Vision Based 3D Model Calibration for Articulated Objects  

Oleh:

Jinshi, Cui ; Huabin, Tang ; Zengqi, Sun

Jenis: Article from Article
Dalam koleksi: Final Program and Book of Abstracts: The 4th Asian Control Conference, September 25-27, 2002 (Sep. 2002), page 1467-1470.
Topik: CCD Camera; Articulated Objects; Robot; Enviroment
Fulltext: AC021590.PDF (402.22KB)

Isi artikel

A computer vision system that consists of two CCD cameras is fixed in the partly-known robot work environment. The aim of the intelligent vision system is to catch the sequence stereo images within the robot motion, then work out the appearance shape and pose of the robot manipulator accurately and timely. In this paper, rotational joint axes of motoman-Up6 manipulator should be accurately calibrated, followed by the pose estimation of each link. Through tracking the feature points on the appropriate link with stereo cameras, the motion curve of each feature point can be obtained and thus the relative pose of rotational axes of each link can be determined with two methods, one is position based approach and the other is image based approach, which can both leads to the estimation of the kinemat ics parameters, then the pose of related link can be estimated by matching the feature position in the image frame with that on the model frame.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali