Learning a visuomotor controller for real world robotic grasping using simulated depth images
Explore this paper's citation graph
Summary
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.
- Type
- preprint
- Published
- 2017-06-14
- Cited by
- 198
- References
- 23
- Access
- Open access
- OpenAlex
- https://openalex.org/W2626073992
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:23439582
Keywords
Artificial intelligence, Computer vision, Computer science, Robotic hand, Controller (irrigation)
References
- Leveraging big data for grasp planning
- Real-time grasp detection using convolutional neural networks
- Deep learning for detecting robotic grasps
- 3DNet: Large-scale object class recognition from CAD models
- Computer vision for fruit harvesting robots - state of the art and challenges ahead
- Gradient-based learning applied to document recognition
- Combined shape, appearance and silhouette for simultaneous manipulator and object tracking
- Human-level control through deep reinforcement learning
- Constructing force-closure grasps
- Caffe: Convolutional Architecture for Fast Feature Embedding
- OpenRAVE: A Planning Architecture for Autonomous Robotics
- Visual servoing for humanoid grasping and manipulation tasks
- Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours
- An Introduction to Reinforcement Learning
- High precision grasp pose detection in dense clutter
- Learning hand-eye coordination for robotic grasping with deep learning and large-scale data collection
- Active vision for dexterous grasping of novel objects
- Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics
- End-to-End Training of Deep Visuomotor Policies
- Springer Handbook of Robotics
Cited by
- Sim-to-real Transfer of Visuo-motor Policies for Reaching in Clutter: Domain Randomization and Adaptation with Modular Networks
- Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
- Pick and Place Without Geometric Object Models
- Domain Randomization and Generative Models for Robotic Grasping
- Asymmetric Actor Critic for Image-Based Robot Learning
- Learning Deep Policies for Robot Bin Picking by Simulating Robust Grasping Sequences
- Tactile Regrasp: Grasp Adjustments via Simulated Tactile Transformations
- Learning Kinematic Descriptions using SPARE: Simulated and Physical ARticulated Extendable dataset
- Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach
- Review of Deep Learning Methods in Robotic Grasp Detection
- Adversarial discriminative sim-to-real transfer of visuo-motor policies
- Learning task-oriented grasping for tool manipulation from simulated self-supervision
- QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
- Adapting control policies from simulation to reality using a pairwise loss
- Multi-View Picking: Next-best-view Reaching for Improved Grasping in Clutter
- Learning a High-Precision Robotic Assembly Task Using Pose Estimation from Simulated Depth Images
- A Data-Efficient Framework for Training and Sim-to-Real Transfer of Navigation Policies
- Dealing with Ambiguity in Robotic Grasping via Multiple Predictions
- Robotic grasping in multi-object stacking scenes based on visual reasoning
- Efficient Policy Learning for Robust Robot Grasping
Related papers
- Foreground object segmentation from binocular stereo video
- 6-DOF object localization by combining monocular vision and robot arm kinematics
- An Object Detection and Pose Estimation Approach for Position Based Visual Servoing
- Object-oriented stripe structured-light vision-guided robot
- Tracking in 3D: Image Variability Decomposition for Recovering Object Pose and Illumination
- Hand-eye calibration using a single image and robotic picking up using images lacking in contrast
- Motion-based Object Detection and Tracking in Color Image Sequence
- Transparent object detection and location based on RGB-D camera
- Stereo object tracking with fusion of texture, color and disparity information
- Robust object tracking based on RGB-D camera