Generalization in Dexterous Manipulation via Geometry-Aware Multi-Task Learning

• URL: http://arxiv.org/abs/2111.03062v1
• Date: Thu, 4 Nov 2021 17:59:56 GMT
• Title: Generalization in Dexterous Manipulation via Geometry-Aware Multi-Task Learning
• Authors: Wenlong Huang, Igor Mordatch, Pieter Abbeel, Deepak Pathak
• Abstract summary: We show that policies learned by existing reinforcement learning algorithms can in fact be generalist. We show that a single generalist policy can perform in-hand manipulation of over 100 geometrically-diverse real-world objects. Interestingly, we find that multi-task learning with object point cloud representations not only generalizes better but even outperforms single-object specialist policies.
• Score: 108.08083976908195
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Dexterous manipulation of arbitrary objects, a fundamental daily task for humans, has been a grand challenge for autonomous robotic systems. Although data-driven approaches using reinforcement learning can develop specialist policies that discover behaviors to control a single object, they often exhibit poor generalization to unseen ones. In this work, we show that policies learned by existing reinforcement learning algorithms can in fact be generalist when combined with multi-task learning and a well-chosen object representation. We show that a single generalist policy can perform in-hand manipulation of over 100 geometrically-diverse real-world objects and generalize to new objects with unseen shape or size. Interestingly, we find that multi-task learning with object point cloud representations not only generalizes better but even outperforms the single-object specialist policies on both training as well as held-out test objects. Video results at https://huangwl18.github.io/geometry-dex

Related papers

• Dreamitate: Real-World Visuomotor Policy Learning via Video Generation [49.03287909942888]
  We propose a visuomotor policy learning framework that fine-tunes a video diffusion model on human demonstrations of a given task. We generate an example of an execution of the task conditioned on images of a novel scene, and use this synthesized execution directly to control the robot.
  arXiv  Detail & Related papers  (2024-06-24T17:59:45Z)
• Entity-Centric Reinforcement Learning for Object Manipulation from Pixels [22.104757862869526]
  Reinforcement Learning (RL) offers a general approach to learn object manipulation. In practice, domains with more than a few objects are difficult for RL agents due to the curse of dimensionality. We propose a structured approach for visual RL that is suitable for representing multiple objects and their interaction.
  arXiv  Detail & Related papers  (2024-04-01T16:25:08Z)
• Learning Generalizable Manipulation Policies with Object-Centric 3D Representations [65.55352131167213]
  GROOT is an imitation learning method for learning robust policies with object-centric and 3D priors. It builds policies that generalize beyond their initial training conditions for vision-based manipulation. GROOT's performance excels in generalization over background changes, camera viewpoint shifts, and the presence of new object instances.
  arXiv  Detail & Related papers  (2023-10-22T18:51:45Z)
• SpawnNet: Learning Generalizable Visuomotor Skills from Pre-trained Networks [52.766795949716986]
  We present a study of the generalization capabilities of the pre-trained visual representations at the categorical level. We propose SpawnNet, a novel two-stream architecture that learns to fuse pre-trained multi-layer representations into a separate network to learn a robust policy.
  arXiv  Detail & Related papers  (2023-07-07T13:01:29Z)
• Efficient Representations of Object Geometry for Reinforcement Learning of Interactive Grasping Policies [29.998917158604694]
  We present a reinforcement learning framework that learns the interactive grasping of various geometrically distinct real-world objects. Videos of learned interactive policies are available at https://maltemosbach.org/io/geometry_aware_grasping_policies.
  arXiv  Detail & Related papers  (2022-11-20T11:47:33Z)
• Learning Generalizable Dexterous Manipulation from Human Grasp Affordance [11.060931225148936]
  Dexterous manipulation with a multi-finger hand is one of the most challenging problems in robotics. Recent progress in imitation learning has largely improved the sample efficiency compared to Reinforcement Learning. We propose to learn dexterous manipulation using large-scale demonstrations with diverse 3D objects in a category.
  arXiv  Detail & Related papers  (2022-04-05T16:26:22Z)
• Learning Category-Level Generalizable Object Manipulation Policy via Generative Adversarial Self-Imitation Learning from Demonstrations [14.001076951265558]
  Generalizable object manipulation skills are critical for intelligent robots to work in real-world complex scenes. In this work, we tackle this category-level object manipulation policy learning problem via imitation learning in a task-agnostic manner. We propose several general but critical techniques, including generative adversarial self-imitation learning from demonstrations, progressive growing of discriminator, and instance-balancing for expert buffer.
  arXiv  Detail & Related papers  (2022-03-04T02:52:02Z)
• Generalizing Object-Centric Task-Axes Controllers using Keypoints [15.427056235112152]
  We learn modular task policies which compose object-centric task-axes controllers. These task-axes controllers are parameterized by properties associated with underlying objects in the scene. Our overall approach provides a simple, modular and yet powerful framework for learning manipulation tasks.
  arXiv  Detail & Related papers  (2021-03-18T21:08:00Z)
• COG: Connecting New Skills to Past Experience with Offline Reinforcement Learning [78.13740204156858]
  We show that we can reuse prior data to extend new skills simply through dynamic programming. We demonstrate the effectiveness of our approach by chaining together several behaviors seen in prior datasets for solving a new task. We train our policies in an end-to-end fashion, mapping high-dimensional image observations to low-level robot control commands.
  arXiv  Detail & Related papers  (2020-10-27T17:57:29Z)
• Learning Dexterous Grasping with Object-Centric Visual Affordances [86.49357517864937]
  Dexterous robotic hands are appealing for their agility and human-like morphology. We introduce an approach for learning dexterous grasping. Our key idea is to embed an object-centric visual affordance model within a deep reinforcement learning loop.
  arXiv  Detail & Related papers  (2020-09-03T04:00:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.