Data-Driven Precision Control of Soft Helicoid Robots
Organizations: Institute for Mechanical Systems, ETH Zürich, Leonhardstrasse 21, 8092, Zürich, Switzerland. · Department of Aeronautics and Astronautics, Stanford University, 496 Lomita Mall, Stanford, 94305, CA, USA. · Department of Industrial Engineering, University of Naples Federico II, P.le V. Tecchio, 80, 80125, Naples, Italy.
Abstract
A purely data-driven method grounded in the theory of adiabatic spectral submanifolds (aSSMs) was recently used to model and control nonlinear soft robotic simulators. Here, we extend the methodology to learn aSSM-reduced models from data collected from hardware soft robotic platforms. Specifically, we learn a five dimensional (5D) aSSM-reduced nonlinear ordinary differential equation (ODE) that accurately models the dynamics of a soft-helicoid trunk robot. We then use this ODE in a model predictive control scheme to demonstrate real-time precision control across various challenging closed-loop trajectory-tracking tasks, ranging from 2D to 3D trajectories with static and dynamic constraints. We find that our methodology enhances accuracy and precision across all control tasks compared to available data-driven linear methods and model-free reinforcement learning. While we focus here on a soft-helicoid trunk robot, our methodology relies solely on hardware data and is therefore applicable to a broader class of soft robotic systems.
Figures & tables
| Method | Model | Data | Training | Maximum | Constraints | Controller |
|---|---|---|---|---|---|---|
| dim. | collection time | time | error [cm] | handling | freq. [Hz] | |
| aSSM (ours) | 5 | 2 hours | 5 minutes | 5 | Yes | 50 |
| Koopman-pregain ( Haggerty et al., 2023 ) | 6 | 2 hours | 30 seconds | 38 | No | 100 |
| DMDc ( Proctor et al., 2016 ) | 6 | 2 hours | 30 seconds | 18 | Yes | 100 |
| MZ-Koopman ( Catalasan and He, 2026 ) | 33 | 2 hours | 15 minutes | 17 | Yes | 20 |
| RL (PPO) ( Schulman et al., 2017 ) | NA | 6 hours | 6 hours | 10 | No | 100 |
Data-Driven Control
Data-driven control uses measured data and learned models or policies to control systems when accurate first-principles models are unavailable.
See all 42 papers