Computer vision is gaining significant importance as a cheap, passive, and information-rich sensor in research areas such as unmanned vehicles, medical robotics, human-machine interaction, autonomous navigation,robotic manipulation and grasping. However, a current trend is to build computer vision systems that are used to perform a specific task which makes it hard to reuse the ideas across different disciplines. In this paper, we concentrate on vision strategies for robotic manipulation tasksin a domestic environment. This work is an extension of our ongoing work on a development of a general vision system for robotic applications. Inparticular, given fetch-and-carry type of tasks, the issues related to the whole detect-approach-grasp loop are considered.
© 2006 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. QC 20120111