cs.ROOct 1, 2026

Learning a Resolution-Consistent Jacobian Field for Bio-Inspired Rigid-Soft Finger

Authors: Tianyou Liang, Haisen Zeng, Shanjun Chen, YiMing Zhu, Zhongyue Lu, Zirong Luo

Organizations: College of Intelligence Science and Technology, National University of Defense Technology,, Changsha, 410073, China · National Key Laboratory of Equipment State Sensing and Smart Support, National University of Defense Technology,, Changsha, 410073, China

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Bio-inspired tendon-driven rigid-soft coupled dexterous fingers exhibit strong nonlinearity and configuration-dependent sensitivity, making accurate modeling challenging. In discrete-time control, Jacobian-based kinematic algorithms typically rely on point-wise local linear approximations, which makes their performance sensitive to sensor sampling frequency and controller update frequency. To address this issue, we propose Jacobian Flow Matching (JFM), a structured learning framework based on Conditional Flow Matching (CFM), to learn a resolution-consistent Jacobian field that models actuation-to-motion transitions as a dynamical flow. The proposed framework supports both single-step prediction and continuous rollout via ODE integration, enabling consistent inference across temporal resolutions. Experiments on a tendon-driven rigid-soft finger show that the proposed method suppresses outlier errors and improves single-step prediction accuracy, reducing the global average RMSE by over 53% compared with a baseline discrete Jacobian learning approach. For long-horizon prediction, trajectories recovered via ODE integration achieve higher fidelity under sparse sampling (Stride = 8), reducing the RMSE median by 14.43% and the error variance by 24.87%. These results demonstrate that the learned flow-based Jacobian field provides an effective local model for offline multi-step trajectory optimization in rigid-soft coupled nonlinear systems.

Figures & tables

Figure 1: Bioinspired anthropomorphic robotic finger [ 18 , 19 ]
Figure 2: Overall framework of the proposed learning and inference pipeline
Figure 3: Servo-to-joint angle mapping (Z-axis flexion)
Figure 4: Trajectory prediction for different strides.
Figure 5: the open-loop trajectory prediction results’ box
Figure 6: the open-loop trajectory prediction results’ box
Figure 7: the inverse prediction results
Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 1 : Data collection pipeline overview
ParameterValue
Total samples∼104\sim 10^{4}
Hand pose frequency4545 Hz
Servo state frequency5555 Hz
Resampling methodNearest-neighbor
Output formatHDF5
Appendix
Table 1 : Data acquisition parameters

Explore similar work

CardsList

  1. Reference-Augmented Learning for Precise Tracking Policy of Tendon-Driven Continuum Robots

    Apr 28, 2026Ziqing Zou, Ke Qiu, Haojian Lu +2Robust Trajectory

    Department of Control Science and Engineering, Zhejiang University, Hangzhou, China

  2. Learning Control as Enabling Layer for Embodied Intelligence Research explored with Soft Robotic Swimming in diverse Flow Speeds

    Jun 9, 2026Fabian Schwab, Federico Allione, Bingcheng Wang +5Soft RoboticsLearning-Based Control

    1Engineering Sciences Department, Swiss Federal Laboratories for Materials Science and Technology, Switzerland · Institute for NeuroInformatics, University of Zurich, Switzerland · Department of Electronic Engineering, University of Rome Tor Vergata, Italy

  3. Learning-Based Dynamics Modeling and Robust Control for Tendon-Driven Continuum Robots

    Apr 28, 2026Ziqing Zou, Ke Qiu, Fei Wang +3Learning-Based ControlRobust Control

    Department of Control Science and Engineering, Zhejiang University, Hangzhou, China