To Stir or Not to Stir: Online Estimation of Liquid Properties for Pouring Actions

Our brains are able to exploit coarse physical models of fluids to solve complex manipulation tasks. There has been considerable interest in developing such a capability in robots so that they can autonomously manipulate fluids adapting to different conditions. In this paper, we investigate the problem of adaptation to liquids with different characteristics. We develop a simple training task (stirring with a stick) that enables rapid inference of the parameters of the liquid with relatively inexpensive measurement equipment (standard webcams) that robots may be assumed to have access to in the wild. We perform the inference in the space of simulation parameters rather than on physically accurate parameters. This facilitates prediction and optimization tasks since the inferred parameters may be fed directly to the simulator. First, we +demonstrate that our “stirring” learner performs better than when the robot is trained with pouring actions. Further, we show that our method is able to infer properties of three very different liquids – water, glycerin and gel – and improve spillage with increased training. We present various experimental results performed by executing stirring and pouring actions on a UR10. We believe that this decoupling of the training actions from the goal task is a significant first step towards simple and autonomous learning of the behavior of different fluids in unstructured environments.