Fachbereich Informatik

Multisensory Robot Pouring Perception

Master Thesis at group TAMS


Robotic pouring is a crucial robotic task in both domestic and industrial environments. A robot is required to pour a liquid from one container to another while preventing it from spilling. Therefore, the robust and accurate perception will play an essential role in this task, especially in estimating the liquid height in the target container. In previous work, we built a large-scale multimodal dataset (Containing video, audio, force/torque and position information collected during human pouring.) with a focus on the perception task for robotic pouring, which contains more than 3000 human pouring sequences. Therefore, making use of the force, motion trajectories and visual data from our multimodal dataset and studying the complementarity and interaction between multiple modalities in robotic pouring would also be an exciting direction of research.