Scholarlink

<p>The widespread adoption of automated material handling systems in wafer fabrication has imposed stricter positioning accuracy requirements on wafer transfer robotic arms. However, the complex and dynamic operational environments of these systems pose substantial challenges to the efficiency and cost-effectiveness of conventional calibration techniques. To address this, we propose a rapid calibration method that integrates geometric and non-geometric error compensation, enabling high-precision robotic arm positioning across diverse working conditions. First, forward kinematic equations are derived using kinematic modeling theory to systematically analyze how geometric errors affect end-effector positioning accuracy; kinematic calibration is then achieved through parameter identification. Second, a random forest-based error prediction model is developed, incorporating multi-source operational variables such as speed and load to enable real-time compensation of non-geometric dynamic errors. Experimental results demonstrate that the combined compensation significantly improves the robotic arm’s positioning accuracy, reducing the peak error by over 70% and the average error by more than 80%. The proposed methodology provides a practical and cost-effective solution for the efficient calibration of wafer transfer robotic arms in advanced manufacturing.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Method for Enhancing the Position Accuracy of Semiconductor Wafer Handling Robots via Kinematic and Multi-condition Prediction Models

  • 期刊论文

  • 发表日期：

    2026年2月20日

查看全文
  • Huijun Yue,
  • Yongshuo Li,
  • Ruiji Wang,
  • Haibo Lv,
  • Guiping Xie,
  • Yaoting Wu

摘要

The widespread adoption of automated material handling systems in wafer fabrication has imposed stricter positioning accuracy requirements on wafer transfer robotic arms. However, the complex and dynamic operational environments of these systems pose substantial challenges to the efficiency and cost-effectiveness of conventional calibration techniques. To address this, we propose a rapid calibration method that integrates geometric and non-geometric error compensation, enabling high-precision robotic arm positioning across diverse working conditions. First, forward kinematic equations are derived using kinematic modeling theory to systematically analyze how geometric errors affect end-effector positioning accuracy; kinematic calibration is then achieved through parameter identification. Second, a random forest-based error prediction model is developed, incorporating multi-source operational variables such as speed and load to enable real-time compensation of non-geometric dynamic errors. Experimental results demonstrate that the combined compensation significantly improves the robotic arm’s positioning accuracy, reducing the peak error by over 70% and the average error by more than 80%. The proposed methodology provides a practical and cost-effective solution for the efficient calibration of wafer transfer robotic arms in advanced manufacturing.