Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach
Explore this paper's citation graph
Summary
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.
- Type
- preprint
- Published
- 2018-04-14
- Cited by
- 660
- References
- 37
- Access
- Open access
- OpenAlex
- https://openalex.org/W2798255267
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:4891707
Keywords
GRASP, Artificial intelligence, Computer science, Convolutional neural network, Clutter
References
- Survey on Visual Servoing for Manipulation
- Handling Objects by Their Handles
- Robotic grasping and contact: a review
- Real-time grasp detection using convolutional neural networks
- Fully convolutional networks for semantic segmentation
- SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
- Robot Grasp Synthesis Algorithms: A Survey
- Playing catch and juggling with a humanoid robot
- Deep learning for detecting robotic grasps
- An overview of 3D object grasp synthesis algorithms
- Data-Driven Grasp Synthesis—A Survey
- Robotic Grasping of Novel Objects using Vision
- Automatic grasp planning using shape primitives
- Efficient grasping from RGBD images: Learning using a new rectangle representation
- Visually guided object grasping
- Grasp Planning via Decomposition Trees
- Visual servoing for humanoid grasping and manipulation tasks
- Deep Residual Learning for Image Recognition
- Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours
- Generating multi-fingered robotic grasps via deep learning
Cited by
- QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
- Zero-shot Sim-to-Real Transfer with Modular Priors
- Multi-View Picking: Next-best-view Reaching for Improved Grasping in Clutter
- Densely Supervised Grasp Detector (DSGD)
- Training Frankenstein's Creature to Stack: HyperTree Architecture Search
- Learning ambidextrous robot grasping policies
- On the choice of grasp type and location when handing over an object
- On-Policy Dataset Synthesis for Learning Robot Grasping Policies Using Fully Convolutional Deep Networks
- Efficient Fully Convolution Neural Network for Generating Pixel Wise Robotic Grasps With High Resolution Images
- Mechanical Search: Multi-Step Retrieval of a Target Object Occluded by Clutter
- Towards Robust Product Packing with a Minimalistic End-Effector
- The CoSTAR Block Stacking Dataset: Learning with Workspace Constraints
- A Space-Variant Visual Pathway Model for Data Efficient Deep Learning
- Prospection: Interpretable plans from language by predicting the future
- Vision for Robust Robot Manipulation
- Learning Probabilistic Multi-Modal Actor Models for Vision-Based Robotic Grasping
- Learning Manipulation Skills via Hierarchical Spatial Attention
- An overview of robot vision
- Learning the signatures of the human grasp using a scalable tactile glove
- Learning robust, real-time, reactive robotic grasping
Related papers
- Deep learning for detecting robotic grasps
- Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours
- Real-time grasp detection using convolutional neural networks
- 1Data-Driven Grasp Synthesis- A Survey
- Efficient grasping from RGBD images: Learning using a new rectangle representation
- Learning ambidextrous robot grasping policies