动校正系统,包括北斗惯导组合定位单元、主动光学参照场生成单元、AI视觉感知单元以及预见控制单元。本系统通过结合北斗全局引导与AI视觉对投射光学图形的精测,解算出位姿偏差,克服了传统定位方法精度与可靠性不足的问题,通过采用可在线更新的神经网络混合动力学模型,能根据运动误差自主补偿因工况变化引入的模型漂移,保证了模型预见控制算法的长期精确性与适应性,此外,系统通过立体视觉重建局部三维环境以识别障碍物,并将其转化为实时安全约束,实现了主动智能避障,显著提升了自动化作业的安全性。
Need to check novelty before this filing date? Find Prior Art