Elements of information theory
Elements of information theory
Map learning with uninterpreted sensors and effectors
Map learning with uninterpreted sensors and effectors
Map learning with uninterpreted sensors and effectors
Artificial Intelligence
Fast, On-Line Learning of Globally Consistent Maps
Autonomous Robots
Diffusion Kernels on Statistical Manifolds
The Journal of Machine Learning Research
A Mathematical Theory of Communication
A Mathematical Theory of Communication
Learning to interpret pointing gestures: experiments with four-legged autonomous robots
Biomimetic Neural Learning for Intelligent Robots
| Hi-index | 0.00 |
In a study of how the concept of affordances could be applied to interaction behaviour, we introduce the notion of "interpersonal maps", a geometrical representation of the relationships between a set of proprioceptive and heteroceptive information sources, thus creating a common representation space for comparing one's own behaviour and the behaviour of others. Such maps can be used to detect specific types of interactions between agents such as imitation. Moreover, in cases of strong couplings between agents, such representations permit to map directly an agent's body structure onto the structure of an observed body, thus addressing the body correspondence problem. These various cases are studied with several robotic experiments using four-legged robots either acting independently or being engaged in delayed imitation. Through a precise study of the effects of the imitation delay on the structure of the interpersonal maps, we show the potential of this "we-centric" space to account for both imitative and non imitative interactions.