Corpus ID: 202558635

MAT: Multi-Fingered Adaptive Tactile Grasping via Deep Reinforcement Learning

  title={MAT: Multi-Fingered Adaptive Tactile Grasping via Deep Reinforcement Learning},
  author={B. Wu and Iretiayo Akinola and J. Varley and P. Allen},
  • B. Wu, Iretiayo Akinola, +1 author P. Allen
  • Published in CoRL 2019
  • Computer Science
  • Vision-based grasping systems typically adopt an open-loop execution of a planned grasp. This policy can fail due to many reasons, including ubiquitous calibration error. Recovery from a failed grasp is further complicated by visual occlusion, as the hand is usually occluding the vision sensor as it attempts another open-loop regrasp. This work presents MAT, a tactile closed-loop method capable of realizing grasps provided by a coarse initial positioning of the hand above an object. Our… CONTINUE READING
    12 Citations

    Figures, Tables, and Topics from this paper

    Deep Reinforcement Learning for Tactile Robotics: Learning to Type on a Braille Keyboard
    • 1
    • PDF
    Blind Bin Picking of Small Screws Through In-finger Manipulation With Compliant Robotic Fingers
    Learning the sense of touch in simulation: a sim-to-real strategy for vision-based tactile sensing
    • 9
    • PDF
    Reactive grasping using high-resolution tactile sensors
    Generation of GelSight Tactile Images for Sim2Real Learning
    • PDF
    Observation Space Matters: Benchmark and Optimization Algorithm
    • PDF
    Machine Learning for Robotic Manipulation
    • PDF


    QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
    • 375
    • Highly Influential
    • PDF
    Pixel-Attentive Policy Gradient for Multi-Fingered Grasping in Cluttered Scenes
    • 8
    • PDF
    Leveraging Contact Forces for Learning to Grasp
    • 12
    • PDF
    More Than a Feeling: Learning to Grasp and Regrasp Using Vision and Touch
    • 89
    • PDF
    Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach
    • 144
    • PDF
    Planning Multi-Fingered Grasps as Probabilistic Inference in a Learned Deep Network
    • 39
    • PDF
    Learning of grasp adaptation through experience and tactile sensing
    • 88
    • PDF
    Learning to Grasp Without Seeing
    • 24
    • PDF
    Tactile Regrasp: Grasp Adjustments via Simulated Tactile Transformations
    • 42
    • PDF