Challenge: We want to enable robots to learn everyday manipulation tasks from the Web and thus make abstract knowledge actionable.
For a robot to successfully perform such a breadth of maniulation actions, it needs to understand given tasks. On this website we show how robots can use a knowledge graph to understand a given task by answering questions such as
"What is the result state when performing a specific action?"
"What object can be used to perform a given task?" "Is there a preferred object to use for this task?"
"Can the task be performed on a given object?" etc.
We start tackling this challenge by focusing on selected task domains, starting with cutting. The idea is that our methodology can teach robots to cut any fruit with any to for any purpose, as visualised below. For this, the robot needs to query the knowledge graph to parameterise its general action plan.
Usage: Please choose a cutting verb and a food object from the dropdown lists below. Based on a food cutting knowledge graph, a sequence of motions will be returned that a robot can use for action parameterisation.