Physical Simulation

Physical simulation aims to accurately model and predict the behavior of physical systems, often using computational methods to solve complex equations governing their dynamics. Current research emphasizes developing more efficient and accurate simulation techniques, focusing on graph neural networks (GNNs), message-passing transformers, and physics-informed machine learning models to improve speed and accuracy, particularly for complex systems like fluids and deformable bodies. These advancements are crucial for various applications, including robotics, autonomous driving, and scientific discovery, by enabling faster and more realistic simulations for design, testing, and analysis.

Papers

May 21, 2024

Equivariant Spatio-Temporal Attentive Graph Networks to Simulate Physical Dynamics
Liming Wu, Zhichao Hou, Jirui Yuan, Yu Rong, Wenbing Huang
Spatio Temporal Physical Simulation Equivariant Graph Neural Network Equivariant Graph Attention

April 11, 2024

QuasiSim: Parameterized Quasi-Physical Simulators for Dexterous Manipulations Transfer
Xueyi Liu, Kangbo Lyu, Jieqiong Zhang, Tao Du, Li Yi
Physical Simulation

March 3, 2024

ML4PhySim : Machine Learning for Physical Simulations Challenge (The airfoil design)
Mouadh Yagoubi, Milad Leyli-Abadi, David Danan, Jean-Patrick Brunet, Jocelyn Ahmed Mazari, Florent Bonnet, Asma Farjallah, Marc Schoenauer, Patrick Gallinari
Machine Learning Physical Simulation Airfoil Design Machine Learning Based Surrogate

February 29, 2024

Learning a Generalized Physical Face Model From Data
Lingchen Yang, Gaspard Zoss, Prashanth Chandran, Markus Gross, Barbara Solenthaler, Eftychios Sifakis, Derek Bradley
LeArning Abstract Raw Data Facial Animation Physical Simulation 3D Facial Animation

February 2, 2024

Efficient Physically-based Simulation of Soft Bodies in Embodied Environment for Surgical Robot
Zhenya Yang, Yonghao Long, Kai Chen, Wang Wei, Qi Dou
Robotic Surgery Physical Simulation Surgical Robot Elastomer Deformation Soft Body Embodied Environment Surgical Robot Learning

January 27, 2024

SimFair: Physics-Guided Fairness-Aware Learning with Simulation Models
Zhihao Wang, Yiqun Xie, Zhili Li, Xiaowei Jia, Zhe Jiang, Aolin Jia, Shuo Xu
Physical Simulation Mechanistic Model Fairness Aware Simulation Model Fairness Aware Learning Fairness Verification

January 22, 2024

First-principles Based 3D Virtual Simulation Testing for Discovering SOTIF Corner Cases of Autonomous Driving
Lehang Li, Haokuan Wu, Botao Yao, Tianyu He, Shuohan Huang, Chuanyi Liu
Autonomous Driving Physical Simulation LiDAR Sensor Sensing Model 3D Simulator Safety of the Intended Functionality

January 17, 2024

Space and Time Continuous Physics Simulation From Partial Observations
Janny Steeven, Nadri Madiha, Digne Julie, Wolf Christian
Deep Space Fluid Dynamic Physical Simulation Differentiable Physic Partial Observation Large Scale Machine Learning Numerical Scheme Continuous Prediction

December 21, 2023

SimLM: Can Language Models Infer Parameters of Physical Systems?
Sean Memery, Mirella Lapata, Kartic Subr
Physical System Physical Simulation Parameter Inference Complex Physical System

November 29, 2023

DisMech: A Discrete Differential Geometry-based Physical Simulator for Soft Robots and Structures
Andrew Choi, Ran Jing, Andrew Sabelhaus, Mohammad Khalid Jawed
Soft Robot Soft Robotics Soft Actuator Discrete Environment Physical Simulation Simple Structure Soft Structure Soft Continuum Robot

November 9, 2023

Diffusion-Generative Multi-Fidelity Learning for Physical Simulation
Zheng Wang, Shibo Li, Shikai Fang, Shandian Zhe
Multi Fidelity Physical Simulation Generative Diffusion

October 12, 2023

Reinforcement Learning of Display Transfer Robots in Glass Flow Control Systems: A Physical Simulation-Based Approach
Hwajong Lee, Chan Kim, Seong-Woo Kim
Reinforcement Learning Physical Simulation Flow Control Robot Scheduling

October 11, 2023

DeepSimHO: Stable Pose Estimation for Hand-Object Interaction via Physics Simulation
Rong Wang, Wei Mao, Hongdong Li
Hand Object Interaction Physic Informed Physical Simulation Hand Object Contact Dynamic Continuous Pose DeepSim Network

October 9, 2023

Examining the simulation-to-reality gap of a wheel loader digging in deformable terrain
Koji Aoshima, Martin Servin
Real Time Physical Simulation Reality Gap Deformable Terrain

September 25, 2023

SuPerPM: A Large Deformation-Robust Surgical Perception Framework Based on Deep Point Matching Learned from Physical Constrained Simulation Data
Shan Lin, Albert J. Miao, Ali Alabiad, Fei Liu, Kaiyuan Wang, Jingpei Lu, Florian Richter, Michael C. Yip
Physical Simulation Surgical Scene Surgical Datasets Non Rigid Point Cloud Point Matching

August 5, 2023

Physics-Based Task Generation through Causal Sequence of Physical Interactions
Chathura Gamage, Vimukthini Pinto, Matthew Stephenson, Jochen Renz
Physic Based Physical Reasoning Physical Simulation Physical Interaction Causal Order

June 12, 2023

Differentiable Multi-Fidelity Fusion: Efficient Learning of Physics Simulations with Neural Architecture Search and Transfer Learning
Yuwen Deng, Wang Kang, Wei W. Xing
Deep Learning Transfer Learning Neural Architecture Search Surrogate Model Efficient Learning Multi Fidelity Physical Simulation Closed Form Differentiable Expression Training Sample

May 18, 2023

PETAL: Physics Emulation Through Averaged Linearizations for Solving Inverse Problems
Jihui Jin, Etienne Ollivier, Richard Touret, Matthew McKinley, Karim G. Sabra, Justin K. Romberg
Inverse Problem Surrogate Modeling Physical Simulation Reverse Graph Linearization Forward Model Ocean Dynamic Acoustic Imaging

March 31, 2023

On the Relationships between Graph Neural Networks for the Simulation of Physical Systems and Classical Numerical Methods
Artur P. Toshev, Ludger Paehler, Andrea Panizza, Nikolaus A. Adams
Machine Learning Graph Neural Network Machine Learning Model Simulation Study Human Relationship Physical System Physical Simulation Numerical Method Particle Method

March 17, 2023

Component reusability evaluation and requirement tracing for agent-based cyber-physical-simulated systems
Wojciech Dudek, Narcis Miguel, Tomasz Winiarski
Robotic System Physical Simulation Cyber Physical Quantitative Evaluation Meta Model RecSys Challenge Language SysML

Physical Simulation

Papers

Equivariant Spatio-Temporal Attentive Graph Networks to Simulate Physical Dynamics

QuasiSim: Parameterized Quasi-Physical Simulators for Dexterous Manipulations Transfer

ML4PhySim : Machine Learning for Physical Simulations Challenge (The airfoil design)

Learning a Generalized Physical Face Model From Data

Efficient Physically-based Simulation of Soft Bodies in Embodied Environment for Surgical Robot

SimFair: Physics-Guided Fairness-Aware Learning with Simulation Models

First-principles Based 3D Virtual Simulation Testing for Discovering SOTIF Corner Cases of Autonomous Driving

Space and Time Continuous Physics Simulation From Partial Observations

SimLM: Can Language Models Infer Parameters of Physical Systems?

DisMech: A Discrete Differential Geometry-based Physical Simulator for Soft Robots and Structures

Diffusion-Generative Multi-Fidelity Learning for Physical Simulation

Reinforcement Learning of Display Transfer Robots in Glass Flow Control Systems: A Physical Simulation-Based Approach

DeepSimHO: Stable Pose Estimation for Hand-Object Interaction via Physics Simulation

Examining the simulation-to-reality gap of a wheel loader digging in deformable terrain

SuPerPM: A Large Deformation-Robust Surgical Perception Framework Based on Deep Point Matching Learned from Physical Constrained Simulation Data

Physics-Based Task Generation through Causal Sequence of Physical Interactions

Differentiable Multi-Fidelity Fusion: Efficient Learning of Physics Simulations with Neural Architecture Search and Transfer Learning

PETAL: Physics Emulation Through Averaged Linearizations for Solving Inverse Problems

On the Relationships between Graph Neural Networks for the Simulation of Physical Systems and Classical Numerical Methods

Component reusability evaluation and requirement tracing for agent-based cyber-physical-simulated systems