Autonomous exploration of motor skills is a key capabilityoflearningroboticsystems.Learningmotorskills can be formulated as inverse modeling problem, which targets at finding an inverse model that maps desired outcomes in some task space, e.g., via points of a motion, to appropriateactions, e.g.,motion control policy parameters. Inthis paper, autonomous exploration of motor skills is achieved by incrementally learning inverse models starting from an initial demonstration. The algorithm is referred to as skill babbling, features sample-efficient learning, and scales to high-dimensionalactionspaces.Skillbabblingextendsideas of goal-directed exploration, which organizes exploration in thespaceofgoals.Theproposedapproachprovidesamodularframeworkforautonomousskillexplorationbyseparating thelearningoftheinversemodelfromtheexplorationmechanism and a model of achievable targets, i.e. the workspace. The effectiveness of skill babbling is demonstrated for a range of motor tasks comprising the autonomous bootstrapping of inverse kinematics and parameterized motion primitives.

EB00000003506K

Available

• Autonomous exploration of motor skills by skill babbling