Task-Difficulty-Aware Efficient Object Arrangement Leveraging Tossing Motions

TL;DR

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, and simultaneously learns the tossing motion through self-supervised learning and the task deter-mination policy via brute-force search.

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.

Figures (9)

• Figure 1: Training pipeline for PP-and-PT-based object arrangement.
• Figure 2: Possible C-patterns
• Figure 3: Overlooking of C1F1M0
• Figure 5: Task assignment table for policy-contact patterns
• Figure 6: Target objects. $O_i~(i\in\{1,2,3\})$ and $\bar{O}_j~(j\in\{1,2,3\})$ are the objects used (Trained) and not used (Unknown) in training