Accounts of using the Tustin-Net architecture on a rotary inverted pendulum

2024/08/22 by Stijn van Esch, van Esch, Stijn, Fabio Bonassi +3
Engineering · #Advanced Theoretical and Applied Studies in Material Sciences and Geometry #Engineering Technology and Methodologies #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Mechanics and Biomechanics Studies #Systems and Control (eess.SY) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2408.12266

openalex publication_date 2024/08/22 · openalex created_date 2024/12/20 · openalex updated_date 2026/07/28

Abstract

In this report we investigate the use of the Tustin neural network architecture (Tustin-Net) for the identification of a physical rotary inverse pendulum. This physics-based architecture is of particular interest as it builds on the known relationship between velocities and positions. We here aim at discussing the advantages, limitations and performance of Tustin-Nets compared to first-principles grey-box models on a real physical apparatus, showing how, with a standard training procedure, the former can hardly achieve the same accuracy as the latter. To address this limitation, we present a training strategy based on transfer learning that yields Tustin-Nets that are competitive with the first-principles model, without requiring extensive knowledge of the setup as the latter.

Related

• Tustin neural networks: a class of recurrent nets for adaptive MPC of mechanical systems
• An introduction to Neural Networks for Physicists
• ServoLNN: Lagrangian Neural Networks Driven by Servomechanisms
• Physics-based polynomial neural networks for one-shot learning of dynamical systems from one or a few samples
• Newtonian and Lagrangian Neural Networks: A Comparison Towards Efficient Inverse Dynamics Identification