ABSTRACT
The positioning of a robot hand in order to grasp an object is a problem fundamental to robotics. The task want to perform can be described as follows: given a visual scene the robot arm must reach an indicated point in that visual scene. This point indicates the observed object that has to be grasped. In order to accomplish this task, a mapping from the visual scene to the corresponding robot joint values must be available. In order to design a system which can be successfully used in real-world applications, there are two important issues which have to be considered. Firstly, in real-world applications of robot systems, a ‘reasonable’ training time must be ensured. Secondly, the added value of self-learning systems must be fully exploited: it is essential that the method adapt to unforeseen gradual or sudden changes in the robot–camera system. The vision problem is different: we faced with huge numbers of data.
