Saved in:
Bibliographic Details
Main Authors: Cho, Dong Heon, Chen, Boyuan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.22039
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915883500175360
author Cho, Dong Heon
Chen, Boyuan
author_facet Cho, Dong Heon
Chen, Boyuan
contents Differentiable simulators enable gradient-based optimization of soft robots over material parameters, control, and morphology, but accurately modeling real systems remains challenging due to the sim-to-real gap. This issue becomes more pronounced when geometry is itself a design variable. System identification reduces discrepancies by fitting global material parameters to data; however, when constitutive models are misspecified or observations are sparse, identified parameters often absorb geometry-dependent effects rather than reflect intrinsic material behavior. More expressive constitutive models can improve accuracy but substantially increase computational cost, limiting practicality. We propose a residual acceleration field learning (RAFL) framework that augments a base simulator with a transferable, element-level corrective dynamics field. Operating on shared local features, the model is agnostic to global mesh topology and discretization. Trained end-to-end through a differentiable simulator using sparse marker observations, the learned residual generalizes across shapes. In both sim-to-sim and sim-to-real experiments, our method achieves consistent zero-shot improvements on unseen morphologies, while system identification frequently exhibits negative transfer. The framework also supports continual refinement, enabling simulation accuracy to accumulate during morphology optimization.
format Preprint
id arxiv_https___arxiv_org_abs_2603_22039
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RAFL: Generalizable Sim-to-Real of Soft Robots with Residual Acceleration Field Learning
Cho, Dong Heon
Chen, Boyuan
Robotics
Machine Learning
Differentiable simulators enable gradient-based optimization of soft robots over material parameters, control, and morphology, but accurately modeling real systems remains challenging due to the sim-to-real gap. This issue becomes more pronounced when geometry is itself a design variable. System identification reduces discrepancies by fitting global material parameters to data; however, when constitutive models are misspecified or observations are sparse, identified parameters often absorb geometry-dependent effects rather than reflect intrinsic material behavior. More expressive constitutive models can improve accuracy but substantially increase computational cost, limiting practicality. We propose a residual acceleration field learning (RAFL) framework that augments a base simulator with a transferable, element-level corrective dynamics field. Operating on shared local features, the model is agnostic to global mesh topology and discretization. Trained end-to-end through a differentiable simulator using sparse marker observations, the learned residual generalizes across shapes. In both sim-to-sim and sim-to-real experiments, our method achieves consistent zero-shot improvements on unseen morphologies, while system identification frequently exhibits negative transfer. The framework also supports continual refinement, enabling simulation accuracy to accumulate during morphology optimization.
title RAFL: Generalizable Sim-to-Real of Soft Robots with Residual Acceleration Field Learning
topic Robotics
Machine Learning
url https://arxiv.org/abs/2603.22039