Long-Horizon Planning and Execution with Functional Object-Oriented Networks
This work addresses the challenge of bridging abstract symbolic planning to robot execution, enabling more flexible task handling in robotics, though it is incremental as it builds on existing FOON representations.
The paper tackled the problem of executing abstract task plans from functional object-oriented networks (FOON) by introducing a method to automatically transform FOON into PDDL and using a hierarchical planning pipeline to generate executable plans for robots, demonstrating it on long-horizon tasks in CoppeliaSim with extensions to unseen scenarios.
Following work on joint object-action representations, functional object-oriented networks (FOON) were introduced as a knowledge graph representation for robots. A FOON contains symbolic concepts useful to a robot's understanding of tasks and its environment for object-level planning. Prior to this work, little has been done to show how plans acquired from FOON can be executed by a robot, as the concepts in a FOON are too abstract for execution. We thereby introduce the idea of exploiting object-level knowledge as a FOON for task planning and execution. Our approach automatically transforms FOON into PDDL and leverages off-the-shelf planners, action contexts, and robot skills in a hierarchical planning pipeline to generate executable task plans. We demonstrate our entire approach on long-horizon tasks in CoppeliaSim and show how learned action contexts can be extended to never-before-seen scenarios.