Tidiness Score-Guided Monte Carlo Tree Search for Visual Tabletop Rearrangement
Hogun Kee, Wooseok Oh, Minjae Kang, Hyemin Ahn, Songhwai Oh
AI summary
Problem
Tabletop tidying research lacks public datasets and objective metrics to evaluate tidiness, while defining explicit goal configurations for unseen objects remains difficult.
Approach
The authors introduce a simulated dataset and train a vision-based discriminator to score tidiness, which guides a Monte Carlo tree search planner to generate pick-and-place trajectories without predefined goals.
Key results
- 88.5% tidying success rate in simulation
- 85% success rate in real-world robot experiments
- Human evaluation confirms tidied scenes match human perception of organization
- Release of the TTU dataset containing 224,225 simulated tidying trajectories
Why it matters
Enables goal-free robotic tidying in diverse real-world environments, advancing practical embodied AI for household and office automation.
Abstract
In this paper, we present the tidiness score-guided Monte Carlo tree search (TSMCTS), a novel framework designed to address the tabletop tidying up problem using only an RGB-D camera. We address two major problems for tabletop tidying up problem: (1) the lack of public datasets and benchmarks, and (2) the difficulty of specifying the goal configuration of unseen objects. We address the former by presenting the tabletop tidying up (TTU) dataset, a structured dataset collected in simulation. Using this dataset, we train a vision-based discriminator capable of predicting the tidiness score. This discriminator can consis- tently evaluate the degree of tidiness across unseen configurations, including real-world scenes. Addressing the second problem, we employ Monte Carlo tree search (MCTS) to find tidying trajectories without specifying explicit goals. Instead of providing specific goals, we demonstrate that our MCTS-based planner can find diverse tidied configurations using the tidiness score as a guidance. Consequently, we propose TSMCTS, which integrates a tidiness discriminator with an MCTS-based tidying planner to find optimal tidied arrangements. TSMCTS has successfully demonstrated its capability across various environments, includ- ing coffee tables, dining tables, office desks, and bathrooms. The TTU dataset is available at: https://github.com/rllab-snu/ TTU-Dataset.