Neural Networks with Physics-Informed Architectures and Constraints for Dynamical Systems Modeling

09/14/2021
by   Franck Djeumou, et al.
0

Effective inclusion of physics-based knowledge into deep neural network models of dynamical systems can greatly improve data efficiency and generalization. Such a-priori knowledge might arise from physical principles (e.g., conservation laws) or from the system's design (e.g., the Jacobian matrix of a robot), even if large portions of the system dynamics remain unknown. We develop a framework to learn dynamics models from trajectory data while incorporating a-priori system knowledge as inductive bias. More specifically, the proposed framework uses physics-based side information to inform the structure of the neural network itself, and to place constraints on the values of the outputs and the internal states of the model. It represents the system's vector field as a composition of known and unknown functions, the latter of which are parametrized by neural networks. The physics-informed constraints are enforced via the augmented Lagrangian method during the model's training. We experimentally demonstrate the benefits of the proposed approach on a variety of dynamical systems – including a benchmark suite of robotics environments featuring large state spaces, non-linear dynamics, external forces, contact forces, and control inputs. By exploiting a-priori system knowledge during training, the proposed approach learns to predict the system dynamics two orders of magnitude more accurately than a baseline approach that does not include prior knowledge, given the same training dataset.

READ FULL TEXT

page 14

page 15

research
06/24/2022

ModLaNets: Learning Generalisable Dynamics via Modularity and Physical Inductive Bias

Deep learning models are able to approximate one specific dynamical syst...
research
04/02/2019

A comparative study of physics-informed neural network models for learning unknown dynamics and constitutive relations

We investigate the use of discrete and continuous versions of physics-in...
research
04/27/2023

Some of the variables, some of the parameters, some of the times, with some physics known: Identification with partial information

Experimental data is often comprised of variables measured independently...
research
12/02/2022

Guaranteed Conformance of Neurosymbolic Models to Natural Constraints

Deep neural networks have emerged as the workhorse for a large section o...
research
05/24/2020

Physics-based polynomial neural networks for one-short learning of dynamical systems from one or a few samples

This paper discusses an approach for incorporating prior physical knowle...
research
05/24/2020

Physics-based polynomial neural networks for one-shot learning of dynamical systems from one or a few samples

This paper discusses an approach for incorporating prior physical knowle...
research
02/27/2020

Tensor network approaches for learning non-linear dynamical laws

Given observations of a physical system, identifying the underlying non-...

Please sign up or login with your details

Forgot password? Click here to reset