Forward Dynamics Estimation from Data-Driven Inverse Dynamics Learning

TL;DR

This paper introduces a data-driven approach to estimate forward dynamics of mechanical systems by learning inverse dynamics models and extracting physical components, improving accuracy and interpretability.

Contribution

The paper presents a novel method to derive forward dynamics from inverse dynamics models, leveraging classical rigid body dynamics for better structure and physical interpretability.

Findings

Effective in learning forward dynamics on simulated robots

Outperforms direct forward dynamics learning in accuracy

Enables structured modeling of physical components

Abstract

In this paper, we propose to estimate the forward dynamics equations of mechanical systems by learning a model of the inverse dynamics and estimating individual dynamics components from it. We revisit the classical formulation of rigid body dynamics in order to extrapolate the physical dynamical components, such as inertial and gravitational components, from an inverse dynamics model. After estimating the dynamical components, the forward dynamics can be computed in closed form as a function of the learned inverse dynamics. We tested the proposed method with several machine learning models based on Gaussian Process Regression and compared them with the standard approach of learning the forward dynamics directly. Results on two simulated robotic manipulators, a PANDA Franka Emika and a UR10, show the effectiveness of the proposed method in learning the forward dynamics, both in terms of accuracy as well as in opening the possibility of using more structured~models.