
TL;DR
This paper discusses the use of Functional Object Oriented Networks (FOON) for structured knowledge representation in robots, focusing on task tree generation, traversal algorithms, and achieving goal nodes in robotic tasks.
Contribution
It introduces FOON as a structured knowledge representation and compares various algorithms for traversing task trees to improve robotic task execution.
Findings
FOON effectively models task relationships in robotics.
Different search algorithms have been evaluated for task tree traversal.
FOON can be used to enhance robot understanding and execution of complex tasks.
Abstract
Robots are man made machines which are used to accomplish the tasks. Robots are mainly used to do complex tasks and work in hazardous environment where humans are difficult to work. They are not only designed to use in hazardous environment but also in the environment where humans are performing the same task repeatedly. These are also used for cooking purpose some tasks can be completed with the interaction of both the human and robot one of such things is cooking where human should help robot in making dishes. This paper mainly focusses on Functional Object Oriented Network which is structured knowledge representation using the input output and motion nodes. Task tress are generated using the task tree FOON is produced and collections of all FOONS forms the universal FOON. Different algorithms to traverse the tree in order to get the best output are also discussed in this paper. The…
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
TopicsRobotics and Automated Systems · Advanced Neural Network Applications · Robot Manipulation and Learning
