Research

Paper

TESTING February 18, 2026

SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation

Authors

Kushal Kedia, Tyler Ga Wei Lum, Jeannette Bohg, C. Karen Liu

Abstract

The ability to manipulate tools significantly expands the set of tasks a robot can perform. Yet, tool manipulation represents a challenging class of dexterity, requiring grasping thin objects, in-hand object rotations, and forceful interactions. Since collecting teleoperation data for these behaviors is challenging, sim-to-real reinforcement learning (RL) is a promising alternative. However, prior approaches typically require substantial engineering effort to model objects and tune reward functions for each task. In this work, we propose SimToolReal, taking a step towards generalizing sim-to-real RL policies for tool manipulation. Instead of focusing on a single object and task, we procedurally generate a large variety of tool-like object primitives in simulation and train a single RL policy with the universal goal of manipulating each object to random goal poses. This approach enables SimToolReal to perform general dexterous tool manipulation at test-time without any object or task-specific training. We demonstrate that SimToolReal outperforms prior retargeting and fixed-grasp methods by 37% while matching the performance of specialist RL policies trained on specific target objects and tasks. Finally, we show that SimToolReal generalizes across a diverse set of everyday tools, achieving strong zero-shot performance over 120 real-world rollouts spanning 24 tasks, 12 object instances, and 6 tool categories.

Metadata

arXiv ID: 2602.16863
Provider: ARXIV
Primary Category: cs.RO
Published: 2026-02-18
Fetched: 2026-02-21 18:51

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.16863v1</id>\n    <title>SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation</title>\n    <updated>2026-02-18T20:42:39Z</updated>\n    <link href='https://arxiv.org/abs/2602.16863v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.16863v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>The ability to manipulate tools significantly expands the set of tasks a robot can perform. Yet, tool manipulation represents a challenging class of dexterity, requiring grasping thin objects, in-hand object rotations, and forceful interactions. Since collecting teleoperation data for these behaviors is challenging, sim-to-real reinforcement learning (RL) is a promising alternative. However, prior approaches typically require substantial engineering effort to model objects and tune reward functions for each task. In this work, we propose SimToolReal, taking a step towards generalizing sim-to-real RL policies for tool manipulation. Instead of focusing on a single object and task, we procedurally generate a large variety of tool-like object primitives in simulation and train a single RL policy with the universal goal of manipulating each object to random goal poses. This approach enables SimToolReal to perform general dexterous tool manipulation at test-time without any object or task-specific training. We demonstrate that SimToolReal outperforms prior retargeting and fixed-grasp methods by 37% while matching the performance of specialist RL policies trained on specific target objects and tasks. Finally, we show that SimToolReal generalizes across a diverse set of everyday tools, achieving strong zero-shot performance over 120 real-world rollouts spanning 24 tasks, 12 object instances, and 6 tool categories.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.RO'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-02-18T20:42:39Z</published>\n    <arxiv:primary_category term='cs.RO'/>\n    <author>\n      <name>Kushal Kedia</name>\n    </author>\n    <author>\n      <name>Tyler Ga Wei Lum</name>\n    </author>\n    <author>\n      <name>Jeannette Bohg</name>\n    </author>\n    <author>\n      <name>C. Karen Liu</name>\n    </author>\n  </entry>"
}