Authors: Haili Xu; Longbiao Zhu; Jian Zhuang; Sun'an Wang
Addresses: School of Mechanical Engineering, Nantong University, Jiangsu 226019, China ' School of Mechanical Engineering, Nantong University, Jiangsu 226019, China ' School of Mechanical Engineering, Xi'an Jiaotong University, Shaanxi 710049, China ' School of Mechanical Engineering, Xi'an Jiaotong University, Shaanxi 710049, China
Abstract: A novel method of object pose estimation based on three dimensional registration for assembly robot is presented. The spatial registration parameters can be computed by finding which projection of the volume is the most similar to the camera image. We compute the Fourier Transform (FT) of the volume projection by using FT projection theorem. A cost function is designed according to the magnitude of projection FT and image FT. Registration parameters is obtained by using Complicated System Genetic Algorithm (CSGA) to minimise the cost function. Assembly experiment made on our robot system shows our method is efficient and practical.
Keywords: pose estimation; assembly robots; 3D registration; three dimensional models; 3D modelling; volume projection; Fourier transform; camera images; CSGA; complicated system genetic algorithms; robot pose; assembly automation.
International Journal of Signal and Imaging Systems Engineering, 2012 Vol.5 No.1, pp.11 - 17
Available online: 04 May 2012 *Full-text access for editors Access for subscribers Purchase this article Comment on this article