Nowadays, Multi-Robot Systems (MRS) control represents a great challenge in the research community: the interest in MRS is justified by the numerous advantages that a group of autonomous robots offers compared to a single robot, in particular when complex missions are assigned. However, these systems need a complex control architecture and a method of coordination of the autonomous robots. In this paper, a Multi-Agent System (MAS) based approach is proposed to solve this problem: the purpose is to realize a decentralized system, where each robot, using a Reinforcement Learning (RL) mechanism, autonomously acquires the appropriate behaviour and suitable rules through its experience of interaction with the environment and the other robots.
Agent Control Technology for Multi-Robot Systems
CAGGIANO A;
2007-01-01
Abstract
Nowadays, Multi-Robot Systems (MRS) control represents a great challenge in the research community: the interest in MRS is justified by the numerous advantages that a group of autonomous robots offers compared to a single robot, in particular when complex missions are assigned. However, these systems need a complex control architecture and a method of coordination of the autonomous robots. In this paper, a Multi-Agent System (MAS) based approach is proposed to solve this problem: the purpose is to realize a decentralized system, where each robot, using a Reinforcement Learning (RL) mechanism, autonomously acquires the appropriate behaviour and suitable rules through its experience of interaction with the environment and the other robots.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
