Learn a consistent jacobian model for bio-inspired soft robotic fingers
Learning a Resolution-Consistent Jacobian Field for Bio-Inspired Rigid-Soft Finger
Robotics
Summary
Modeling bio-inspired robotic fingers that combine rigid and soft parts is difficult because their movements are complex and change depending on position. The authors developed a new learning method called Jacobian Flow Matching that produces a smooth and consistent model of how finger movements respond to actions over time. This model works well at different speeds and improves prediction accuracy, helping control the finger better during long sequences of motion. Their tests showed large reductions in error compared to traditional methods, making the model more reliable for tasks involving these soft-rigid fingers.
What this means in practice
- •For soft robotic engineers: Improve control algorithms for robotic fingers that combine hard and soft parts by using a model that predicts finger movement consistently at varying control speeds.
- •For robotic manipulation teams: Enable more accurate offline trajectory planning for complex tendon-driven fingers through a learned Jacobian field supporting long-term motion predictions.
Authors
Tianyou Liang, Haisen Zeng, Shanjun Chen, YiMing Zhu, Zhongyue Lu, Zirong Luo
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.