No abstract

We aim at a robot capable to learn sequences of actions to achieve a field of complex tasks. In this paper, we are considering the learning of a set of interrelated complex tasks hierarchically organized. To learn this high-dimensional mapping between a continuous high-dimensional space of tasks and an infinite dimensional space of unbounded sequences of actions, we introduce a new framework called “procedures”, which enables the autonomous discovery of how to combine previously learned skills in order to learn increasingly complex combinations of motor policies. We propose an active learning algorithmic architecture, capable of organizing its learning process in order to achieve a field of complex tasks by learning sequences of primitive motor policies. Based on heuristics of active imitation learning, goal-babbling and strategic learning using intrinsic motivation, our algorithmic architecture leverages our procedures framework to actively decide during its learning process which outcome to focus on and which exploration strategy to apply. We show on a simulated environment that our new architecture is capable of tackling the learning of complex motor policies by adapting the complexity of its policies to the task at hand. We also show that our “procedures” enable the learning agent to discover the task hierarchy and exploit his experience of previously learned skills to learn new complex tasks.

No abstract

In this paper, we present a model of emotions that we proposed in EmotiRob project. We make a comparison of recent models of emotions and show that our model is generic in basing on the theories of emotions of Ortony et al., Lazarus, Scherer and then the personality theory of Meyers-Brigg and Meyers.

No abstract

To cite this version:Sébastien Saint-Aimé, Marine Grandgeorge, Brigitte Le Pevedic, Dominique Duhaut. Evaluation of Emi interaction with non-disabled children in nursery school using wizard of Oz technique. IEEE Robio-2011, IEEE International Conference on Robotics and Biomimetics, Dec 2011, Thailand. 2011. Evaluation of Emi interaction with non-disabled children in nursery school using wizard of Oz techniqueSébastien Saint-Aimé and Marine Grandgeorge and Brigitte Le-Pévédic and Dominique DuhautAbstract-Research in the field of emotional interaction is discussed here, for the EmotiRob project, to maintain interaction with children in the 4-to-8 year old age range. The objective of this project is to give comfort to vulnerable children and/or those undergoing long-term hospitalisation through the help of an emotional robot companion. The studies carried out on perception and emotional synthesis have allowed us to develop an experimental stuffed robot Emi, using an emotional model, iGrace, allowing for emotional reaction based on the speech of the user. This paper briefly presents the EmotiRob project and how emotion has been used for Emi. The last experiment done with children to evaluate their interaction with Emi is then described.

In this paper we present StimCards: an interactive game for cognitive training exercises. To increase the impact of this game we experiment four kinds of interfaces: a basic computer, an embodied conversational agent and a robot with two different appearances. The report of these experiments shows that the robot is the best positive feedback for cognitive game.

Abstract-This article presents the evaluation of the emotional expressiveness of EmI companion robot in the EmotiRob project. We describe iGrace emotional computational model of emotion to generate an emotional response based on the speech of the interlocutor, the mechanical design and implementation of EmI, and experimentation to evaluate the expressiveness of EmI with 52 school children aged 7 to 9 years.