Discovering efficient periodic behaviors in mechanical systems via neural approximators

DOI10.1002/OCA.3025MaRDI QIDQ6180273FDO

Authors Christoph Brune, Stefano Stramigioli, F. Califano

Publication date 19 January 2024

Published in Optimal Control Applications \& Methods (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2212.14253

zbMATH Keywords

neural networksstabilizationnonlinear oscillation

Mathematics Subject Classification ID

Control of mechanical systems (70Q05) Nonlinear systems in control theory (93C10) Stabilization of systems by feedback (93D15)

Abstract: It is well known that conservative mechanical systems exhibit local oscillatory behaviours due to their elastic and gravitational potentials, which completely characterise these periodic motions together with the inertial properties of the system. The classification of these periodic behaviours and their geometric characterisation are in an on-going secular debate, which recently led to the so-called eigenmanifold theory. The eigenmanifold characterises nonlinear oscillations as a generalisation of linear eigenspaces. With the motivation of performing periodic tasks efficiently, we use tools coming from this theory to construct an optimization problem aimed at inducing desired closed-loop oscillations through a state feedback law. We solve the constructed optimization problem via gradient-descent methods involving neural networks. Extensive simulations show the validity of the approach.

Recommendations

• An optimal control approach to the design of periodic orbits for mechanical systems with impacts
• Periodic motion planning and control for underactuated mechanical systems
• Optimal control for nonlinear oscillations of natural mechanical systems
• Optimal Energy Shaping via Neural Approximators
• Neural control for coordinated natural oscillation patterns

Cites work

• Computing Lyapunov functions using deep neural networks
• Data-driven discovery of coordinates and governing equations
• Data-driven science and engineering. Machine learning, dynamical systems, and control
• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• Iterative learning control and repetitive control for engineering practice
• Nonlinear robust periodic output regulation of minimum phase systems
• Normal Modes for Non-Linear Vibratory Systems
• Optimal Energy Shaping via Neural Approximators
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Reinforcement learning. An introduction

This page was built for publication: Discovering efficient periodic behaviors in mechanical systems via neural approximators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6180273)