首页 | 本学科首页   官方微博 | 高级检索  
     


Symbolic computation of robot models for geometric parameters identification with singularity analysis
Authors:Said M Megahed
Affiliation:(1) Mechanical Design and Production Department, Faculty of Engineering, Cairo University, 12316 Giza, Egypt
Abstract:The computation of the geometrical parameters identification model of a robot arm leads to an intrinsic extended Jacobian matrix. This matrix is often singular and its manual computation is tedious and may lead to error. A computer assisted procedure is developed to compute this matrix symbolically and to check its singularity. The program GPIM is developed in Pascal and works on PC's for this purpose. Jacobian singularity is checked using its symbolic expressions with the aid of a proposed table for robot arm parameters to obtain a minimum set of identified parameters. This program can be used for simple chain robot arms having revolute (R) and/or prismatic (P) joints including the case of consecutive near parallel axes. TH8 robot arm of type RPPRRR having two pairs of consecutive near parallel axes is studied to show the program efficiency and how singularity is eliminated.Nomenclature a i Denavit-Hartenberg parameter. - C thetavi ,C agri andC thetav12 cos theta i , cos agr i and cos (theta 1 + theta 2). - J 0, J Basic Jacobian and Jacobian matrices. - J a , J r , J theta, J agr and J beta Sub-Jacobian matrices. - J 0a , J 0r , J 0theta, J 0agr and J 0beta Intrinsic sub-Jacobian matrices. - m Work space dimensions. - n Number of moving links of the robot arm. - p Number of pairs of consecutive near parallel axes. - P i,i+1 and P li 3×1] position vectors. - Q pi , Q ai , Q ri , Q agri , Q betai and Q thetai 4×4] differential operator matrices. - r i Denavit-Hartenberg parameter. - R l , R i , R i+1 and R n+1 Base, links i–1, i and n coordinate frames. - R i,i+1 and R 1i 3×3] orientation matrices. - S thetai , S agri and S theta12 sin theta i , sin agr i and sin theta 1 + theta 2. - T i,i+1 and T 1i 4×4] homogeneous transformation matrices. - X m×1] robot arm operational coordinates. - 
$$X_{i,i + 1} ,\hat X_{i,i + 1}  and X_{1i} ,\hat X_{1i}$$
1st columns of the orientation matrices, and associated tensors. - 
$$Y_{i,i + 1} ,\hat Y_{i,i + 1}  and Y_{1i} ,\hat Y_{1i}$$
2nd columns of the orientation matrices, and associated tensors. - 
$$Z_{i,i + 1} ,\hat Z_{i,i + 1}  and Z_{1i} ,\hat Z_{1i}$$
3rd columns of the orientation matrices, and associated tensors. - agr i Denavit-Hartenberg parameter. - beta i Twist angle parameter. - deltaX Changes in robot arm operational coordinates. - deltaPHgr, deltaPHgrls and deltaPHgrrr Changes in robot arm geometrical parameters, and their least square and ridge regression estimated changes. - theta i Denavit-Hartenberg parameter. - PHgr, PHgr c and PHgr n Robot geometrical parameters and their correct and nominal values. - OHgrpi, OHgrai, OHgrri, OHgragri, OHgrbetai and OHgrthetai 4×4] differential operator matrices. - pgr and pgr–1 3×3] conversion matrix and its inverse.
Keywords:Parameter identification  estimation techniques  robot geometrical parameters  Jacobian matrix
本文献已被 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号