Neural Network Augmented Physics Models for Systems with Partially Unknown Dynamics: Application to Slider-Crank Mechanism

De Groote, Wannes; Kikken, Edward; Hostens, Erik; Van Hoecke, Sofie; Crevecoeur, Guillaume

doi:10.48550/arxiv.1910.12212

Cited by 2 publications

(2 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dimensions of the setup are detailed in Table I. This system exhibits highly nonlinear behavior [31] and is often plagued by unidentified load disturbances and unknown interactions with the environment. To achieve an adequate control under these conditions, typically some form of PID controller is used.…”

Section: Resultsmentioning

confidence: 99%

Adaptive control of a mechatronic system using constrained residual reinforcement learning

Staessens¹,

Lefebvre²,

Crevecoeur³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

We propose a simple, practical and intuitive approach to improve the performance of a conventional controller in uncertain environments using deep reinforcement learning while maintaining safe operation. Our approach is motivated by the observation that conventional controllers in industrial motion control value robustness over adaptivity to deal with different operating conditions and are suboptimal as a consequence. Reinforcement learning on the other hand can optimize a control signal directly from input-output data and thus adapt to operational conditions, but lacks safety guarantees, impeding its use in industrial environments. To realize adaptive control using reinforcement learning in such conditions, we follow a residual learning methodology, where a reinforcement learning algorithm learns corrective adaptations to a base controller's output to increase optimality. We investigate how constraining the residual agent's actions enables to leverage the base controller's robustness to guarantee safe operation. We detail the algorithmic design and propose to constrain the residual actions relative to the base controller to increase the method's robustness. Building on Lyapunov stability theory, we prove stability for a broad class of mechatronic closed-loop systems. We validate our method experimentally on a slider-crank setup and investigate how the constraints affect the safety during learning and optimality after convergence.

show abstract

Section: Resultsmentioning

confidence: 99%

Adaptive control of a mechatronic system using constrained residual reinforcement learning

Staessens¹,

Lefebvre²,

Crevecoeur³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…where the output of physics model is corrected by machine learning (so-called residual physics). Some studies consider more complex combinations of f P and f A , for example, S = solve[f P,2 (f A (f P,1 )) = 0] (Raissi, 2018;De Groote et al, 2019;Heiden et al, 2020;Kaltenbach and Koutsourelakis, 2021). A trickier case appears in Jiang et al (2018), where discrete state of contact dynamics is first determined by a data-driven classifier, which is then used for choosing one of physics models (also including learnable ones) to be used.…”

Section: Architecture Examples From Literaturementioning

confidence: 99%

Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modeling

Takeishi

Kalousis

2021

Preprint

View full text Add to dashboard Cite

Integrating physics models within machine learning holds considerable promise toward learning robust models with improved interpretability and abilities to extrapolate. In this work, we focus on the integration of incomplete physics models into deep generative models, variational autoencoders (VAEs) in particular. A key technical challenge is to strike a balance between the incomplete physics model and the learned components (i.e., neural nets) of the complete model, in order to ensure that the physics part is used in a meaningful manner. To this end, we propose a VAE architecture in which a part of the latent space is grounded by physics. We couple it with a set of regularizers that control the effect of the learned components and preserve the semantics of the physics-based latent variables as intended. We not only demonstrate generative performance improvements over a set of synthetic and real-world datasets, but we also show that we learn robust models that can consistently extrapolate beyond the training distribution in a meaningful manner. Moreover, we show that we can control the generative process in an interpretable manner.

show abstract

Neural Network Augmented Physics Models for Systems with Partially Unknown Dynamics: Application to Slider-Crank Mechanism

Cited by 2 publications

References 28 publications

Adaptive control of a mechatronic system using constrained residual reinforcement learning

Adaptive control of a mechatronic system using constrained residual reinforcement learning

Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modeling

Contact Info

Product

Resources

About