Multiagent robotics: toward energy autonomy

In this paper, we propose a novel trend in multiagent robotics: energy autonomy. A definition of energy autonomy is developed from an original concept, “potential energy,” that is under the constraints of remaining energy capacity and the relative distance among robotic agents. Toward energy autonomy, we initially present a simulation of a multiagent robotic system in which each robot is capable of exchanging energy cells with other robots. Our simulation points out that: (1) each robot is able not only to act as an autonomous agent, but also to interact with others to be beyond the individual capabilities; (2) in order to adapt to changes in the environment, each robot is situated as an adaptive agent in a network of neighboring robots, which leads to a state of energy autonomy. Finally, based on the results of the simulation, we adjust the rules for our real multirobot system. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007