Composing Dextrous Grasping and In-hand Manipulation via Scoring with a Reinforcement Learning Critic

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Röstel, Lennart, Winkelbauer, Dominik, Pitz, Johannes, Sievers, Leon, Bäuml, Berthold
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916948529381376
author Röstel, Lennart
Winkelbauer, Dominik
Pitz, Johannes
Sievers, Leon
Bäuml, Berthold
author_facet Röstel, Lennart
Winkelbauer, Dominik
Pitz, Johannes
Sievers, Leon
Bäuml, Berthold
contents In-hand manipulation and grasping are fundamental yet often separately addressed tasks in robotics. For deriving in-hand manipulation policies, reinforcement learning has recently shown great success. However, the derived controllers are not yet useful in real-world scenarios because they often require a human operator to place the objects in suitable initial (grasping) states. Finding stable grasps that also promote the desired in-hand manipulation goal is an open problem. In this work, we propose a method for bridging this gap by leveraging the critic network of a reinforcement learning agent trained for in-hand manipulation to score and select initial grasps. Our experiments show that this method significantly increases the success rate of in-hand manipulation without requiring additional training. We also present an implementation of a full grasp manipulation pipeline on a real-world system, enabling autonomous grasping and reorientation even of unwieldy objects.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13253
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Composing Dextrous Grasping and In-hand Manipulation via Scoring with a Reinforcement Learning Critic
Röstel, Lennart
Winkelbauer, Dominik
Pitz, Johannes
Sievers, Leon
Bäuml, Berthold
Robotics
Artificial Intelligence
In-hand manipulation and grasping are fundamental yet often separately addressed tasks in robotics. For deriving in-hand manipulation policies, reinforcement learning has recently shown great success. However, the derived controllers are not yet useful in real-world scenarios because they often require a human operator to place the objects in suitable initial (grasping) states. Finding stable grasps that also promote the desired in-hand manipulation goal is an open problem. In this work, we propose a method for bridging this gap by leveraging the critic network of a reinforcement learning agent trained for in-hand manipulation to score and select initial grasps. Our experiments show that this method significantly increases the success rate of in-hand manipulation without requiring additional training. We also present an implementation of a full grasp manipulation pipeline on a real-world system, enabling autonomous grasping and reorientation even of unwieldy objects.
title Composing Dextrous Grasping and In-hand Manipulation via Scoring with a Reinforcement Learning Critic
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2505.13253