From Biological Curves to Soft Modular Robots: Learning-Based Inverse Design with Uncertainty-Guided Refinement
From Biological Curves to Soft Modular Robots: Learning-Based Inverse Design with Uncertainty-Guided Refinement
This research presents a learning-based real-to-sim-to-real framework that translates biological curves into robotic configurations through surrogate-assisted inverse design and uncertainty-guided refinement. The method reduces deformation prediction time from 1.53 × 10⁵ ms to 0.0008 ms and design time from 1.99 × 10⁶ ms to 4.18 ms. It achieves a 93% shape reconstruction success rate on 5,000 test samples and is validated through physical demonstrations with four bioinspired robots.
This paper introduces a learning-based framework for translating biological curves into robotic configurations, using surrogate models and uncertainty-guided refinement to significantly reduce design and prediction times. The method achieves a 93% shape reconstruction success rate on test samples and is validated through physical demonstrations with bioinspired robots.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
deformation prediction time
0.0008
View original evidence
refinement. The proposed method combines learned inverse prediction with targeted candidate exploration, using a fast forward surrogate to evaluate and select suitable designs. The learned forward surrogate reduces deformation prediction time from $1.53 \times 10^5$ ms to 0.0008 ms, achieving approximately a $1.9 \times 10^8$-fold speedup. At the same time, the proposed inverse design method reduces design time from
Open source S4design time
4.18
View original evidence
$1.99 \times 10^6$ ms to 4.18 ms. On 5,000 test samples generated from previously unseen threading patterns, the method achieves a 93% shape reconstruction success rate under surrogate-based evaluation. Physical demonstrations with four bioinspired robots further illustrate the translation of biological curves into realizable robotic shapes.
Open source S5test samples
5,000
View original evidence
$1.99 \times 10^6$ ms to 4.18 ms. On 5,000 test samples generated from previously unseen threading patterns, the method achieves a 93% shape reconstruction success rate under surrogate-based evaluation. Physical demonstrations with four bioinspired robots further illustrate the translation of biological curves into realizable robotic shapes.
Open source S5forward surrogate prediction time
0.0008
View original evidence
refinement. The proposed method combines learned inverse prediction with targeted candidate exploration, using a fast forward surrogate to evaluate and select suitable designs. The learned forward surrogate reduces deformation prediction time from $1.53 \times 10^5$ ms to 0.0008 ms, achieving approximately a $1.9 \times 10^8$-fold speedup. At the same time, the proposed inverse design method reduces design time from
Open source S4
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
$1.99 \times 10^6$ ms to 4.18 ms. On 5,000 test samples generated from previously unseen threading patterns, the method achieves a 93% shape reconstruction success rate under surrogate-based evaluation. Physical demonstrations with four bioinspired robots further illustrate the translation of biological curves into realizable robotic shapes.
Open source S5
refinement. The proposed method combines learned inverse prediction with targeted candidate exploration, using a fast forward surrogate to evaluate and select suitable designs. The learned forward surrogate reduces deformation prediction time from $1.53 \times 10^5$ ms to 0.0008 ms, achieving approximately a $1.9 \times 10^8$-fold speedup. At the same time, the proposed inverse design method reduces design time from
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org