An Object Detection and Pose Estimation Approach for Position Based Visual Servoing

Abstract

 In this paper, an object recognition method and a pose estimation approach using stereo vision is presented. The proposed approach will be used for position based visual servoing of a 6 DoF manipulator. The object detection method is designed in the purpose of increasing robustness based on using RGB color-based object descriptor and an online correction. Pose is estimated by using the depth information derived from stereo vision camera and an SVD based method. Transformation between desired pose and object pose will be calculated and later it is used for position based visual servoing. The proposed object recognition method is invariant to scale, orientation and lighting condition which increases the level of robustness. The error of stereo vision camera can reach 1 mm. The accuracy of stereo vision camera is adequate for tasks like grasping and manipulation.