Speaker: Pierre-Brice Wieber, Chargé de Recherche at Bipop team, INRIA Rhône-Alpes
Abstract: When a robot has more than one way to realize a given objective, the general approach is to leverage this redundancy and introduce additional objectives, that the robot could try to satisfy at the same time. We are going to see some of the geometric and numerical problems that this approach raises, and face the never-ending problem when dealing with machines: they do very precisely what they are told to, only for you to realize that what you told them to do is in fact absolutely not what you wanted.
This talk will be preceded by a brief panorama of robotics research in the Grenoble area.