Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods

2018-02-28 · Deirdre Quillen, Eric Jang, Ofir Nachum, Chelsea Finn, Julian Ibarz, Sergey Levine

In this paper, we explore deep reinforcement learning algorithms for vision-based robotic grasping. Model-free deep reinforcement learning (RL) has been successfully applied to a range of challenging environments, but the proliferation of algorithms makes it difficult to discern which particular approach would be best suited for a rich, diverse task like grasping. To answer this question, we propose a simulated benchmark for robotic grasping that emphasizes off-policy learning and generalization to unseen objects. Off-policy learning enables utilization of grasping data over a wide variety of objects, and diversity is important to enable the method to generalize to new objects that were not seen during training. We evaluate the benchmark tasks against a variety of Q-function estimation methods, a method previously proposed for robotic grasping with deep neural network models, and a novel approach based on a combination of Monte Carlo return estimation and an off-policy correction. Our results indicate that several simple methods provide a surprisingly strong competitor to popular algorithms such as double Q-learning, and our analysis of stability sheds light on the relative tradeoffs between the algorithms.

📄 PDF Abstract BibTeX arXiv:1802.10264

Code (1)

smrjan/robotic-vision pytorch

Tasks

Deep Reinforcement LearningDiversityQ-Learningreinforcement-learning Reinforcement Learning Reinforcement Learning (RL)Robotic Grasping

Acceleration of Actor-Critic Deep Reinforcement Learning for Visual Grasping in Clutter by State Representation Learning Based on Disentanglement of a Raw Input Image

2020-02-27 · Taewon Kim, Yeseong Park, Youngbin Park, Il Hong Suh

For a robotic grasping task in which diverse unseen target objects exist in a cluttered environment, some deep learning-based methods have achieved state-of-the-art results using visual input directly. In contrast, actor…

Deep Reinforcement Learning Disentanglement Reinforcement Learning Reinforcement Learning (RL)+2

Robotics and Computer-Integrated Manufacturing

2024-04-01 · Robotics and Computer-Integrated Manufacturing 2024 4 · Yu Huang, Daxin Liu *, Zhenyu Liu, Ke Wang 외

To grasp the randomly moving objects in unstructured environment, a novel robotic grasping method based on multi-agent TD3 with high-quality memory (MA-TD3H) is proposed. During the grasping process, the MA-TD3H algo…

Robotic Grasping

Distributed Reinforcement Learning of Targeted Grasping with Active Vision for Mobile Manipulators

2020-07-16 · Yasuhiro Fujita, Kota Uenishi, Avinash Ummadisingu, Prabhat Nagarajan 외

Developing personal robots that can perform a diverse range of manipulation tasks in unstructured environments necessitates solving several challenges for robotic grasping systems. We take a step towards this broader goa…

Deep Reinforcement Learning GPU reinforcement-learning Reinforcement Learning (RL)+1

RL-CycleGAN: Reinforcement Learning Aware Simulation-To-Real

2020-06-16 · CVPR 2020 6 · Kanishka Rao, Chris Harris, Alex Irpan, Sergey Levine 외

Deep neural network based reinforcement learning (RL) can learn appropriate visual representations for complex tasks like vision-based robotic grasping without the need for manually engineering or prior learning a percep…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)Robotic Grasping+1

Quantile QT-Opt for Risk-Aware Vision-Based Robotic Grasping

2019-10-01 · Cristian Bodnar, Adrian Li, Karol Hausman, Peter Pastor 외

The distributional perspective on reinforcement learning (RL) has given rise to a series of successful Q-learning algorithms, resulting in state-of-the-art performance in arcade game environments. However, it has not yet…

Q-Learning Reinforcement Learning Reinforcement Learning (RL)Robotic Grasping