FEELIX GROWING

Hiolle, A. & Cañamero, L. (2008). Why should you care? An arousal-based model of exploratory behavior for autonomous robots. In S. Bullock, J. Noble, R. A. Watson, and M. A. Bedau (Eds.) Proceedings of the Eleventh International Conference on Artificial Life, pp. 242-248, MIT Press, Cambridge, MA.
Abstract:
The question of how autonomous robots could be part of our everyday life is of a growing interest. We present here an experiment in which an autonomous robot explores its environment and tries to familiarize itself with the features available using a neural-network-based architecture. The lack of stability of its learning structures increases the arousal level of the robot, pushing the robot to look for comfort from its caretaker to reduce the arousal. In this paper, we studied how the behavior of the caretaker influences the course of the robot exploration and learning experience by providing certain amount of comfort during this exploration. We then draw some conclusions on how to use this architecture together with related work, to enhance the adaptability of autonomous robots development.

Hiolle, A. & Cañamero, L. (2008). Conscentious caretaking for autonomous robots: an arousal-based model of exploratory behavior . In Schlesinger, M., Berthouze, L. and Balkenius, C. (Eds.) Proceedings of the Eighth International Conference on Epigenetic Robotics: Modeling Cognitive Development in Robotic Systems, pages 45–52. Lund University Cognitive Studies, 139.
Abstract:
The question of how autonomous robots could be part of our everyday life is gaining increasing interest. We present here an experiment in which an autonomous robot explores its environment and tries to familiarize itself with its novel elements using a neural-network-based architecture. When confronted with novelty, the lack of stability of its learning structures increases the arousal level of the robot, pushing it to look
for comfort from its caretaker in order to reduce this arousal. In this paper, we studied how the behavior of the caretaker—and in particular the amount of comfort it provides to the robot during its exploration of the environment—influences the course of the robot’s exploration and learning experience. This work takes inspiration from early mother-infant interactions and the impact that the primary caretaker has on the development of children—at least in mainstream Western culture. The underlying hypothesis
is that the behavior of a caregiver, and particularly his/her role in modulating arousal, will influence the development of an autonomous robot, and that arousal regulation will also depend on how accurately the robot signals its internal state and how the caretaker (or human user) responds to these signals.

Hiolle, A. and Cañamero, L. (2007). Developing sensorimotor associations through attachment bonds. In Prince, C., Balkenius, C., Berthouze, L., Kozima, H., and Littman, M. (Eds.) Proceedings of the Seventh International Workshop on Epigenetic Robotics: Modeling Cognitive Development in Robotic Systems, pages 45–52. Lund University Cognitive Studies, 134.
Abstract:
Attachment bonds and positive affect help cognitive development and social interactions in infants and animals. In this paper we present a neural architecture to enable a robot to develop an attachment bond with a person or an object, and to discover the correct sensorimotor associations to maintain a desired affective state of well-being using a minimum amount of prior knowledge about the possible interactions with this object. We also discuss how our research on attachment bonds could further developmental robotics in the near future.

Calinon, S. and Billard, A. (2008), "A Probabilistic Programming by Demonstration Framework Handling Constraints in Joint Space and Task Space" in Proceedings of IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS), Nice, France, September 2008.

Abstract
We present a probabilistic architecture for solving generically
the problem of extracting the task constraints through a
Programming by Demonstration (PbD) framework and for generalizing
the acquired knowledge to various situations. We propose an
approach based on Gaussian Mixture Regression (GMR) to find
automatically a controller for the robot reproducing the essential
characteristics of the skill by handling simultaneously
constraints in joint space and in task space. Experiments with two
5-DOFs Katana robots are then presented with two manipulation
tasks consisting of handling and displacing a set of objects.

Contact details of the coordinator:
Dr. Lola Cañamero
School of Computer Science
University of Hertfordshire
College Lane
Hatfield, Herts, AL10 9AB, UK

L DOT CANAMERO AT HERTS DOT AC DOT UK

web page

For queries about the project, please contact:

FGMANAGER AT GOOGLEMAIL DOT COM