TY - CONF T1 - An active learning approach for assessing robot grasp reliability T2 - Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on Y1 - 2004 A1 - Antonio Morales A1 - Eris Chinellato A1 - Fagg, A.H. A1 - Angel P. del Pobil KW - active learning approach KW - Costs KW - Grasping KW - Haptic interfaces KW - Intelligent robots KW - Laboratories KW - learning (artificial intelligence) KW - manipulators KW - motor control KW - Motor drives KW - online estimation KW - Reliability KW - reliability assessment capabilities KW - robot grasp reliability KW - Robot sensing systems KW - Torso KW - Training data KW - Uncertainty KW - visually-guided grasp selection AB -
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.
JF - Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on ER - TY - CONF T1 - Experimental prediction of the performance of grasp tasks from visual features T2 - Intelligent Robots and Systems, 2003. (IROS 2003). Proceedings. 2003 IEEE/RSJ International Conference on Y1 - 2003 A1 - Antonio Morales A1 - Eris Chinellato A1 - Fagg, A.H. A1 - Angel P. del Pobil KW - adaptive behavior KW - Barrett hand KW - dexterous manipulators KW - estimation rule KW - feature extraction KW - Geometry KW - grasp configuration KW - Grasping KW - hand kinematics KW - humanoid robot KW - Humans KW - Image reconstruction KW - Intelligent robots KW - Kinematics KW - Laboratories KW - manipulator kinematics KW - object image KW - performance prediction KW - prediction theory KW - Reliability KW - Robot sensing systems KW - robot vision KW - Robustness KW - Service robots KW - three finger grasps KW - unmodeled objects KW - visual features KW - visually guided grasping AB -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.
JF - Intelligent Robots and Systems, 2003. (IROS 2003). Proceedings. 2003 IEEE/RSJ International Conference on ER -