Newtonian and Lagrangian Neural Networks: A Comparison Towards Efficient Inverse Dynamics Identification

TL;DR

This paper compares Newtonian and Lagrangian neural networks for inverse dynamics modeling of industrial robots, highlighting that Newtonian networks perform better when estimating motor torques due to their explicit modeling of dissipative effects.

Contribution

The study provides guidance on choosing between Newtonian and Lagrangian neural networks for inverse dynamics, especially in the context of motor torque estimation.

Findings

Newtonian networks outperform Lagrangian networks when estimating motor torques.

Lagrangian networks are less effective due to lack of explicit dissipative torque modeling.

Comparison conducted on data from an industrial robot.

Abstract

Accurate inverse dynamics models are essential tools for controlling industrial robots. Recent research combines neural network regression with inverse dynamics formulations of the Newton-Euler and the Euler-Lagrange equations of motion, resulting in so-called Newtonian neural networks and Lagrangian neural networks, respectively. These physics-informed models seek to identify unknowns in the analytical equations from data. Despite their potential, current literature lacks guidance on choosing between Lagrangian and Newtonian networks. In this study, we show that when motor torques are estimated instead of directly measuring joint torques, Lagrangian networks prove less effective compared to Newtonian networks as they do not explicitly model dissipative torques. The performance of these models is compared to neural network regression on data of a MABI MAX 100 industrial robot.