Comprehension of Operating a Robot by Enactive Learning: Exemplary Approaches with Programming-by-Demonstration

Abstract

Robot Programming-by-Demonstration demands two skills from the user, understanding the task that is to be programmed, and comprehending how to use the robotic system. This article is about training the user those skills that are necessary for the latter requirement. The robotic system is based on the DLR/KUKA light-weight robot, which allows direct human-machine interaction and compliant motions. An augmented reality and a virtual reality setup are presented that aim to demonstrate and evaluate skills transfer of two different sub-tasks of this system: avoiding robot singularities and setting correct compliance parameters. For this purpose training accelerators are introduced for (1) visualising robot singularities, (2) exploring robot singularities, and (3) feeling compliance parameters. An evaluation procedure for these accelerators is suggested.