3D grating mounting self-learning error compensation method and system

By employing a single industrial camera and a three-level layered error decoupling control method during the 3D grating mounting process, the problem of high-precision mounting error between 3D gratings and display panels was solved, achieving sub-micron level precision and high-efficiency mass production, which is suitable for advanced packaging in optical communication.

CN122415754APending Publication Date: 2026-07-17SHENZHEN GOLD TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN GOLD TECH CO LTD
Filing Date
2026-06-16
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies for micron-level high-precision mounting of 3D gratings and display panels suffer from problems such as dual-camera coordinate positioning errors, contradictions between linear compensation models and nonlinear multi-source coupling errors, offline calibration schemes being unable to adapt to mass production dynamic errors, continuous decay of compensation effects, and disconnect between algorithms and mounting processes.

Method used

A single industrial camera is used to acquire images in the same visual coordinate system. Combined with a three-level hierarchical error decoupling and collaborative control architecture, including mechanical error pre-compensation, glass warpage real-time compensation and self-learning dynamic closed-loop compensation methods, dynamic suppression of multi-source errors and accuracy maintenance are achieved through improved DH transformation matrix method, binocular stereo vision method and incremental PID control.

Benefits of technology

It achieves submicron-level multi-source error suppression, improves 3D display performance, reduces equipment costs, increases mass production yield and cycle time efficiency, breaks through the precision bottleneck, and is suitable for advanced optical communication packaging scenarios such as NPO/CPO optical engines.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415754A_ABST
    Figure CN122415754A_ABST
Patent Text Reader

Abstract

本发明涉及一种3D光栅贴装自学习误差补偿方法及系统,该方法包括:利用设置在固定基准上的单台工业相机,分别采集COG显示面板和3D光栅玻璃的Mark点图像,计算获得初始对位偏差;构建三级分层误差解耦与协同控制架构,实现对多源耦合误差的全链路动态抑制与稳态精度维持;基于修正参数驱动机械手和运动平台完成贴合,并检测实际贴合偏差,并更新补偿模型参数,形成闭环控制。本发明通过同坐标单相机硬件架构与三级分层自学习补偿算法的协同创新,从源头消除系统误差并实现亚微米级多源误差抑制,在大幅降低设备成本、提升量产良率与节拍效率的同时,突破了倒装贴装的精度瓶颈,显著提升了3D显示性能并具备向CPO光引擎等光通讯封装场景延展的能力。
Need to check novelty before this filing date? Find Prior Art