This 35' professional scientific documentary which introduces the ECAgents project has been released in September 2006 (during the course of the project itself). The documentary gently introduce the users to the theme of the emergence of communication and language in embodied and situated agents, the interdisciplinary nature of this research endeavour, the open research questions, the results achieved up to this phase of the project, and the potential applications of the research conducted in the project. A high resolution DVD version of the documentary is available on requests. The documentary has been disseminated in scientific and industrial exhibition and has been broadcasted in its full duration or in part through TV channels including the Italian RAI Educational channel and the RAI TG2 News program.
Living organisms acquire, store, process and transmit information and, as such, information transfer plays a central role in the biology of most species. The conditions conducive to the evolution of communication and the paths by which reliable systems of communication become established remain largely unknown because communication does not leave a fossil record. This is a particularly challenging problem because efficient communication requires tight co-evolution between the signal emitted and the response elicited. Here we conducted repeated trials of experimental evolution with robots that could produce visual signals to provide information on food location. We found that communication readily evolves when colonies are composed of genetically similar individuals and/or when selection acts at the colony level. Under individual selection, the ability to produce visual signals resulted in the evolution of deceptive communication strategies in colonies of unrelated robots and a concomitant decrease in colony performance. This study generates predictions on the evolutionary conditions that may lead to the emergence of sophisticated forms of communication, including cooperative communication and deceptive signalling, and it provides guidelines for designing artificial evolutionary systems that may display spontaneous communication. For more information see: Floreano D., Mitri S., Magnenat S., Keller L. (2008).
Evolution and complexification of implicit and explicit communication
The research described in De Greef and Nolfi (2008) designed and investigate an experimental scenario which is simple enough to be analysed systematically but which at the same time includes all the elements which are necessary to investigate important questions concerning the evolution of communication such as: which are the conditions that might lead to the evolution of communication skills in a population of initially non-communicating individuals? What is the relation between agents' communicative and non-communicative behaviours and between different communication modalities (e.g. implicit and explicit communication)? How does the 'meaning' of signals originate and evolve and how is this grounded in agents' sensory experience?
In these experiments the robots are allowed to communicate explicitly (by sending and receiving signals through the wireless connections consisting of floating point values in the range [0.0, 1.0]) and implicitly (i.e. by visually perceiving the relative position of the other robots and my moving so to influence the perceptual environment of the other robot). The videos available from the following web-page show how robots evolved for an ability to reach the two target areas and to switch areas are fast as possible use implicit and explicit communication to share information about the environment and about their relative positions. For an analysis of the robots behavioural and communication skills and on how these skills originate evolutionarily see: De Greef and Nolfi (2008)
Authors: Joachim de Greef and Stefano Nolfi, CNR-ISTC, Italy
Emergence of coordinated/communicative behaviors
In these experiments we investigated whether coordinated/ communicative behaviour can emerge in group of robots evolved on the basis of a task-independent selection criteria. More specifically, the selection criteria considered consists in the mutual information between the motor actions performed by a team of robots. The rationale behind this study is the hypothesis that implicit and general purpose utility functions---fitness functions or reward/error measures---can allow evolution or learning to explore their search space more freely, without being constrained by an explicit description of the desired solution. In this way, it might be possible to discover behavioral and cognitive skills that play useful functionalities, and that might be hard to identify beforehand by the experimenter without an a priori knowledge of the system under study.
As can be seen by the videos available from the following web page the evolved robots display behaviours which are structured (i.e. which are formed by sequence of different atomic movements), periodic, and coordinated. This is achieved by exploiting communication. Robots are allowed to communicate through radio signals (i.e. continuous values in the range [0.0, 1.0]) which are transmitted and received through the wireless connection) and visual signals (i.e. light signals which are produced by turning the robots led ring on or off and by visually detecting colour blobs through the camera)
All human languages allow speakers to adopt perspectives other than their own in conceptualising the world, and to mark perspective explicitly through words or grammatical constructions. How can we explain this universal of human languages? We argue that speakers and hearers are able to configure and apply different strategies to deal with communicative tasks and retain their use if and only if this increases communicative success and expressive power while minimizing effort in terms of memory and information processing. Each strategy requires the recruitment of neural mechanisms, which are not specific to language. This video illustrates a concrete case study a model in which embodied robot agents are capable of creating and learning concepts, and language to express them. The agents play language games,
describing objects moving about in their immediate surroundings to each other. Because the agents are embodied and situated, the perspective from which they perceive and describe a scene influences their individual perception of it. We show that in this case, giving agents the capacity to project each other's visual perspectives increases their communicative success, as does allowing them to encode perspective in language. Because human beings are embodied agents as well, situated in the physical world, their communicative success will analogically increase when they use perspective-marking language. The model thus correctly predicts that perspective marking will be culturally selected as a universal strategy in the evolution of human language.
What kind of motivation drives child language development? In this research we presents a computational model and a robotic experiment which support the hypothesis that children discover communication as a result of exploring and playing with their environment. The considered robotic agent is intrinsically motivated towards situations in which it optimally progresses in learning. To experience optimal learning progress, it must avoid situations already familiar but also situations where nothing can be learned. The robot is placed in an environment in which both communicating and non-communicating objects are present.A s a consequence of its intrinsic motivation, the robot explores this environment in an organized manner focussing first on non-communicative activities and then discovering the learning potential of certain types of interactive behavior. In this experiment, the agent ends up being interested by communication through vocal interactions without having a specific drive for communication. For more information see: Oudeyer, Kaplan (2006).
The future of robotics will be about how it emerges in our everyday life and provides a sustainable interest. We want to rethink the role of robots from servants doing our bidding, or animal imitations that comes far from our expectations - to more realistic visions where robots will be able to amaze us. GlowBots are small round robots showing visually appealing shapes on a colour-intense hypnotizing display. Interacting with the robot encourages it to seek out and communicate its shape to other robots. Through their rich interaction a flora of shapes emerges that entices interest and engagement in an open ended fashion for joy and entertainment. GlowBots have been demonstrated at several venues and exhibitions including Emerging Technologies at SIGGRAPH 2007 and Wired NextFest. They have also appeared in different scientific news media e.g. Discovery News and New Scientist Technology blog.
Authors: Mattias Jacobsson, Lars Holmquist, VICKTORIA Institute, Sweeden
ECAgents exhibition, 2004, "The Hague" Information Society Technologies event
This video review the ECAgents exhibition organized at the Information Society Technologies event organized in November 2004 at "The Hague" under the EU's Dutch Presidency by the European Commission in partnership with the Dutch Ministry of Economic Affairs. The exhibition included contributions of SONY-CSL, EPFL, and VIKTORIA.