Reasoning About Liquids via Closed-Loop Simulation

Authors: Connor Schenck, Dieter Fox

Simulators are a powerful tool for reasoning abouta robot’s interactions with its environment. However, whensimulations diverge from reality, that reasoning becomes lessuseful. In this paper, we evaluate closing the loop betweensimulation and reality by using observations of real liquids tocorrect errors when tracking the liquid’s state in a simulator.Our results show that closed-loop simulation is an effective wayto prevent large divergence between the simulated and real liquidstates, and additionally that this method can enable reasoningabout liquids that would otherwise be infeasible.