• Corpus ID: 49470584

QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation

@article{Kalashnikov2018QTOptSD,
  title={QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation},
  author={Dmitry Kalashnikov and Alex Irpan and Peter Pastor and Julian Ibarz and Alexander Herzog and Eric Jang and Deirdre Quillen and Ethan Holly and Mrinal Kalakrishnan and Vincent Vanhoucke and Sergey Levine},
  journal={ArXiv},
  year={2018},
  volume={abs/1806.10293}
}
In this paper, we study the problem of learning vision-based dynamic manipulation skills using a scalable reinforcement learning approach. [...] Key Method In contrast to static learning behaviors that choose a grasp point and then execute the desired grasp, our method enables closed-loop vision-based control, whereby the robot continuously updates its grasp strategy based on the most recent observations to optimize long-horizon grasp success.Expand
Robotic Grasping using Deep Reinforcement Learning
TLDR
This work uses an off-policy reinforcement learning framework along with a novel GraspQ-Network to output grasp probabilities used to learn grasps that maximize the pick success.
MAT: Multi-Fingered Adaptive Tactile Grasping via Deep Reinforcement Learning
TLDR
MAT, a tactile closed-loop method capable of realizing grasps provided by a coarse initial positioning of the hand above an object, is presented, a deep reinforcement learning (RL) policy optimized through the clipped surrogate objective within a maximum entropy RL framework to balance exploitation and exploration.
Learning of Long-Horizon Sparse-Reward Robotic Manipulator Tasks with Base Controllers
TLDR
This paper proposes a method of learning long-horizon sparse-reward tasks utilizing one or more existing traditional controllers named base controllers built upon Deep Deterministic Policy Gradients (DDPG).
Robotic Pushing and Grasping Knowledge Learning via Attention Deep Q-learning Network
TLDR
The results show that the proposed attention deep Q-learning network for robotic grasping with the assistance of pushing actions with non-sparse rewards can not only improve robotic grasping performance but also avoid unnecessary pushing actions to improve action efficiency.
Robotic grasp manipulation using evolutionary computing and deep reinforcement learning
TLDR
This paper has developed learning-based pose estimation by decomposing the problem into both position and orientation learning and develops a deep reinforcement learning (DRL) model which is named as grasp deep Q-network (GDQN).
Learn to grasp unknown objects in robotic manipulation
TLDR
This paper introduces a robotic grasping strategy based on the model-free deep reinforcement learning, named Deep Reinforcement Grasp Policy, which demands minimal training time and limited simple objects in simulation and generalizes efficiently on novel objects in real-world scenario.
Learning robust, real-time, reactive robotic grasping
TLDR
A novel approach to perform object-independent grasp synthesis from depth images via deep neural networks overcomes shortcomings in existing techniques, namely discrete sampling of grasp candidates and long computation times, and achieves better performance, particularly in clutter.
Q-attention: Enabling Efficient Learning for Vision-based Robotic Manipulation
TLDR
This work presents the Attentiondriven Robotic Manipulation (ARM) algorithm, which is a general manipulation algorithm that can be applied to a range of sparse-rewarded tasks, given only a small number of demonstrations.
Ensemble Bootstrapped Deep Deterministic Policy Gradient for Vision-Based Robotic Grasping
TLDR
The simulation results show that, under the same number of interactions, the manipulators’ success rate in grabbing 1000 random objects has increased more than double and reached state-of-the-art (SOTA) performance.
A Framework for Efficient Robotic Manipulation
TLDR
It is shown that, given only 10 demonstrations, a single robotic arm can learn sparse-reward manipulation policies from pixels, such as reaching, picking, moving, pulling a large object, flipping a switch, and opening a drawer in just 15-50 minutes of real-world training time.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 46 REFERENCES
Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods
TLDR
This paper proposes a simulated benchmark for robotic grasping that emphasizes off-policy learning and generalization to unseen objects, and indicates that several simple methods provide a surprisingly strong competitor to popular algorithms such as double Q-learning.
Learning Synergies Between Pushing and Grasping with Self-Supervised Deep Reinforcement Learning
TLDR
This work demonstrates that it is possible to discover and learn complex synergies between non-prehensile and prehensile actions from scratch through model-free deep reinforcement learning, and achieves better grasping success rates and picking efficiencies than baseline alternatives after a few hours of training.
Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach
TLDR
The proposed Generative Grasping Convolutional Neural Network (GG-CNN) predicts the quality and pose of grasps at every pixel, overcomes limitations of current deep-learning grasping techniques by avoiding discrete sampling of grasp candidates and long computation times.
Learning Hand-Eye Coordination for Robotic Grasping with Large-Scale Data Collection
TLDR
A large convolutional neural network is trained to predict the probability that task-space motion of the gripper will result in successful grasps, using only monocular camera images and independently of camera calibration or the current robot pose.
Learning a visuomotor controller for real world robotic grasping using simulated depth images
TLDR
This paper proposes an approach to learning a closed-loop controller for robotic grasping that dynamically guides the gripper to the object and finds that this approach significantly outperforms the baseline in the presence of kinematic noise, perceptual errors and disturbances of the object during grasping.
End-to-End Training of Deep Visuomotor Policies
TLDR
This paper develops a method that can be used to learn policies that map raw image observations directly to torques at the robot's motors, trained using a partially observed guided policy search method, with supervision provided by a simple trajectory-centric reinforcement learning method.
Regrasping Using Tactile Perception and Supervised Policy Learning
TLDR
This paper presents a machine learning method that is able to learn a regrasping behavior that corrects failed grasps based on tactile perception and improves over time and uses simple linear policies to guide the general policy to avoid poor local minima and to learn the generalpolicy from smaller amounts of data.
Continuous Deep Q-Learning with Model-based Acceleration
TLDR
This paper derives a continuous variant of the Q-learning algorithm, which it is called normalized advantage functions (NAF), as an alternative to the more commonly used policy gradient and actor-critic methods, and substantially improves performance on a set of simulated robotic control tasks.
Deep learning for detecting robotic grasps
TLDR
This work presents a two-step cascaded system with two deep networks, where the top detections from the first are re-evaluated by the second, and shows that this method improves performance on an RGBD robotic grasping dataset, and can be used to successfully execute grasps on two different robotic platforms.
Self-Supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigation
TLDR
A generalized computation graph is proposed that subsumes value-based model-free methods and model-based methods, and is instantiate to form a navigation model that learns from raw images and is sample efficient, and outperforms single-step and double-step double Q-learning.
...
1
2
3
4
5
...