Abstract:Modelling complex deformation for soft robotics provides a guideline to understand their behaviour, leading to safe interaction with the environment. However, building a surrogate model with high accuracy and fast inference speed can be challenging for soft robotics due to the nonlinearity from complex geometry, large deformation, material nonlinearity etc. The reality gap from surrogate models also prevents their further deployment in the soft robotics domain. In this study, we proposed a physics-informed Neural Networks (PINNs) named PINN-Ray to model complex deformation for a Fin Ray soft robotic gripper, which embeds the minimum potential energy principle from elastic mechanics and additional high-fidelity experimental data into the loss function of neural network for training. This method is significant in terms of its generalisation to complex geometry and robust to data scarcity as compared to other data-driven neural networks. Furthermore, it has been extensively evaluated to model the deformation of the Fin Ray finger under external actuation. PINN-Ray demonstrates improved accuracy as compared with Finite element modelling (FEM) after applying the data assimilation scheme to treat the sim-to-real gap. Additionally, we introduced our automated framework to design, fabricate soft robotic fingers, and characterise their deformation by visual tracking, which provides a guideline for the fast prototype of soft robotics.

What problem does this paper attempt to address?

### Problems Addressed by the Paper This paper aims to address the complex deformation issues in soft robot modeling. Specifically, the nonlinear structure, material properties, and large deformations of soft robots make it highly challenging to construct high-precision and fast inference surrogate models. Additionally, the gap between existing surrogate models and reality (sim-to-real gap) also limits their further application in the field of soft robotics. ### Solution To tackle these challenges, the authors propose a method based on Physics-Informed Neural Networks (PINNs), called PINN-Ray. PINN-Ray improves the generalization ability and robustness to data scarcity by embedding the principle of minimum potential energy in elasticity mechanics and high-fidelity experimental data into the loss function of the neural network during training. The method has been improved in the following aspects: 1. **Combining Physical Laws and Machine Learning**: PINN-Ray uses physical laws from elasticity mechanics theory as constraints, embedding them into the optimization process of the neural network to ensure the model adheres to physical principles. 2. **Data Assimilation**: By introducing experimental data assimilation techniques, actual measured data is integrated into the training process, reducing the gap between simulation and reality. 3. **Efficient Data Processing**: A real-time image processing technique is proposed to monitor the deformation of key points, improving the efficiency of performance evaluation. ### Experimental Validation The authors validate the effectiveness of the method by comparing the performance of PINN-Ray and the Finite Element Method (FEM) in modeling the deformation of Fin Ray soft fingers. Experimental results show that PINN-Ray, with data assimilation, significantly outperforms FEM and the standard version of PINN-Ray without data assimilation in terms of prediction accuracy. ### Main Contributions 1. **Proposed PINN-Ray**: Combining physical laws and machine learning to predict the deformation of soft robots. 2. **Data Assimilation**: Achieving sim-to-real processing by modifying the loss function. 3. **Image Processing Scheme**: For intrinsic key point tracking, efficiently characterizing the deformation of soft robots. 4. **In-depth Evaluation and Comparison**: Detailed evaluation and comparison of PINN-Ray, FEM, and experimental results in modeling the deformation of soft robots. Through these improvements, PINN-Ray not only enhances the accuracy of the model but also accelerates the convergence speed and more accurately estimates key variables such as strain and stress, providing strong support for the design and optimization of soft robots.

PINN-Ray: A Physics-Informed Neural Network to Model Soft Robotic Fin Ray Fingers

A Deep Learning Method for Vision Based Force Prediction of a Soft Fin Ray Gripper Using Simulation Data

A Physics-Informed Neural Network for Simulation of Finite Deformation in Hyperelastic-Magnetic Coupling Problems

Intrinsic Contact Sensing and Object Perception of an Adaptive Fin-Ray Gripper Integrating Compact Deflection Sensors

A Spatial-Temporally Adaptive PINN Framework for 3D Bi-Ventricular Electrophysiological Simulations and Parameter Inference

Physics-informed Neural Network Estimation of Material Properties in Soft Tissue Nonlinear Biomechanical Models

Physics-informed Neural Networks for Encoding Dynamics in Real Physical Systems

Learning physics-informed simulation models for soft robotic manipulation: A case study with dielectric elastomer actuators

Modeling finite-strain plasticity using physics informed neural network and assessment of the network performance

Energy-based physics-informed neural network for frictionless contact problems under large deformation

A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics

A stepwise physics‐informed neural network for solving large deformation problems of hypoelastic materials

Sim-to-Real of Soft Robots with Learned Residual Physics

A physics-informed neural network for Kresling origami structures

Development of a cascaded multitask physics-informed neural network (CM-PINN) to construct the muti-physical field model of rubber bushing press fitting

Scalable, Simulation-Guided Compliant Tactile Finger Design

PerSival: Neural-network-based visualisation for pervasive continuum-mechanical simulations in musculoskeletal biomechanics

Enhanced physics‐informed neural networks for hyperelasticity

Identifying Constitutive Parameters for Complex Hyperelastic Materials using Physics-Informed Neural Networks

Versatile 3D-printed fin-ray effect soft robotic fingers: lightweight optimization and performance analysis