Adaptable Pouring: Teaching Robots Not to Spill using Fast but Approximate Fluid Simulation

Tatiana López Guevara, Nicholas Kenelm Taylor, Michael Gutmann, Subramanian Ramamoorthy, Kartic Subr

Research output: Chapter in Book/Report/Conference proceedingConference contribution

390 Downloads (Pure)


Humans manipulate fluids intuitively using intuitive approximations of the underlying physical model. In this paper, we explore a general methodology
that robots may use to develop and improve strategies for overcoming manipulation tasks associated with appropriately defined loss functions. We focus on the specific task of pouring a liquid from a container (pourer) to another container (receiver) while minimizing the mass of liquid that spills outside the receiver. We present a solution, based on guidance from approximate simulation, that is fast, flexible and adaptable to novel containers as long as their shapes can be sensed. Our key idea is to decouple the optimization of the parameter space of the simulator from the optimization over action space for determining robot control actions. We perform the former in a training (calibration) stage and the latter during run-time (deployment). For the purpose of this paper we use pouring in both stages, even though separate actions could be chosen. We compare four different strategies for
calibration and three different strategies for deployment. Our results demonstrate that fast fluid simulations are effective, even if they are only approximate, in guiding automatic strategies for pouring liquids.
Original languageEnglish
Title of host publicationProceedings of Machine Learning Research: Conference on Robot Learning, 13-15 November 2017
EditorsSergey Levine, Vincent Vanhoucke, Ken Goldberg
Number of pages10
Publication statusPublished - 13 Nov 2017
Event1st Conference on Robot Learning 2017 - Mountain View, United States
Duration: 13 Nov 201715 Nov 2017

Publication series

NameProceedings of Machine Learning Research
ISSN (Electronic)2640-3498


Conference1st Conference on Robot Learning 2017
Abbreviated titleCoRL 2017
Country/TerritoryUnited States
CityMountain View
Internet address


  • Fluid simulation
  • Pouring


Dive into the research topics of 'Adaptable Pouring: Teaching Robots Not to Spill using Fast but Approximate Fluid Simulation'. Together they form a unique fingerprint.

Cite this