A Multiagent System Based on Heterogeneous Robots

The paper presents the introduction of heterogeneity into a robotic Multiagent System. The system is based on ideas from Artificial Life; it forms a kind of artificial ecosystem where the animats are linked together in their search for “food” in form of electrical energy. Within this view, different robot-types can be seen as different species. Given a basic robotic ecosystem with homogeneous agents — the so called “moles” —, the two new species “mouse” and “head” are introduced. The differences between species are quite substantial. The “moles” for example have simple sensing capabilities whereas the “mouse” is equipped with vision. Some agents do not have social capabilities, whereas “heads” depend on cooperation. The paper describes how these differences and the common dependence on a global energy source interfere, and which conceptual and technological choices have to be made to keep a kind of ecological balance.

[1]  W. Ashby.  Design for a Brain , 1954 .

[2]  David McFarland,et al.  Quantitative ethology : the state space approach , 1981 .

[3]  Stewart W. Wilson.  The animat path to AI , 1991 .

[4]  Pattie Maes,et al.  A bottom-up mechanism for behavior selection in an artificial creature , 1991 .

[5]  Jean-Arcady Meyer,et al.  Simulation of adaptive behavior in animats: review and prospect , 1991 .

[6]  Rodney A. Brooks,et al.  Intelligence Without Reason , 1991, IJCAI.

[7]  Long-Ji Lin,et al.  Self-improving reactive agents: case studies of reinforcement learning frameworks , 1991 .

[8]  John E. W. Mayhew,et al.  Computer simulation of an animal environment , 1991 .

[9]  F. Varela,et al.  Toward a Practice of Autonomous Systems: Proceedings of the First European Conference on Artificial Life , 1992 .

[10]  Luc Steels,et al.  The artificial life roots of artificial intelligence , 1993 .

[11]  David R. Jefferson,et al.  Artificial Life as a Tool for Biological Inquiry, in Artificial Life: an Overview , 1993 .

[12]  Maja J. Matarić,et al.  Designing emergent behaviors: from local interactions to collective intelligence , 1993 .

[13]  Ian Horswill,et al.  Polly: A Vision-Based Artificial Agent , 1993, AAAI.

[14]  Luc De Raedt,et al.  Learning to survive , 1993 .

[15]  Hugues Bersini.  Reinforcement learning for homeostatic endogenous variables , 1994 .

[16]  Luc Steels,et al.  A case study in the behavior-oriented design of autonomous agents , 1994 .

[17]  Bruce Blumberg,et al.  Action-selection in hamsterdam: lessons from ethology , 1994 .

[18]  Jukka Riekki,et al.  Vision-based behaviors for multi-robot cooperation , 1994, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'94).

[19]  Gregory M. Werner,et al.  Using second order neural connections for motivation of behavioral choices , 1994 .

[20]  Lynne E. Parker,et al.  On the design of behavior-based multi-robot teams , 1995, Adv. Robotics.

[21]  Pattie Maes,et al.  Synthetic Robotic Language Acquisition by Observation , 1996 .

[22]  Pattie Maes,et al.  Robotic “Food” Chains: Externalization of State and Program for Minimal-Agent Foraging , 1996 .

[23]  Luc Steels,et al.  Discovering the Competitors , 1996, Adapt. Behav..

[24]  Luc Steels,et al.  Grounding adaptive language games in robotic agents , 1997 .

[25]  Inman Harvey,et al.  Fourth European Conference on Artificial Life , 1997 .

[26]  Luc Steels,et al.  A selectionist mechanism for autonomous behavior acquisition , 1997, Robotics Auton. Syst..