Generalizable and Fast Surrogates: Model Predictive Control of Articulated Soft Robots Using Physics-Informed Neural Networks
- 12 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 42 (15523098) , 619-636
Abstract
Soft robots can revolutionize several applications with high demands on dexterity and safety. When operating these systems, real-time estimation and control require fast and accurate models. However, prediction with first-principles (FP) models is slow, and learned black-box models have poor generalizability. Physics-informed machine learning offers excellent advantages here, but it is currently limited to simple, often simulated systems without considering changes after training. We propose physics-informed neural networks (PINNs) for articulated soft robots (ASRs) with a focus on data efficiency. The amount of expensive real-world training data is reduced to a minimum — one dataset in one system domain. Two hours of data in different domains are used for a comparison against two gold-standard approaches: In contrast to a recurrent neural network, the PINN provides a high generalizability. The prediction speed of an accurate FP model is exceeded with the PINN by up to a factor of 467 at slightly reduced accuracy. This enables nonlinear model predictive control (MPC) of a pneumatic ASR. Accurate position tracking with the MPC running at 47 Hz is achieved in six dynamic experiments.Keywords
All Related Versions
Funding Information
- Deutsche Forschungsgemeinschaft
- DFG, German Research Foundation (433586601)
- DFG, German Research Foundation (INST 187/742-1)
This publication has 65 references indexed in Scilit:
- Learning Controllers for Continuum Soft Manipulators: Impact of Modeling and Looming ChallengesAdvanced Intelligent Systems, 2024
- Physics‐Informed Neural Networks to Model and Control Robots: A Theoretical and Experimental InvestigationAdvanced Intelligent Systems, 2024
- Physics-informed neural nets for control of dynamical systemsNeurocomputing, 2024
- Nonlinear Model-Based Control of a Pneumatically Driven RobotIFAC-PapersOnLine, 2023
- Physics-informed Neural Networks-based Model Predictive Control for Multi-link ManipulatorsIFAC-PapersOnLine, 2022
- Soft Robots2020
- Nonlinear System Identification2020
- Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equationsJournal of Computational Physics, 2019
- CasADi: a software framework for nonlinear optimization and optimal controlMathematical Programming Computation, 2018
- Solving Ordinary Differential Equations II1996