基于并联理论的单目视觉位姿估计

耿明超1; 赵铁石1; 边辉1; 唐启敬3

doi:10.3788/OPE.20132110.2617

您当前的位置：

首页 >

文章列表页 >

基于并联理论的单目视觉位姿估计

微纳技术与精密机械 | 更新时间：2020-08-12

- 基于并联理论的单目视觉位姿估计
- Pose estimation of monocular vision based on parallel theory
- 光学精密工程 2013年21卷第10期页码：2617-2626
- 作者机构：
  
  1. 燕山大学河北省并联机器人与机电系统实验室，河北秦皇岛&nbsp
  2. Key Laboratory of Advanced Forging &amp
  3. Technology and Science of the Ministry of Education
  4. 上海航天电子技术研究所上海,201109
  5. 先进锻压成形技术与科学教育部重点实验室，河北秦皇岛&nbsp
- 作者简介：
- 基金信息：
  
  非惯性系并联式多轴稳定平台机构学理论与实验研究()
- DOI：10.3788/OPE.20132110.2617
  中图分类号： TP242.62;TP273
- 收稿日期：2013-03-19，
  
  修回日期：2013-05-08，
  
  网络出版日期：2012-10-19，
  
  纸质出版日期：2013-10-15
- 稿件说明：
移动端阅览
耿明超赵铁石边辉唐启敬. 基于并联理论的单目视觉位姿估计[J]. 光学精密工程, 2013,21(10): 2617-2626

GENG Ming-chao ZHAO Tie-shi BIAN Hui TANG Qi-jing. Pose estimation of monocular vision based on parallel theory[J]. Editorial Office of Optics and Precision Engineering, 2013,21(10): 2617-2626
耿明超赵铁石边辉唐启敬. 基于并联理论的单目视觉位姿估计[J]. 光学精密工程, 2013,21(10): 2617-2626 DOI： 10.3788/OPE.20132110.2617.

GENG Ming-chao ZHAO Tie-shi BIAN Hui TANG Qi-jing. Pose estimation of monocular vision based on parallel theory[J]. Editorial Office of Optics and Precision Engineering, 2013,21(10): 2617-2626 DOI： 10.3788/OPE.20132110.2617.

摘要

搭建了一种主动式测量系统用于测试伺服稳定平台的伺服精度。以此系统为基础，提出了基于并联理论的单目视觉位姿测量方法。介绍了系统的硬件组成和系统测量原理，分析了系统的分辨率。给出了激光点在投影靶坐标系下的计算原理及过程。然后，从并联机构的自由度出发，将视觉测量系统等效地转化为一个并联机构，将视觉测量系统的位姿估计问题转换为并联机构的正解问题。利用并联机构的运动影响系数迭代求解系统的位姿，由并联机构关节螺旋直接得到影响系数，由此简化了推导过程。仿真和实验结果表明：系统的姿态测量精度为0.05。该方法能够快速稳定的收敛，基本达到了系统的设计要求。

Abstract

An active measurement system was put up to measure the servo accuracy of a servo stable platform. On the basis of the system

a new pose estimation method of monocular vision was proposed according to the parallel theory. First

the hardware composition and measuring principle of the system were introduced and its resolution was analyzed. The calculation principle and calculation process of the world coordinates of laser points were given. Then

from the freedom of a parallel mechanism

the visual measurement system was transformed into a parallel mechanism equivalently

and the pose estimation problem was transformed into a forward solution problem of parallel mechanism. Furthermore

the kinematics influence coefficient of the parallel mechanism was used to estimate the pose of the system. Instead of complex derivations

the kinematics influence coefficient could be obtained directly. The simulation and experimental results show that the orientation measurement accuracy of the system is 0.05

which concludes that the method has a rapid and stable convergence for the design requirements of the system.

关键词

Keywords

references

朱枫,何雷,郝颖明,等. 视觉位姿测量中单目视觉与双目立体视觉测量精度比较[J].仪器仪表学报，2007，28(4):165-169.ZHU F, HE L, HAO Y M, et al.. Contrast of calculated accuracy between monocular vision an binocular vision without modeling [J]. Chinese Journal of Scientific Instrument, 2007, 28(4):165-169. (in Chinese)[2]马颂德，张正友. 计算机视觉计算理论与算法基础[M].北京：科学出版社，1997.MA S D, ZHANG ZH Y. Computer Vision-The Theory of Computation and Basis of Algorithm [M]. Beijing: Science Press, 1997. (in Chinese)[3]刘峰，王向军，许薇，等. 并联跟踪台单目视觉自标定方法[J].红外与激光工程，2009，38(1): 175-179.LIU F, WANG X J, XU W, et al.. Self-calibration method for the parallel tracking platform based on the monocular vision [J]. Infrared and Laser Engineering, 2009, 38(1):175-179. (in Chinese)[4]ROBERT M H. Determining camera parameters from the perspective projection of a rectangle [J]. Pattern Recognition, 1989, 22(3):225-230.[5]高学海，徐科军，张翰，等. 基于单目视觉和激光测距仪的位姿测量算法[J].仪器仪表学报，2007，28(8): 1479-1485.GAO X H, XU K J, ZHANG H, et al.. Position-pose measurement algorithm based on single camera and laser range-finder [J]. Chinese journal of Scientific Instrument, 2007, 28(8):396-401. (in Chinese)[6]魏振忠，赵征，张广军. 空间圆姿态识别二义性的角度约束消除[J].光学精密工程，2010，18(3): 1669-1673.WEI ZH ZH, ZHAO ZH, ZHANG G J. Solution of duality in pose esteimation of single circle using Euclidean angular constraint [J]. Opt. Precision Eng., 2010, 18(3): 685-691. (in Chinese)[7]张磊，徐科军，赵锐，等. 一种基于圆特征的单目视觉位姿测量算法的改进[J].合肥工业大学学报：自然科学版，2009，32(11): 685-691.ZHANG L, XU K J, ZHAO R, et al.. Improvement of position and orientation measurement algorithm of monocular vision based on circle features [J]. Journal of Hefei University of Technology：Natural Science, 2009, 32(11):685-691. (in Chinese)[8]许微. 基于几何特征量的自主位姿测量方法的研究[D].天津：天津大学，2008.XU W. Study on the self-measurement method of the attitude and position based on geometrical characteristic parameter [D].Tianjin: Tianjin University, 2008. (in Chinese)[9]肖旭. 单目视觉位姿测量方法的研究[D].天津：天津大学，2008.XIAO X. Study on Position-pose measurement method of monocular vision [D].Tianjin: Tianjin University, 2008. (in Chinese)[10]ZHANG SH J, GAO X B, ZHANG F, et al.. Monocular vision based iterative pose estimation algorithm from corresponding feature points [J]. Science China, 2010, 53(8):1682-1696. [11]孙泽林，王昭，李明. 火炮稳定精度图像测试系统[J].光学精密工程，2012，20(1):157-164.SUN Z L, WANG ZH, LI M. Image test system for gun stabilization accuracy [J]. Opt. Precision Eng., 2012, 20(1): 157-164. (in Chinese) [12]BATISTA J,ARAUJO H,ALMEIDA A T. Pose view stability analysis for camera look angles computation [C]. Proceedings of the 13th International Conference on Pattern Recognition, Vienna, Austria, ICPR, 1996: 171-175.[13]韩延祥，张志胜，戴敏. 用于目标测距的单目视觉测量方法[J].光学精密工程，2011，19(5): 1110-1116.HAN Y X, ZHANG ZH SH, DAI M. Monocular vision system for distance measurement based on feature points [J]. Opt. Precision Eng., 2011, 19(5):1110-1116. (in Chinese) [14]黄真，赵永生，赵铁石. 高等空间机构学[M].北京：高等教育出版社，2005.HUANG ZH, ZHAO Y SH, ZHAO T SH. Advanced spatial mechanism [M]. Beijing: Higher Education Press, 1997. (in Chinese)

浏览量

131

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于CAD模型的单目六自由度位姿测量

2UU-PSU上肢康复机构运动学与动力学研究

并联坐标测量机研究进展

基于单目视觉的火箭回收高度测量技术研究

面向激光跟踪测量的大范围高精度姿态测量