Title: Towards Learning by Demonstration in Multi-robot Systems 

Speaker: Murilo Fernandes Martins 

Abstract: 
A robots become more accessible to humans, more intuitive and human-friendly
ways of programming them with interactive and group-aware behaviours are
needed. This talk will address the gap between Learning by Demonstration
and Multi-robot systems. In particular, the work presented in this talk
tackles the fundamental problem of learning multi-robot cooperative behaviour 
from concurrent multi-teacher demonstrations, problem which had not been fully
addressed in the literature as yet.

This talk will briefly introduce the design and implementation of a novel,
multi-layered framework for multi-robot learning from simultaneous demonstra-
tions, capable of deriving control policies at two different levels of
abstraction. The lower level learns models of joint-actions at trajectory
level, adapting such
models to new scenarios via feature mapping. The higher level extracts the
structure of cooperative tasks at symbolic level, generating a sequence of
robot actions composing multi-robot plans. To the best of the author's 
knowledge, the presented framework was the rst Learning by Demonstration 
system to enable multiple human demonstrators to simultaneously teach group 
behaviour to multiple robots learners.

A series of experimental tests were conducted using real robots in a real hu-
man workspace environment. The results obtained from a comprehensive com-
parison conrm the applicability of the joint-action model adaptation method
utilised. What is more, the results of several trials provide evidence that
the framework effectively extracts reasonable multi-robot plans from demonstra-
tions.