This paper addresses the role of imitation as a means to enhance the learning of communication skills in autonomous robots. A series of robotic experiments are presented in which autonomous mobile robots are taught a synthetic proto-language. Learning of the language occurs through an imitative scenario where the robot replicates the teacher's movements. Imitation is here an implicit attentional mechanism which allows the robot imitator to share a similar set of proprio- and exteroceptions with the teacher. The robot grounds its understanding of the teacher's words, which describe the teacher's current observations, upon its own perceptions which are similar to those of the teacher. Learning of the robot is based on a dynamical recurrent associative memory architecture (DRAMA). Learning is unsupervised and results from the self-organization of the robot's connectionist architecture. Results show that the imitative behavior greatly improves the efficiency and speed of the learning. More...

This paper proposes a class of software agent architectures capable of playing some of beneficial roles embodiment accomplishes in living beings. These are component based architectures including sensors, actuators, emotion eliciting mechanisms and action control mechanisms. These architectures through their yellow pages and log services, which maintain updated information about the agent body and its dynamics, provide the means for the soft-visual appearance required for non-verbal communication and for learning through imitation. Component based architectures fulfil the requirement of efficient execution of agent functions in parallel and the possibility of using simplifying sensory-motor coordination. All components of architectures of the proposed class share the same goal of preserving the plans for the adaptive and situated construction of the agent. This feature plays the role of the body as an inner source of motives, enables the agent to have feelings and emotions, and allows the agent to centrifugally construct itself, avoiding non-grounding meta-control processes. Finally the proposed class of architectures integrate several learning mechanisms necessary for learning through imitation, for learning through direct manipulation of objects, for learning correlations among and computational complexity of the several aspects of the task and the environment, which allows the agent to replace slow with fast computations. 1.

In this paper, we propose a robotics framework for studying the coevolution of signaling. Our motivation is twofold. First, we propose a situated and embodied framework for signaler-receiver interaction, and second, we provide a promising approach for thestudyofmechanisms that would enable adaptive systems to access new information channels and to exploit implicit information in their environments. We present experimental results on a successful coevolution of signals that enable a very simple communication between two robots. Finally, we delineate some aspects of forthcoming research.

be combined to form complex behaviors (Matari'c, M. 95), how such systems can achieve tasks that require global knowledge (Werger and Matari'c 96), how robot teams and tasks can be organized for efficient operation (Fontan and Matari'c 96),(Goldberg and Matari'c 97), and how robots can learn behavior selection (Matari'c 97) and learn through observation of their history of behavior activation (Michaud and Matari'c 97). Figure 1: The Interaction Lab Robots. In front are the Pioneers, surrounded by the four R2Es and twenty R1s. Our Robots The Interaction Lab has twenty-six robots, including RWI Pioneers and ISR R1s and R2Es. We are sure that the Pioneers will participate in all three events, and are investigating feasible means of incorporating some of the other robots. We'd like to field the largest teams we can. The three Pioneers - Ben, Mae, and Ullanta 1 - are manufactured by Real World Interface, Inc., and are differentially steered bases with seven son