Object and relation centric representations for push effect prediction

Ahmet Ercan Tekden, Aykut Erdem, Erkut Erdem, Tamim Asfour, Emre Ugur

Abstract: Highlights•Graph neural network based framework for parameter estimation and physics prediction.•Articulation based graph representation for modeling multi-part objects.•6D Effect Prediction.•Tool Manipulation Planning.•Model predictive control with a learned model.