视觉与倾角传感器组合相对位姿测量方法研究

发布时间：2018-01-21 16:51

本文关键词： 位姿测量倾角传感器参数标定组合测量单目视觉　出处：《天津大学》2016年硕士论文　论文类型：学位论文

【摘要】：空间物体的位姿是反映物体的重要参数,随着空间物体位姿测量需求的不断提升,空间物体的位姿测量越来越受到人们的重视,广泛应用在工业、军事、大型装备制造、航空航天等领域,传统的空间物体相对位姿测量方法从原理上可以概括为四大类:基于激光跟踪测量系统的位姿测量方法、基于惯性导航系统的位姿测量方法、基于机械定位约束的位姿测量方法和基于计算机视觉系统的视觉位姿测量方法。本文针对视觉相对位姿测量中存在的问题,研究了一种视觉与倾角传感器组合的相对位姿测量方法,通过在相机和被测物上各放置一个倾角传感器,利用固定倾角传感器所提供的角度信息提供额外的约束来求解Pn P(Perspective-n-Point)问题,文章介绍了组合相对位姿测量方法的原理及其结构参数标定,为了提高算法的精度和抗噪能力,在求得线性解的基础上进一步进行非线性优化,仿真实验验证了该方法的正确性,并与现存的Pn P问题求解方法比较,表明本课题研究的Pn P问题求解方法具有较高精度并且对噪声干扰具有较高的稳定性,最后,实验结果表明本文研究的方法对空间15个特征点在x方向和y方向的平均反投影误差均小于为0.2 pixel,能够满足空间物体位姿测量精度需求。本文主要的研究工作和创新点如下:1、广泛查阅国内外相关研究文献资料,分析位姿测量研究的意义和国内外研究现状,阐述本课题的研究价值和意义。2、针对传统视觉位姿测量求解过程中存在的问题,研究了视觉与倾角传感器组合相对位姿测量方法,利用视觉和倾角传感器的测量信息求解Pn P问题,详细推导了组合位姿测量原理,设计组合位姿测量算法,为了提高算法的求解精度和抗噪能力,在求得线性解的基础上进一步进行非线性优化,提高求解精度,并进行仿真实验,验证组合测量原理的正确性。3、归纳总结了组合位姿测量图像传感器标定的基本理论和方法,根据相机成像模型,推导相机内外参数标定原理,完成了相应算法的编写与仿真分析。4、根据视觉与倾角传感器组合位姿测量原理,研究了组合位姿测量系统结构参数标定方法,推导组合位姿测量系统结构参数标定原理,并利用仿真实验验证组合位姿测量系统结构参数标定原理的正确性。5、设计编写视觉与倾角传感器组合位姿测量系统相关算法,并进行实验分析各测量因素对组合位姿测量精度的影响。
[Abstract]:The pose of space object is an important parameter to reflect the object. With the increasing demand of position and attitude measurement of space object, more and more attention has been paid to the pose measurement of space object, which has been widely used in industry and military. In the field of large-scale equipment manufacturing, aerospace and other fields, the traditional relative position and attitude measurement methods of space objects can be summarized into four categories in principle: the position and attitude measurement method based on laser tracking measurement system. The position and pose measurement method based on inertial navigation system, the position and pose measurement method based on mechanical positioning constraint and the vision position and attitude measurement method based on computer vision system. In this paper, a method of measuring the relative position and attitude of the combination of vision and inclination sensor is studied. One inclination sensor is placed on the camera and the object being tested. The angle information provided by the fixed inclination sensor is used to provide additional constraints to solve the problem of PN Perspective-n-Point. This paper introduces the principle of combined relative pose measurement method and its structural parameter calibration. In order to improve the accuracy and anti-noise ability of the algorithm, the nonlinear optimization is carried out on the basis of finding the linear solution. The simulation results show that the method is correct and compared with the existing PN P problem solving method. It shows that the solution method of PN P problem studied in this paper has high accuracy and high stability to noise interference. Finally. The experimental results show that the average backprojection error of the method is less than 0. 2 pixel in x direction and y direction for 15 feature points in space. The main research work and innovation of this paper are as follows: 1. This paper analyzes the significance of the position and pose measurement research and the current research situation at home and abroad, expounds the research value and significance of this topic, aiming at the problems existing in the traditional vision pose measurement process. The relative pose measurement method of the combination of vision and inclination sensor is studied. The measurement information of vision and inclination sensor is used to solve the PN P problem, and the principle of combined position and attitude measurement is deduced in detail. In order to improve the accuracy and anti-noise ability of the combined pose measurement algorithm, the nonlinear optimization is carried out on the basis of the linear solution, and the accuracy of the solution is improved, and the simulation experiment is carried out. Verify the correctness of the combined measurement principle. 3. Summarize the basic theory and method of the combined position and attitude measurement image sensor calibration. According to the camera imaging model, the principle of camera internal and external parameters calibration is deduced. The corresponding algorithm is compiled and simulated. 4. According to the principle of combined position and attitude measurement of vision and inclination sensor, the calibration method of structural parameters of combined position and attitude measurement system is studied. The principle of structure parameter calibration of combined position and attitude measurement system is deduced, and the correctness of structure parameter calibration principle of combined position and attitude measurement system is verified by simulation experiment. The related algorithms of the combined position and pose measurement system of the vision and inclination sensor are designed and the influence of various measurement factors on the combined position and pose measurement accuracy is analyzed experimentally.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.41;TP212

【相似文献】