Intrinsically-Motivated Robot Learning of Bayesian Probabilistic Movement Primitives

We present an approach for internally-guided learning in the context of a multi-task robot skill acquisition framework. More specifically, we focus on learning a parametrized distribution of robot movement primitives by using active intrinsically-motivated learning. We focus on the case where the learning process is initialized with human demonstrations, and refined through experiences. Such approach aims at combining experiential and observational learning. We demonstrate the effectiveness of our approach on a waste throwing task with a simulated 7-DoF Franka Emika robot.