A Comparison of Imitation Learning Algorithms for Bimanual Manipulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913478584827904 |
|---|---|
| author | Drolet, Michael Stepputtis, Simon Kailas, Siva Jain, Ajinkya Peters, Jan Schaal, Stefan Amor, Heni Ben |
| author_facet | Drolet, Michael Stepputtis, Simon Kailas, Siva Jain, Ajinkya Peters, Jan Schaal, Stefan Amor, Heni Ben |
| contents | Amidst the wide popularity of imitation learning algorithms in robotics, their properties regarding hyperparameter sensitivity, ease of training, data efficiency, and performance have not been well-studied in high-precision industry-inspired environments. In this work, we demonstrate the limitations and benefits of prominent imitation learning approaches and analyze their capabilities regarding these properties. We evaluate each algorithm on a complex bimanual manipulation task involving an over-constrained dynamics system in a setting involving multiple contacts between the manipulated object and the environment. While we find that imitation learning is well suited to solve such complex tasks, not all algorithms are equal in terms of handling environmental and hyperparameter perturbations, training requirements, performance, and ease of use. We investigate the empirical influence of these key characteristics by employing a carefully designed experimental procedure and learning environment. Paper website: https://bimanual-imitation.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_06536 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Comparison of Imitation Learning Algorithms for Bimanual Manipulation Drolet, Michael Stepputtis, Simon Kailas, Siva Jain, Ajinkya Peters, Jan Schaal, Stefan Amor, Heni Ben Robotics Machine Learning Amidst the wide popularity of imitation learning algorithms in robotics, their properties regarding hyperparameter sensitivity, ease of training, data efficiency, and performance have not been well-studied in high-precision industry-inspired environments. In this work, we demonstrate the limitations and benefits of prominent imitation learning approaches and analyze their capabilities regarding these properties. We evaluate each algorithm on a complex bimanual manipulation task involving an over-constrained dynamics system in a setting involving multiple contacts between the manipulated object and the environment. While we find that imitation learning is well suited to solve such complex tasks, not all algorithms are equal in terms of handling environmental and hyperparameter perturbations, training requirements, performance, and ease of use. We investigate the empirical influence of these key characteristics by employing a carefully designed experimental procedure and learning environment. Paper website: https://bimanual-imitation.github.io/ |
| title | A Comparison of Imitation Learning Algorithms for Bimanual Manipulation |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2408.06536 |