Embodiment is a key element in artificial intelligence and robotics, as illustrated by several authors (Brooks, Pfeifer). We will here demonstrate some examples of this key aspect by showing some mechatronic design elements that have been developed during the  swarm-bots projet, a European Project sponsored by the Information Society Technologies program. We will demonstrate how collective performances are strongly related to the robot design and technology. Scalability, efficience and global performances will be linked to size, mobility and mechatronic aspects. Finally we will give some comments on lesson learned during this projet. This introduction will be linked to some aspects of other presentations.

Collective behaviors in social insects can be very impressive. They range from the coordinated displacement of thousands of individuals [5, 6] to the building of complex structures [8, 13] or to the proper allocation of tasks between the members of a group [11, 4,2]. During the last forty years, a growing body of studies was interested in understanding the mechanisms underlying these biological systems. We now know that most of these collective behaviors can be seen as decentralized systems made of autonomous units that are distributed in the environment and that follow simple probabilistic stimulus-response behaviors [3]. This peculiar mode of organization, often based on self-organized processes, combines e_ciency with flexibility, robustness and distributedness [1].

Aiming at controlling the behaviors of groups of robots, collective robotics was often inspired by the collective abilities demonstrated by social animals, and particularly by social insects [15]. Indeed, social insects represent promising models for the decentralized organization and coordination of many autonomous robots [1]. For fifteen years, several studies have used bio-inspired robot controllers to deal with collective behaviors as manifold as aggregation [14], foraging [16], task allocation [12], stick pulling [10], object sorting[9] or place selection [7].

Therefore this lecture aims at reviewing the development of this recent bug-inspired collective robotics. Through di_erent examples of insects-like collective behaviors embodied in groups of robots, it will emphasize the challenges of coordinating the work of many autonomous agents and will show the interest of such a research field for both biologists and roboticists.

References

[1] Bonabeau, E., Dorigo, M., & Theraulaz, G. (1999). Swarm intelligence : from natural to artificial systems. Oxford: Oxford University Press.

[2] Bonabeau, E. & Theraulaz, G. (1999). Role and variability of response thresholds in the regulation of division of labor in insect societies. In C. Detrain, J.-L. Deneubourg, & J. M. Pasteels (Eds.), Information Processing in Social Insects. Basel: Birkhäuser Verlag, pp. 141–163.

[3] Camazine, S., Deneubourg, J.-L., Franks, N. R., Sneyd, J., Theraulaz, G., & Bonabeau, E. (2001). Self-organization in biological systems. Princeton: Princeton University Press.

[4] Deneubourg, J.-L., Goss, S., Pasteels, J. M., Fresneau, D., & Lachaud, J.-P. (1987). Selforganization mechanisms in ant societies (ii): learning in foraging and division of labor. In J. M. Pasteels & J.-L. Deneubourg (Eds.), From individual to collective behavior in social insects, vol. 54. Basel: Birkhäuser, pp. 177–196.

[5] Dussutour, A., Fourcassié, V., Helbing, D., & Deneubourg, J.-L. (2004). Optimal tra_c organization in ants under crowded conditions. Nature, 428(6978), 70–3.

[6] Franks, N. R. & Fletcher, C. R. (1983). Spatial patterns in army ant foraging and migration: Eciton burchelli on barro colorado island, panama. Behavioral Ecology and Sociobiology, V12(4), 261–270.

[7] Garnier, S., Jost, C., Jeanson, R., Gautrais, J., Asadpour, M., Caprari, G., & Theraulaz, G. (2005). Aggregation behaviour as a source of collective decision in a group of cockroach-like robots. In M. Capcarrere (Ed.), Proceedings of the 8th European Conference on Artificial Life, vol. 3630 of Lecture Notes in Artificial Intelligence. Berlin Heidelberg: Springer-Verlag, pp. 169–178.

[8] Grassé, P.-P. (1984). Termitologia, Tome II. Fondation des Sociétés. Construction. Paris: Masson.

[9] Holland, O. E. & Melhuish, C. (1999). Stigmergy, self-organisation, and sorting in collective robotics. Artificial Life, 5, 173–202.

[10] Ijspeert, A., Martinoli, A., Billard, A., & Gambardella, L. (2001). Collaboration through the exploitation of local interactions in autonomous collective robotics: The stick pulling experiment. Autonomous Robots, V11(2), 149–171.

[11] Jeanne, R. L. (1996). Regulation of nest construction behaviour in polybia occidentalis. Animal Behaviour, 52(3), 473–488.

[12] Krieger, M. J. B., Billeter, J.-B., & Keller, L. (2000). Ant-like task allocation and recruitment in cooperative robots. Nature, 406(6799), 992–995.

[13] Lüscher, M. (1961). Air-conditioned termite nests. Sientific American, 205, 138–145.

[14] Martinoli, A., Jispeert, A. J., & Mondada, F. (1999). Understanding collective aggregation mechanisms: from probabilistic modelling to experiments with real robots. Robotics and Autonomous Systems, 29, 51–63.

[15] Sahin, E. (2005). Swarm robotics: From sources of inspiration to domains of application. vol. 3342 of Lecture Notes in Computer Science. pp. 10–20.

[16] Sugawara, K. & Sano, M. (1997). Cooperative acceleration of task performance: foraging behavior of interacting multi-robots system. Physica D: Nonlinear Phenomena, 100(3/4), 343–354.

It is a characteristic of swarm robotics that modelling overall emergent swarm behaviours in terms of the low level behaviours of individual robots is very difficult. Yet if swarm robotics is to make the transition from the laboratory to real-world engineering realisation we need such models.  Using a wireless connected swarm as a case study, this talk will outline two possible approaches. The first makes use of temporal logic formalism to specify and perhaps ultimately prove the emergent swarm properties. The second extends a probabilistic modelling approach that geometrically estimates state transition probabilities to build a model of overall swarm connectivity.

This talk will address the problem of multi-robot coordination for the execution of complex tasks in dynamic, inaccessible, non-deterministic environments. The talk will be focused on the application of coordination methodologies for creating soccer playing robotic teams. Thus, the RoboCup Initiative and its main objectives will be described. RoboCup is a joint international project created to promote research in (Distributed) Artificial Intelligence and Intelligent Robotics. Soccer was selected as the central research topic because, besides being a very complex collective game, it is attractive to researchers, to the media and to the general public. Also a huge amount of technologies have to be involved to create soccer playing robots, including, among others: autonomous agents, multi-agent systems, cooperation and communication, robotics, sensor fusion, control systems, real-time reasoning, machine learning. There are nine major senior leagues in RoboCup, including six soccer leagues: Simulation 2D, Simulation 3D, Small-Size, Middle-Size, 4 Legged and Humanoid Robots. Two search and rescue leagues: Simulated Rescue and Robotic Rescue and a RoboCup@Home league complete the major competitions in the RoboCup initiative. Different leagues pose distinct but interrelated problems leading to develop a humanoid soccer playing team with individual and collective abilities similar to a human soccer playing team. Emphasis will be given to advanced coordination methodologies for creating high-level flexible strategies, for soccer playing teams of robots. The coordination methodologies developed by our research group, applied in RoboCup since 1998, enabled us (teams FC Portugal and 5DPO) to win three robotic soccer World championships and six European championships, besides more than twenty other international awards, in distinct leagues (simulation2D, simulation 3D, coach competition, small-size, middle size and rescue). These methodologies include strategies defined using flexible tactics, dynamic formations, player types, flexible setplays, communication and perception methodologies and the concept of active/passive situation. The concept of Strategic Coordination will also be described and the associated coordination algorithms: Situation Based Strategic Positioning, Dynamic Positioning and Role Exchange, Intelligent Perception, Advanced Communication and Mutual Modeling. Examples and results of the implementation of these methodologies, using RoboCup simulators and our robotic platforms, will be shown and deeply discussed.