This article discusses the use of search algorithms to retrieve task trees from a Functional Object-Oriented Network (FOON). FOONs are used by robots to carry out tasks via a series of steps. The study uses Universal FOON, which contains many cooking recipes, to retrieve a task tree that outlines the steps required to reach a goal node (goal output recipe). Two search algorithms, Iterative deepening search (IDS) and Greedy best-first search (GBFS), were used. The performance of the algorithms varied with different goal nodes. The study highlights the potential of FOONs and search algorithms in enabling robots to perform tasks effectively.
Publication date: 12 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.05346
