SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Planning and Control
Explore this paper's citation graph
Summary
The deep dynamics model, a variant of SE3-Nets, learns a low-dimensional pose embedding for visuomotor control via an encoder-decoder structure that runs in real-time, achieves good prediction of scene dynamics and outperforms the baseline methods on multiple control runs.
- Type
- preprint
- Published
- 2017-10-02
- Cited by
- 43
- References
- 22
- Access
- Open access
- OpenAlex
- https://openalex.org/W2763676071
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:27257366
Keywords
Computer science, Artificial intelligence, Embedding, Computer vision, Segmentation
References
- Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
- From Pixels to Torques: Policy Learning with Deep Dynamical Models
- Learning predictive models of a depth camera & manipulator from raw execution traces
- Distinctive Image Features from Scale-Invariant Keypoints
- A tutorial on visual servo control
- A generalized iterative LQG method for locally-optimal feedback control of constrained nonlinear stochastic systems
- DART: Dense Articulated Real-Time Tracking
- SE3-nets: Learning rigid body motion using deep neural networks
- Deep visual foresight for planning robot motion
- Self-Supervised Visual Descriptor Learning for Dense Correspondence
- PVEs: Position-Velocity Encoders for Unsupervised Learning of Structured State Representations
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Learning Visual Servoing with Deep Features and Fitted Q-Iteration
- Learning to Poke by Poking: Experiential Learning of Intuitive Physics
- End-to-End Training of Deep Visuomotor Policies
- Visual Servoing
- Unsupervised Learning for Physical Interaction through Video Prediction
- End-to-end training of deep visuomotor policies
- Embed to control: a locally Linear Latent dynamics model for control from raw images
Cited by
- Combining learned and analytical models for predicting action effects from sensory data
- Visual Robot Task Planning
- Stochastic Adversarial Video Prediction
- The limits and potentials of deep learning for robotics
- On Machine Learning and Structure for Mobile Robots
- Learning, Moving, And Predicting With Global Motion Representations
- Time Reversal as Self-Supervision
- Learning Latent Space Dynamics for Tactile Servoing
- Dynamic Manipulation of Flexible Objects with Torque Sequence Using a Deep Neural Network
- Unsupervised Visuomotor Control through Distributional Planning Networks
- Learning Task Agnostic Sufficiently Accurate Models
- Self-Supervised Learning for Specified Latent Representation
- KeyIn: Discovering Subgoal Structure with Keyframe-based Video Prediction
- Physics-as-Inverse-Graphics: Joint Unsupervised Learning of Objects and Physics from Video
- Motion-Nets: 6D Tracking of Unknown Objects in Unseen Environments using RGB
- PIQA: Reasoning about Physical Commonsense in Natural Language
- Learning Pose Estimation for UAV Autonomous Navigation and Landing Using Visual-Inertial Sensor Data
- Learning Predictive Models From Observation and Interaction
- Predicting the Physical Dynamics of Unseen 3D Objects
- Learning Transformable and Plannable se(3) Features for Scene Imitation of a Mobile Service Robot