A concurrent approach to robot team learning

摘要

Despite the advancement of research and development on multi-robot teams, a key challenge still remains as to how to develop effective mechanisms that enable the robots to autonomously generate, adapt, and enhance team behaviours while improving their individual performance simultaneously. To address this issue, a cooperative learning algorithm is modified in this paper to accommodate for the individualistic Q-Learning as well as the collaborative advice sharing. The developed methods are examined in relation to the performance characteristics of single-robot learning to ascertain if they retain viable learning characteristics despite the integration of individual learning into team behaviour. Further, a modification to the individual learning method was implemented into the proposed multi-robot learning approach to examine the performance improvements gained by the multi-robot learning algorithms.