Movement Primitives are a well-established\ud paradigm for modular movement representation and\ud generation. They provide a data-driven representation\ud of movements and support generalization to novel situations,\ud temporal modulation, sequencing of primitives\ud and controllers for executing the primitive on physical\ud systems. However, while many MP frameworks exhibit\ud some of these properties, there is a need for a uni-\ud fied framework that implements all of them in a principled\ud way. In this paper, we show that this goal can be\ud achieved by using a probabilistic representation. Our\ud approach models trajectory distributions learned from\ud stochastic movements. Probabilistic operations, such as\ud conditioning can be used to achieve generalization to\ud novel situations or to combine and blend movements in\ud a principled way. We derive a stochastic feedback controller\ud that reproduces the encoded variability of the\ud movement and the coupling of the degrees of freedom\ud of the robot. We evaluate and compare our approach\ud on several simulated and real robot scenarios

This paper proposes an interaction learning method for collaborative and assistive robots based on movement primitives. The method allows for both action recognition and human-robot movement coordination. It uses imitation learning to construct a mixture model of human-robot interaction primitives. This probabilistic model allows the assistive trajectory of the robot to be inferred from human observations. The method is scalable in relation to the number of tasks and can learn nonlinear correlations between the trajectories that describe the human-robot interaction. We evaluated the method experimentally with a lightweight robot arm in a variety of assistive scenarios, including the coordinated handover of a bottle to a human, and the collaborative assembly of a toolbox. Potential applications of the method are personal caregiver robots, control of intelligent prosthetic devices, and robot coworkers in factories.Keywords Movement primitives · physical humanrobot interaction · imitation learning · mixture model · action recognition · trajectory generation

Stochastic search algorithms are general black-box optimizers. Due to their ease of use and their generality, they have recently also gained a lot of attention in operations research, machine learning and policy search. Yet, these algorithms require a lot of evaluations of the objective, scale poorly with the problem dimension, are affected by highly noisy objective functions and may converge prematurely. To alleviate these problems, we introduce a new surrogate-based stochastic search approach. We learn simple, quadratic surrogate models of the objective function. As the quality of such a quadratic approximation is limited, we do not greedily exploit the learned models. The algorithm can be misled by an inaccurate optimum introduced by the surrogate. Instead, we use information theoretic constraints to bound the 'distance' between the new and old data distribution while maximizing the objective function. Additionally the new method is able to sustain the exploration of the search distribution to avoid premature convergence. We compare our method with state of art black-box optimization methods on standard uni-modal and multi-modal optimization functions, on simulated planar robot tasks and a complex robot ball throwing task. The proposed method considerably outperforms the existing approaches.

Abstract-Dexterous manipulation enables repositioning of objects and tools within a robot's hand. When applying dexterous manipulation to unknown objects, exact object models are not available. Instead of relying on models, compliance and tactile feedback can be exploited to adapt to unknown objects. However, compliant hands and tactile sensors add complexity and are themselves difficult to model. Hence, we propose acquiring in-hand manipulation skills through reinforcement learning, which does not require analytic dynamics or kinematics models. In this paper, we show that this approach successfully acquires a tactile manipulation skill using a passively compliant hand. Additionally, we show that the learned tactile skill generalizes to novel objects.