Achieving Corresponding Effects on Multiple Robotic Platforms: Imitating in Context Using Different Effect Metrics

One of the fundamental problems in imitation is the correspondence problem, how to map between the actions, states and effects of the model and imitator agents, when the embodiment of the agents is dissimilar. In our approach, the matching is according to different metrics and granularity. This paper presents JABBERWOCKY, a system that uses captured data from a human demonstrator to generate appropriate action commands, addressing the correspondence problem in imitation. Towards a characterization of the space of effect metrics, we are exploring absolute/relative angle and displacement aspects and focus on the overall arrangement and trajectory of manipulated objects. Using as an example a captured demonstration from a human, the system produces a correspondence solution given a selection of effect metrics and starting from dissimilar initial object positions, producing action commands that are then executed by two imitator target platforms (in simulation) to successfully imitate.