%0 Conference Paper %B Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on %D 2004 %T An active learning approach for assessing robot grasp reliability %A Antonio Morales %A Eris Chinellato %A Fagg, A.H. %A Angel P. del Pobil %K active learning approach %K Costs %K Grasping %K Haptic interfaces %K Intelligent robots %K Laboratories %K learning (artificial intelligence) %K manipulators %K motor control %K Motor drives %K online estimation %K Reliability %K reliability assessment capabilities %K robot grasp reliability %K Robot sensing systems %K Torso %K Training data %K Uncertainty %K visually-guided grasp selection %X

Learning techniques in robotic grasping applications have usually been concerned with the way a hand approaches to an object, or with improving the motor control of manipulation actions. We present an active learning approach devised to face the problem of visually-guided grasp selection. We want to choose the best hand configuration for grasping a particular object using only visual information. Experimental data from real grasping actions is used, and the experience gathering process is driven by an on-line estimation of the reliability assessment capabilities of the system. The goal is to improve the selection skills of the grasping system, minimizing at the same time the cost and duration of the learning process.

%B Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on %G eng %R 10.1109/IROS.2004.1389399 %F grasping, learning %0 Conference Paper %B Intelligent Robots and Systems, 2003. (IROS 2003). Proceedings. 2003 IEEE/RSJ International Conference on %D 2003 %T Experimental prediction of the performance of grasp tasks from visual features %A Antonio Morales %A Eris Chinellato %A Fagg, A.H. %A Angel P. del Pobil %K adaptive behavior %K Barrett hand %K dexterous manipulators %K estimation rule %K feature extraction %K Geometry %K grasp configuration %K Grasping %K hand kinematics %K humanoid robot %K Humans %K Image reconstruction %K Intelligent robots %K Kinematics %K Laboratories %K manipulator kinematics %K object image %K performance prediction %K prediction theory %K Reliability %K Robot sensing systems %K robot vision %K Robustness %K Service robots %K three finger grasps %K unmodeled objects %K visual features %K visually guided grasping %X

This paper deals with visually guided grasping of unmodeled objects for robots which exhibit an adaptive behavior based on their previous experiences. Nine features are proposed to characterize three-finger grasps. They are computed from the object image and the kinematics of the hand. Real experiments on a humanoid robot with a Barrett hand are carried out to provide experimental data. This data is employed by a classification strategy, based on the k-nearest neighbour estimation rule, to predict the reliability of a grasp configuration in terms of five different performance classes. Prediction results suggest the methodology is adequate.

%B Intelligent Robots and Systems, 2003. (IROS 2003). Proceedings. 2003 IEEE/RSJ International Conference on %G eng %R 10.1109/IROS.2003.1249685