Decentralized learning control

Abstract

The theoretical results reported are motivated by the problem of maneuvering common objects by multiple robots. The authors show that learning control is possible in a decentralized configuration. Sufficient conditions are given for decentralized learning control convergence. The results are illustrated with a well-known example of interconnected pendulums given by Siljak.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>