A Mobile Peer-to-Peer Network of CBR Agents to Provide e-Assistance

In this paper we combine three recent technologies to provide e-Assistance. On the one hand Ambient Intelligence is an useful paradigm to help people dealing with unknown situations. On the other hand, Case Based Reasoning (CBR) applies knowledge gained through past experiences to solve current problems. Finally, peer-to-peer (P2P) networks let users to share resources among members of a community. A combination of these technologies can be used to build a seamless e-assistance system in order to help people using mobile ad-hoc networks and enabling nodes to establish spontaneous connections among them in order to meet specific needs. Taking this context as a starting point we have developed a CBR multi-agent system (MAS) that exchanges its cases through a P2P mobile ad-hoc network in order to solve daily problems. An example describing an intelligent gym for elder physical training is also provided.