Accurate Robotic Pouring for Serving Drinks
Yongqiang Huang, Yu Sun

TL;DR
This paper presents a neural network-based system for precise robotic pouring, achieving low error rates in pouring water, oil, and syrup, with potential applications in automated serving tasks.
Contribution
The authors develop a recurrent neural network system that accurately controls robotic pouring, demonstrating generalization to unseen containers and liquids.
Findings
Error as low as 4 milliliters in water pouring
Comparable accuracy in pouring oil and syrup
System generalizes to new containers and liquids
Abstract
Pouring is the second most frequently executed motion in cooking scenarios. In this work, we present our system of accurate pouring that generates the angular velocities of the source container using recurrent neural networks. We collected demonstrations of human pouring water. We made a physical system on which the velocities of the source container were generated at each time step and executed by a motor. We tested our system on pouring water from containers that are not used for training and achieved an error of as low as 4 milliliters. We also used the system to pour oil and syrup. The accuracy achieved with oil is slightly lower than but comparable with that of water.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRobot Manipulation and Learning · Smart Agriculture and AI · Robotics and Automated Systems
