Continuum Robot Modeling with Action Conditioned Flow Matching

2026-05-09

AI Summary

This work addresses the challenge of accurately predicting the steady-state shape of tendon-driven continuum robots, which arises from their continuous deformation, intricate tendon routing, compliance, friction, and manufacturing variations. The authors propose a self-modeling approach based on action-conditioned point cloud flow matching. By developing a lightweight 3D-printed robotic platform coupled with a multi-camera RGB-D capture system, and leveraging random quasi-static configuration sampling with conditional generative modeling, they introduce action-conditioned flow matching to the continuum robotics domain for the first time. This enables high-fidelity mapping from motor states to full 3D geometric configurations. The method demonstrates strong generalization within the same design family, extends effectively to loaded scenarios, and significantly outperforms existing approaches across real and simulated robots with 2, 3, and 5 modules, achieving substantial improvements in both Chamfer Distance (CD) and Earth Mover’s Distance (EMD) metrics.

Abstract

Predicting the shape of tendon driven continuum robots (TDCRs) at steady state from actuation remains challenging due to continuous deformation, complex tendon routing, compliance, friction, and fabrication variability. In this paper, we address this problem as kinematic self modeling conditioned on action. We present a lightweight 3D printed TDCR hardware platform and an RGB-D data collection pipeline with multiple cameras, and we learn a point cloud flow matching model that maps motor actuation states to the robot's settled 3D geometry. The model is trained from randomly sampled quasi static configurations and evaluated on test motor commands within the same TDCR design family and actuation range. We compare against prior 3D deformable object and robot self modeling approaches in both MuJoCo simulation and real hardware experiments. Experiments on simulated 2-, 3-, and 5-module TDCRs and real 2- and 3-module robots show improved shape prediction accuracy under CD and EMD metrics. We further show in simulation that the same conditional formulation generalizes to tip payload as a conditioning input, enabling payload conditioned steady-state shape prediction. These results demonstrate a data driven self modeling framework for quasi static TDCR geometry prediction.

Problem

• continuum robot
• shape prediction
• tendon-driven
• steady-state modeling
• kinematic self-modeling

Innovation

• continuum robot
• flow matching
• self-modeling
• point cloud
• tendon-driven