Gaze following is the ability to re-direct one’s gaze to the location where another agent is looking. We present a computational model of how hu-man infants or other agents may acquire gaze following by learning to pre-dict the locations of interesting sights from the looking behavior of other agents through reinforcement learning. The model accounts for many find-ings about the development of gaze following in human infants. During learning, the model develops pre-motor representations that exhibit many properties characteristic of mirror neurons, but they are specific to looking behaviors. The existence of such a new class of mirror neurons is the main prediction of our model. The model also offers a parsimonious account of how these and possibly other mirror neurons may acquire their special re-sponse properties. In this account, visual representations of other agents’ actions become associated with pre-motor neurons that represent the inten-tion to perform corresponding actions. The model also demonstrates how this development may be obstructed in autism spectrum disorder, giving rise to specific physiological and anatomical differences in the mirror system.

This position paper proposes that the study of embodied cognitive agents, such as humanoid robots, can advance our understanding of the cognitive development of complex sensorimotor, linguistic and social learning skills. This in turn will benefit the design of cognitive robots capable of learning to handle and manipulate objects and tools autonomously, to cooperate and communicate with other robots and humans, and to adapt their abilities to changing internal, environmental, and social conditions. Four key areas of research challenges are discussed, specifically for the issues related to the understanding of: (i) how agents learn and represent compositional actions; (ii) how agents learn and represent compositional lexicons; (iii) the dynamics of social interaction and learning; and (iv) how compositional action and language representations are integrated to bootstrap the cognitive system. The review of specific issues and progress in these areas is then translated into a practical roadmap based on a series of milestones. These milestones provide a possible set of cognitive robotics goals and test-scenarios, thus acting as a research roadmap for future work on cognitive developmental robotics.

In this paper, we analyze the different approaches taken to-date within the computer vision, robotics and artificial intelligence communities for the representation, recognition, synthesis and understanding of action. We deal with action at different levels of complexity and provide the reader with the necessary related literature references. We put the literature reference further into context and outline a possible interpretation of action by taking into account the different aspects of action recognition, action synthesis and task-level planning.

Start date of project: 01/09/2005 Duration: 36 months

Robots and humans receive partial, fragmentary hints about the world’s state through their respective sensors. These hints – tiny patches of light intensity, frequency components of sound, etc. – are far removed from the world of objects we feel we perceive so effortlessly around us. The study of infant development and the construction of robots are both deeply concerned with how this apparent gap between the world and our experience of it is bridged. In this paper, we focus on some fundamental problems in perception that have attracted the attention of researchers in both robotics and infant development. Our goal is to identify points of contact already existing between the two fields, and also important questions identified in one field that could fruitfully be addressed in the other. We start with the problem of object segregation: how do infants and robots determine visually where one object ends and another begins? For object segregation, both fields have examined the idea of using “key events ” where perception is in some way simplified and the infant or robot acquires knowledge that can be exploited at other times. We propose that the identification of the key events themselves constitutes a point of contact between the fields. And although the specific algorithms used in robots do not necessarily map directly to infant strategies, the overall “algorithmic skeleton ” formed by the set of algorithms needed to identify and exploit key events may in fact form a basis for mutual dialogue. We then look more broadly at the role of embodiment in humans and robots, and see the opportunities it affords for development.

a theoretical framework for learning

Abstract—In the mirror neuron system, the canonical neurons play a role in object shape and observer-object relation recognition. However, there are almost no functional models of canonical neurons towards the integration of these two functions. We attempt to represent the relative position between the object and the robot in a neural network model. Although at present some generative models based on the Restricted Boltzmann Machine can code the image transformation in continuous images, what we need to accomplish in canonical neuron modeling is different from the requirements of modeling transformation in video frames. As a result, we propose a novel model called “Restricted Boltzmann Machine with Transformation Units”, which can represent the relative object positions based on laser images. The laser sensor provides binary and accurate images and can further be connected with other models to construct a unified architecture of the mirror neuron system. I.