Scholarlink

<p>Positioning accuracy is critical in industrial robotics, and calibration is one of the most effective methods to enhance it. To address the problems of parameter coupling and ill-conditioned parameters in the kinematic calibration of industrial robots, a multi-constraint model is proposed in this paper. The kinematic parameters are identified by grouping based on the Levenberg–Marquardt algorithm and Genetic Algorithm (LM-GA). Firstly, an industrial robot kinematic model is developed based on the Denavit-Hartenberg (D-H) model. Subsequently, a parameter identification model is established using both distance constraints and position constraints. A grouping strategy for parameter identification is utilized to reduce parameter coupling. Furthermore, the effectiveness and robustness of this proposed method in this paper are validated through simulations. Finally, experimental validation is conducted. The method proposed in this study is compared with the calibration method by grouping parameters based on distance constraint and the traditional method based on position constraint. Experimental results show that the proposed method improves the positioning accuracy of the robot by 74.45% (reducing the averaged positioning error at 20 validation positions from 1.55&#xa0;mm to 0.39&#xa0;mm). This method can effectively reduce ill-conditioned parameters, achieving comprehensive calibration of all kinematic parameters, and provides technical support for industrial robots to perform various high-precision positioning tasks.</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 novel calibration method of 6-DOF industrial robots based on a multi-constraint model by grouping parameters

  • 期刊论文

  • 发表日期：

    2025年11月6日

查看全文
  • Zeng Kang,
  • Ling Wang,
  • Chunxiang Zhu,
  • Xiai Chen,
  • Suan Xu,
  • Binrui Wang

摘要

Positioning accuracy is critical in industrial robotics, and calibration is one of the most effective methods to enhance it. To address the problems of parameter coupling and ill-conditioned parameters in the kinematic calibration of industrial robots, a multi-constraint model is proposed in this paper. The kinematic parameters are identified by grouping based on the Levenberg–Marquardt algorithm and Genetic Algorithm (LM-GA). Firstly, an industrial robot kinematic model is developed based on the Denavit-Hartenberg (D-H) model. Subsequently, a parameter identification model is established using both distance constraints and position constraints. A grouping strategy for parameter identification is utilized to reduce parameter coupling. Furthermore, the effectiveness and robustness of this proposed method in this paper are validated through simulations. Finally, experimental validation is conducted. The method proposed in this study is compared with the calibration method by grouping parameters based on distance constraint and the traditional method based on position constraint. Experimental results show that the proposed method improves the positioning accuracy of the robot by 74.45% (reducing the averaged positioning error at 20 validation positions from 1.55 mm to 0.39 mm). This method can effectively reduce ill-conditioned parameters, achieving comprehensive calibration of all kinematic parameters, and provides technical support for industrial robots to perform various high-precision positioning tasks.