2009 · 20 citations · 13 references
EngineeringComputer ScienceRobotics EngineeringMobile RoboticsRobot LearningRoboticsGrasp PlanningManipulation LearningObject ManipulationReach-to-grasp ControlRobot ManipulationRobot MorphologyGrasp LocalizationLearning-based ManipulationManipulation Under UncertaintyRobot GraspingRobotic Manipulation Learning
What it means for an object to be dasiawithin reachpsila depends very much on the morphology and skills of a robot. In this paper, we enable a mobile manipulation robot to learn a concept of PLACE from which successful manipulation is possible through trial-and-error interaction with the environment. Due to this developmental approach, PLACE is very much grounded in observed experience, and takes the hardware and skills of the robot into account. During task-execution, this model is used to determine optimal grasp places in a least-commitment approach. This PLACE takes into account uncertainties in both robot and target object positions, and leads to more robust behavior.
13
Cellular networks underlying human spatial navigation
Arne D. Ekstrom, Michael J. Kahana, Jeremy B. Caplan et al. · Nature · 2003 · 1.5K citations
Large Scale Multiple Kernel Learning
Sören Sonnenburg, Gunnar Rätsch, Christin Schäfer et al. · MPG.PuRe (Max Planck Society) · 2006 · 1.2K citations · Full text