工业机器人快速标定的误差分析研究

发布时间：2018-02-11 18:15

本文关键词： 零位标定运动学参数标定误差分析柔性误差　出处：《南京理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着工业机器人在各领域越来越广泛的应用,其对精度也提出了更高的要求,现有的标定技术具有一定的成熟度且精度较高,但依旧存在着标定方法的误差和标定模型的不完整,这也就给标定精度的提升又提供了一定的空间,因此有关工业机器人快速标定的误差分析研究就有更加显著的意义。本文对工业机器人标定方法的误差分析进行了研究,并基于柔性误差对运动学参数标定的影响,改进了误差模型,进而从两个方面进一步提高了 IRB120工业机器人的绝对定位精度。本文首先对工业机器人的运动学基础知识进行了详细阐述,基于经典的D-H模型和改进的D-H模型(MDH模型)分别建立了运动学模型,并进行了正运动学的推导和逆运动学问题的求解方法的研究,另外,分别从误差模型、约束模型和优化模型三个方面对基于点约束的零位标定和基于球面约束的运动学参数标定进行了阐述,为后面研究提供了基础理论的准备。针对标定方法的误差分析这方面,在原标定系统中为PSD云台建立了工件坐标系,进而对静态位姿和激光线在PSD上投影偏移进行了误差分析,推导出各运动学参数和投影偏移之间的误差关系,并结合IRB120进行了相应的误差实例分析,实现了标定方法的误差分析。另一方面,分析工业机器人的误差来源,针对关节柔性误差进行了较为详细的研究分析工作,包括分析和建立关节柔性误差模型,并将其融合到了 IRB120运动学参数标定仿真系统中;依据建立的误差模型,结合本文具体机型分析了其关节柔性变化量,从而验证了关节柔性的影响不能忽略,并对其变化量进行了误差灵敏度分析,为后续模型改进提供前期基础工作。由于关节柔性误差的影响并不能忽略,从而对原有的零位标定和运动学参数标定的误差模型进行了相应改进,添加了对柔性误差的考虑,并通过Matlab仿真和零位标定实验验证了改进模型的效果确实优于原来的模型,进而最终从原先预期的两个方面提升了标定精度。
[Abstract]:With the more and more extensive application of industrial robot in various fields, its precision is also higher. The existing calibration technology has a certain maturity and high precision. However, there are still errors in calibration methods and incomplete calibration models, which provides some space for the improvement of calibration accuracy. Therefore, the research on the error analysis of industrial robot calibration is more significant. In this paper, the error analysis of the calibration method of industrial robot is studied, and based on the effect of flexible error on kinematic parameter calibration, The error model is improved, and the absolute positioning accuracy of IRB120 industrial robot is further improved from two aspects. Firstly, the basic knowledge of kinematics of industrial robot is described in detail in this paper. Based on the classical D-H model and the modified D-H model / MDH model, the kinematics model is established, and the forward kinematics derivation and inverse kinematics problem solving method are studied. The zero calibration based on point constraint and kinematic parameter calibration based on spherical constraint are described in three aspects of constraint model and optimization model. In the original calibration system, the workpiece coordinate system is established for the PSD cloud head, and then the error analysis of the static pose and the projection migration of the laser line on the PSD is carried out, and the error relationship between the kinematics parameters and the projection offset is deduced. The error analysis of calibration method is realized with IRB120. On the other hand, the error source of industrial robot is analyzed, and the flexible joint error is studied and analyzed in detail. The flexibility error model of joint is analyzed and established, and it is integrated into the IRB120 kinematics parameter calibration simulation system, according to the established error model, the change of joint flexibility is analyzed based on the specific model of this paper. It is proved that the influence of joint flexibility can not be ignored, and the error sensitivity analysis of its variation is carried out, which provides the basis for the subsequent improvement of the model. Because the effect of joint flexibility error can not be ignored, the influence of joint flexibility error can not be ignored. Therefore, the error model of zero calibration and kinematic parameter calibration is improved, and the flexibility error is considered. The effect of the improved model is proved to be better than that of the original model by Matlab simulation and zero calibration experiment. Finally, the calibration accuracy is improved from the two aspects originally expected.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242.2

【相似文献】