Task-Difficulty-Aware Efficient Object Arrangement Leveraging Tossing Motions

TL;DR

This paper proposes a task-difficulty-aware method for efficient object arrangement using pick-and-toss motions, combining self-supervised learning and environment-based decision policies to improve robotic efficiency in diverse settings.

Contribution

It introduces a novel approach that adaptively chooses between pick-and-place and pick-and-toss based on environment difficulty, leveraging self-supervised learning for tossing motion.

Findings

Enhanced efficiency in object arrangement tasks.

Successful real-world implementation with various objects.

Effective environment-based decision policy.

Abstract

This study explores a pick-and-toss (PT) as an alternative to pick-and-place (PP), allowing a robot to extend its range and improve task efficiency. Although PT boosts efficiency in object arrangement, the placement environment critically affects the success of tossing. To achieve accurate and efficient object arrangement, we suggest choosing between PP and PT based on task difficulty estimated from the placement environment. Our method simultaneously learns the tossing motion through self-supervised learning and the task determination policy via brute-force search. Experimental results validate the proposed method through simulations and real-world tests on various rectangular object arrangements.