Spring-Rod System Identification via Differentiable Physics Engine
This addresses system identification for complex physical systems like tensegrity structures, offering an explainable and data-efficient method, though it appears incremental as it builds on traditional physics engines with differentiable adaptations.
The paper tackles system identification for complex spring-rod assemblies by proposing a novel differentiable physics engine that modularizes governing equations and reduces dimensions, resulting in efficient learning with reduced training data and explainable parameters. It demonstrates efficacy on tensegrity systems like NASA's icosahedron.
We propose a novel differentiable physics engine for system identification of complex spring-rod assemblies. Unlike black-box data-driven methods for learning the evolution of a dynamical system \emph{and} its parameters, we modularize the design of our engine using a discrete form of the governing equations of motion, similar to a traditional physics engine. We further reduce the dimension from 3D to 1D for each module, which allows efficient learning of system parameters using linear regression. The regression parameters correspond to physical quantities, such as spring stiffness or the mass of the rod, making the pipeline explainable. The approach significantly reduces the amount of training data required, and also avoids iterative identification of data sampling and model training. We compare the performance of the proposed engine with previous solutions, and demonstrate its efficacy on tensegrity systems, such as NASA's icosahedron.