paper-with-me

Papers

Modality-Driven Design for Multi-Step Dexterous Manipulation: Insights from Neuroscience

2024-12-15 · Naoki Wake, Atsushi Kanehira, Daichi Saito, Jun Takamatsu, Kazuhiro Sasabuchi, Hideki Koike, Katsushi Ikeuchi

Multi-step dexterous manipulation is a fundamental skill in household scenarios, yet remains an underexplored area in robotics. This paper proposes a modular approach, where each step of the manipulation process is addressed with dedicated policies based on effective modality input, rather than relying on a single end-to-end model. To demonstrate this, a dexterous robotic hand performs a manipulation task involving picking up and rotating a box. Guided by insights from neuroscience, the task is decomposed into three sub-skills, 1)reaching, 2)grasping and lifting, and 3)in-hand rotation, based on the dominant sensory modalities employed in the human brain. Each sub-skill is addressed using distinct methods from a practical perspective: a classical controller, a Vision-Language-Action model, and a reinforcement learning policy with force feedback, respectively. We tested the pipeline on a real robot to demonstrate the feasibility of our approach. The key contribution of this study lies in presenting a neuroscience-inspired, modality-driven methodology for multi-step dexterous manipulation.

📄 PDF Abstract BibTeX arXiv:2412.11337

Code (0)

등록된 구현이 없습니다.

Tasks

Vision-Language-Action

Similar Papers 제목 키워드 기반

A Tendon-Driven Five-Fingered Hand with Distributed Tactile Perception for Dexterous Manipulation

2026-08-26 · Huayang Chen, Longhui Qin arxiv

To apply the techniques of embodied artificial intelligence to human-oid robots for complex manipulations, dexterous robotic hands are indispensable, which are restricted by the dexterity and tactile perception capabilit…

Towards Robotic Dexterous Hand Intelligence: A Survey

2026-05-13 · Weiguang Zhao, Tian Liang, Xihao Guo, Rui Zhang 외 arxiv

Robotic dexterous hands are central to contact-rich manipulation, with rapid progress driven by advances in hardware, sensing, control, simulation, and data generation. However, existing studies are often developed under…

Multi-Goal Dexterous Hand Manipulation using Probabilistic Model-based Reinforcement Learning

2025-04-30 · Yingzhuo Jiang, Wenjun Huang, Rongdun Lin, Chenyang Miao 외

This paper tackles the challenge of learning multi-goal dexterous hand manipulation tasks using model-based Reinforcement Learning. We propose Goal-Conditioned Probabilistic Model Predictive Control (GC-PMPC) by designin…

Model-based Reinforcement LearningModel Predictive Control

Learning Robust Dexterous In-Hand Manipulation from Joint Sensors with Proprioceptive Transformer

2026-05-20 · Senlan Yao, Chenyu Yang, Jaehoon Kim, Aristotelis Sympetheros 외 arxiv

In-hand object manipulation is a fundamental yet challenging capability for dexterous robots. Despite significant progress in dexterous manipulation, existing approaches rely heavily on vision or tactile sensing to track…

Reinforcement Learning

MM-Hand: A 21-DOF Multi-modal Modular Dexterous Robotic Hand with Remote Actuation

2026-04-19 · Zhuoheng Li, Qingquan Lin, Checheng Yu, Qiangyu Chen 외 arxiv

High-DOF dexterous hands require compact actuation, rich sensing, and reliable thermal behavior, but conventional designs often occupy valuable in-hand space, increase end-effector mass, and suffer from heat accumulation…