Physics-Informed Learning for the Friction Modeling of High-Ratio Harmonic Drives

TL;DR

This paper introduces a scalable Physics-Informed Neural Network approach for friction modeling in high-ratio harmonic drives, improving control performance without dedicated testing setups.

Contribution

It presents a novel PINN-based friction identification method that leverages intrinsic robot models, eliminating the need for specialized sensors and setups.

Findings

PINN-based models outperform traditional static friction models.

Enhanced control performance with real-time friction compensation.

Method demonstrates scalability across multiple robot joints.

Abstract

This paper presents a scalable method for friction identification in robots equipped with electric motors and high-ratio harmonic drives, utilizing Physics-Informed Neural Networks (PINN). This approach eliminates the need for dedicated setups and joint torque sensors by leveraging the robo\v{t}s intrinsic model and state data. We present a comprehensive pipeline that includes data acquisition, preprocessing, ground truth generation, and model identification. The effectiveness of the PINN-based friction identification is validated through extensive testing on two different joints of the humanoid robot ergoCub, comparing its performance against traditional static friction models like the Coulomb-viscous and Stribeck-Coulomb-viscous models. Integrating the identified PINN-based friction models into a two-layer torque control architecture enhances real-time friction compensation. The results demonstrate significant improvements in control performance and reductions in energy losses, highlighting the scalability and robustness of the proposed method, also for application across a large number of joints as in the case of humanoid robots.