A review of robot learning for manipulation: Challenges, representations, and algorithms

O Kroemer, S Niekum, G Konidaris - Journal of machine learning research, 2021 - jmlr.org
A key challenge in intelligent robotics is creating robots that are capable of directly
interacting with the world around them to achieve their goals. The last decade has seen …

Review of deep reinforcement learning-based object grasping: Techniques, open challenges, and recommendations

MQ Mohammed, KL Chung, CS Chyi - IEEE Access, 2020 - ieeexplore.ieee.org
The motivation behind our work is to review and analyze the most relevant studies on deep
reinforcement learning-based object manipulation. Various studies are examined through a …

Learning synergies between pushing and grasping with self-supervised deep reinforcement learning

A Zeng, S Song, S Welker, J Lee… - 2018 IEEE/RSJ …, 2018 - ieeexplore.ieee.org
Skilled robotic manipulation benefits from complex synergies between non-prehensile (eg
pushing) and prehensile (eg grasping) actions: pushing can help rearrange cluttered objects …

Actionable models: Unsupervised offline reinforcement learning of robotic skills

Y Chebotar, K Hausman, Y Lu, T Xiao… - arXiv preprint arXiv …, 2021 - arxiv.org
We consider the problem of learning useful robotic skills from previously collected offline
data without access to manually specified rewards or additional online exploration, a setting …

Contactnets: Learning discontinuous contact dynamics with smooth, implicit representations

S Pfrommer, M Halm, M Posa - Conference on Robot …, 2021 - proceedings.mlr.press
Common methods for learning robot dynamics assume motion is continuous, causing
unrealistic model predictions for systems undergoing discontinuous impact and stiction …

Augmenting physical simulators with stochastic neural networks: Case study of planar pushing and bouncing

A Ajay, J Wu, N Fazeli, M Bauza… - 2018 IEEE/RSJ …, 2018 - ieeexplore.ieee.org
An efficient, generalizable physical simulator with universal uncertainty estimates has wide
applications in robot state estimation, planning, and control. In this paper, we build such a …

Learning rigid dynamics with face interaction graph networks

KR Allen, Y Rubanova, T Lopez-Guevara… - arXiv preprint arXiv …, 2022 - arxiv.org
Simulating rigid collisions among arbitrary shapes is notoriously difficult due to complex
geometry and the strong non-linearity of the interactions. While graph neural network (GNN) …

Reactive planar non-prehensile manipulation with hybrid model predictive control

FR Hogan, A Rodriguez - The International Journal of …, 2020 - journals.sagepub.com
This article presents an offline solution and online approximation to the hybrid control
problem of planar non-prehensile manipulation. Hybrid dynamics and underactuation are …

Dipn: Deep interaction prediction network with application to clutter removal

B Huang, SD Han, A Boularias… - 2021 IEEE International …, 2021 - ieeexplore.ieee.org
We propose a Deep Interaction Prediction Network (DIPN) for learning to predict complex
interactions that ensue as a robot end-effector pushes multiple objects, whose physical …

Meta-learning priors for efficient online bayesian regression

J Harrison, A Sharma, M Pavone - … of Robotics XIII: Proceedings of the …, 2020 - Springer
Gaussian Process (GP) regression has seen widespread use in robotics due to its
generality, simplicity of use, and the utility of Bayesian predictions. The predominant …