Learning Quasi-Static 3D Models of Markerless Deformable Linear Objects for Bimanual Robotic Manipulation

TL;DR

This paper introduces a novel Transformer-based 3D model for deformable linear objects that outperforms existing models in accuracy and speed, with data augmentation enhancing overall performance for robotic manipulation tasks.

Contribution

A new Transformer architecture for DLO modeling with a scaling method and data augmentation, improving accuracy and efficiency over prior models.

Findings

Transformer model achieves superior accuracy across datasets.

Data augmentation significantly improves model performance.

Simple MLP with augmentation approaches state-of-the-art accuracy.

Abstract

The robotic manipulation of Deformable Linear Objects (DLOs) is a vital and challenging task that is important in many practical applications. Classical model-based approaches to this problem require an accurate model to capture how robot motions affect the deformation of the DLO. Nowadays, data-driven models offer the best tradeoff between quality and computation time. This paper analyzes several learning-based 3D models of the DLO and proposes a new one based on the Transformer architecture that achieves superior accuracy, even on the DLOs of different lengths, thanks to the proposed scaling method. Moreover, we introduce a data augmentation technique, which improves the prediction performance of almost all considered DLO data-driven models. Thanks to this technique, even a simple Multilayer Perceptron (MLP) achieves close to state-of-the-art performance while being significantly faster to evaluate. In the experiments, we compare the performance of the learning-based 3D models of the DLO on several challenging datasets quantitatively and demonstrate their applicability in the task of shaping a DLO.