Lightweight Learning from Actuation-Space Demonstrations via Flow Matching for Whole-Body Soft Robotic Grasping

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Liudi, Bai, Yang, Wang, Yuhao, Alsarraj, Ibrahim, Kutyniok, Gitta, Wang, Zhanchi, Wu, Ke
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917381383651328
author Yang, Liudi
Bai, Yang
Wang, Yuhao
Alsarraj, Ibrahim
Kutyniok, Gitta
Wang, Zhanchi
Wu, Ke
author_facet Yang, Liudi
Bai, Yang
Wang, Yuhao
Alsarraj, Ibrahim
Kutyniok, Gitta
Wang, Zhanchi
Wu, Ke
contents Robotic grasping under uncertainty remains a fundamental challenge due to its uncertain and contact-rich nature. Traditional rigid robotic hands, with limited degrees of freedom and compliance, rely on complex model-based and heavy feedback controllers to manage such interactions. Soft robots, by contrast, exhibit embodied mechanical intelligence: their underactuated structures and passive flexibility of their whole body, naturally accommodate uncertain contacts and enable adaptive behaviors. To harness this capability, we propose a lightweight actuation-space learning framework that infers distributional control representations for whole-body soft robotic grasping, directly from deterministic demonstrations using a flow matching model (Rectified Flow),without requiring dense sensing or heavy control loops. Using only 30 demonstrations (less than 8% of the reachable workspace), the learned policy achieves a 97.5% grasp success rate across the whole workspace, generalizes to grasped-object size variations of +-33%, and maintains stable performance when the robot's dynamic response is directly adjusted by scaling the execution time from 20% to 200%. These results demonstrate that actuation-space learning, by leveraging its passive redundant DOFs and flexibility, converts the body's mechanics into functional control intelligence and substantially reduces the burden on central controllers for this uncertain-rich task.
format Preprint
id arxiv_https___arxiv_org_abs_2511_01770
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lightweight Learning from Actuation-Space Demonstrations via Flow Matching for Whole-Body Soft Robotic Grasping
Yang, Liudi
Bai, Yang
Wang, Yuhao
Alsarraj, Ibrahim
Kutyniok, Gitta
Wang, Zhanchi
Wu, Ke
Robotics
Robotic grasping under uncertainty remains a fundamental challenge due to its uncertain and contact-rich nature. Traditional rigid robotic hands, with limited degrees of freedom and compliance, rely on complex model-based and heavy feedback controllers to manage such interactions. Soft robots, by contrast, exhibit embodied mechanical intelligence: their underactuated structures and passive flexibility of their whole body, naturally accommodate uncertain contacts and enable adaptive behaviors. To harness this capability, we propose a lightweight actuation-space learning framework that infers distributional control representations for whole-body soft robotic grasping, directly from deterministic demonstrations using a flow matching model (Rectified Flow),without requiring dense sensing or heavy control loops. Using only 30 demonstrations (less than 8% of the reachable workspace), the learned policy achieves a 97.5% grasp success rate across the whole workspace, generalizes to grasped-object size variations of +-33%, and maintains stable performance when the robot's dynamic response is directly adjusted by scaling the execution time from 20% to 200%. These results demonstrate that actuation-space learning, by leveraging its passive redundant DOFs and flexibility, converts the body's mechanics into functional control intelligence and substantially reduces the burden on central controllers for this uncertain-rich task.
title Lightweight Learning from Actuation-Space Demonstrations via Flow Matching for Whole-Body Soft Robotic Grasping
topic Robotics
url https://arxiv.org/abs/2511.01770