• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2019, Vol. 55 ›› Issue (24): 187-196.doi: 10.3901/JME.2019.24.187

• 运载工程 • 上一篇    下一篇

扫码分享

基于激光跟踪仪和机器视觉的飞机翼身对接装配偏差动态综合修正

朱永国1, 张文博1, 邓正平1, 刘春锋2   

  1. 1. 南昌航空大学航空制造工程学院 南昌 330063;
    2. 中航工业江西洪都航空工业集团有限责任公司部装一厂 南昌 330024
  • 收稿日期:2018-12-22 修回日期:2019-10-21 出版日期:2019-12-20 发布日期:2020-02-18
  • 通讯作者: 朱永国(通信作者),男,1978年出生,副教授。主要研究方向为飞机数字化装配。E-mail:zhuyongguo_2003@163.com
  • 作者简介:张文博,男,1993年出生,硕士研究生。主要研究方向为飞机数字化装配。E-mail:15797896015@163.com;邓正平,男,1989年出生,博士。主要研究方向为飞机数字化装配。E-mail:dengzhengpingjh@nuaa.edu.cn;刘春锋,男,1983年出生,高级工程师。主要研究方向为飞机装配工艺。E-mail:lcf_005@163.com
  • 基金资助:
    国家自然科学基金(51565042,51865037)、江西省重点研发计划(20143ACE50008,20171BBE50007)、江西省自然基金(20181BAB206029)、江西省教育厅基金(GJJ180532)和南昌市重大科技(2018-ZDKJGG-010)资助项目。

Dynamic Synthesis Correction of Deviation for Aircraft Wing-fuselage Docking Assembly Based on Laser Tracker and Machine Vision

ZHU Yongguo1, ZHANG Wenbo1, DENG Zhengping1, LIU Chunfeng2   

  1. 1. School of Aeronautical Manufacturing Engineering, Nanchang Hangkong University, Nanchang 330063;
    2. First Sub-assembly Factory, AVIC Jiangxi Hongdu Aviation Industry Group Company Ltd, Nanchang 330024
  • Received:2018-12-22 Revised:2019-10-21 Online:2019-12-20 Published:2020-02-18

摘要: 提出一种基于激光跟踪仪和机器视觉的飞机翼身对接装配偏差动态综合修正方法。在对合前机翼调姿阶段,利用人工鱼群粒子滤波算法对机翼调姿控制点坐标进行跟踪与估值,结合多项式轨迹规划与定位器运动学逆解,建立翼身对合前机翼位姿的修正数学模型。在翼身对合阶段,利用加权最小二乘异质传感器信息融合方法,将机器视觉和激光跟踪仪的检测数据进行融合滤波,建立翼身对合质量动态监控和对合偏差综合修正数学模型。试验表明,该方法能显著提升飞机翼身对接装配准确度,实现飞机大部件对接装配偏差的动态修正。

关键词: 机身, 机翼, 装配, 机器视觉, 激光跟踪仪

Abstract: A method of dynamic synthesis correction of deviation error is proposed for aircraft wing-fuselage docking assembly by using laser tracker and vision with the object of fork-ear joint. In the stage of wing pose adjustment, the artificial fish swarm particle filter is used to track and estimate coordinates of points on the wing. Trajectory planning in combination with the locator inverse kinematics, dynamic modification model of the wing pose is built. In the stage of wing-fuselage docking, the characteristic parameters of the joint is dynamically detected to get joint position errors by machine vision. The error based on laser tracker and machine vision are fused and filtered, according to the principle of weighted least square of heterogeneous sensor information fusion. By this way, the models of dynamic monitoring of docking quality and comprehensive correction of docking deviation are also built. The experiments show that the proposed method can effectively improve quality of wing-fuselage docking assembly, and dynamic synthesis correction of deviation of wing-fuselage docking assembly is realized.

Key words: fuselage, wing, docking, vision, laser tracker

中图分类号: