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 …

Transporter networks: Rearranging the visual world for robotic manipulation

A Zeng, P Florence, J Tompson… - … on Robot Learning, 2021 - proceedings.mlr.press
Robotic manipulation can be formulated as inducing a sequence of spatial displacements:
where the space being moved can encompass an object, part of an object, or end effector. In …

Learning skills from demonstrations: A trend from motion primitives to experience abstraction

M Tavassoli, S Katyara, M Pozzi… - … on Cognitive and …, 2023 - ieeexplore.ieee.org
The uses of robots are changing from static environments in factories to encompass novel
concepts such as human–robot collaboration in unstructured settings. Preprogramming all …

Equivariant Learning in Spatial Action Spaces

D Wang, R Walters, X Zhu… - Conference on Robot …, 2022 - proceedings.mlr.press
Recently, a variety of new equivariant neural network model architectures have been
proposed that generalize better over rotational and reflectional symmetries than standard …

Symmetric models for visual force policy learning

C Kohler, AS Srikanth, E Arora… - 2024 IEEE International …, 2024 - ieeexplore.ieee.org
While it is generally acknowledged that force feedback is beneficial to robotic control,
applications of policy learning to robotic manipulation typically only leverage visual …

Multi-task learning with sequence-conditioned transporter networks

MH Lim, A Zeng, B Ichter, M Bandari… - … on Robotics and …, 2022 - ieeexplore.ieee.org
Enabling robots to solve multiple manipulation tasks has a wide range of industrial
applications. While learning-based approaches enjoy flexibility and generalizability, scaling …

Action priors for large action spaces in robotics

O Biza, D Wang, R Platt, JW van de Meent… - arXiv preprint arXiv …, 2021 - arxiv.org
In robotics, it is often not possible to learn useful policies using pure model-free
reinforcement learning without significant reward shaping or curriculum learning. As a …

Learning manipulation skills via hierarchical spatial attention

M Gualtieri, R Platt - IEEE Transactions on Robotics, 2020 - ieeexplore.ieee.org
Learning generalizable skills in robotic manipulation has long been challenging due to real-
world sized observation and action spaces. One method for addressing this problem is …

Visual Foresight With a Local Dynamics Model

C Kohler, R Platt - The International Symposium of Robotics Research, 2022 - Springer
Abstract Model-free policy learning has been shown to be capable of learning manipulation
policies which can solve long-time horizon tasks using single-step manipulation primitives …

[图书][B] Sequential Decision Making under Uncertainty: Optimality Guarantees, Compositional Learning, and Applications to Robotics and Ecology

HJ Lim - 2023 - search.proquest.com
Sequential decision making under uncertainty problems often deal with partially observable
Markov decision processes (POMDPs). POMDPs mathematically capture making decisions …