Review of deep reinforcement learning-based object gras**: 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 …

Sim-to-real transfer of robotic control with dynamics randomization

XB Peng, M Andrychowicz, W Zaremba… - … on robotics and …, 2018 - ieeexplore.ieee.org
Simulations are attractive environments for training agents as they provide an abundant
source of data and alleviate certain safety concerns during the training process. But the …

Deep drone racing: From simulation to reality with domain randomization

A Loquercio, E Kaufmann, R Ranftl… - IEEE Transactions …, 2019 - ieeexplore.ieee.org
Dynamically changing environments, unreliable state estimation, and operation under
severe resource constraints are fundamental challenges that limit the deployment of small …

Driving policy transfer via modularity and abstraction

M Müller, A Dosovitskiy, B Ghanem, V Koltun - ar**, perception and interaction: A survey
S Garg, N Sünderhauf, F Dayoub… - … and Trends® in …, 2020 - nowpublishers.com
For robots to navigate and interact more richly with the world around them, they will likely
require a deeper understanding of the world in which they operate. In robotics and related …

Transfer in deep reinforcement learning using successor features and generalised policy improvement

A Barreto, D Borsa, J Quan, T Schaul… - International …, 2018 - proceedings.mlr.press
The ability to transfer skills across tasks has the potential to scale up reinforcement learning
(RL) agents to environments currently out of reach. Recently, a framework based on two …

An adaptive deep reinforcement learning framework enables curling robots with human-like performance in real-world conditions

DO Won, KR Müller, SW Lee - Science Robotics, 2020 - science.org
The game of curling can be considered a good test bed for studying the interaction between
artificial intelligence systems and the real world. In curling, the environmental characteristics …

Modular networks: Learning to decompose neural computation

L Kirsch, J Kunze, D Barber - Advances in neural …, 2018 - proceedings.neurips.cc
Scaling model capacity has been vital in the success of deep learning. For a typical network,
necessary compute resources and training time grow dramatically with model size …

Bev-seg: Bird's eye view semantic segmentation using geometry and semantic point cloud

MH Ng, K Radia, J Chen, D Wang, I Gog… - arxiv preprint arxiv …, 2020 - arxiv.org
Bird's-eye-view (BEV) is a powerful and widely adopted representation for road scenes that
captures surrounding objects and their spatial locations, along with overall context in the …

Pose-and-shear-based tactile servoing

J Lloyd, NF Lepora - The International Journal of Robotics …, 2024 - journals.sagepub.com
Tactile servoing is an important technique because it enables robots to manipulate objects
with precision and accuracy while adapting to changes in their environments in real-time …