Deep Learning for Control: a non-Reinforcement Learning View

Matei, Ion; Minhas, Raj; Zhenirovskyy, Maksym; Kleer, Johan de; Rai, Rahul

doi:10.23919/acc45564.2020.9147287

Cited by 4 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Care must be taken when learning state dependent control maps for unstable system. If the learning algorithm is not properly initialized, the system instability can induce instability in the learning algorithm as well [30]. In [27], the authors use a hybrid modeling and learning methodology to deal with catching in-flight objects with uneven shapes.…”

Section: State Of the Art In Hybrid Modelingmentioning

confidence: 99%

Hybrid modeling: Applications in real-time diagnosis

Matei¹,

Kleer²,

Feldman³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Reduced-order models that accurately abstract high fidelity models and enable faster simulation is vital for real-time, model-based diagnosis applications. In this paper, we outline a novel hybrid modeling approach that combines machine learning inspired models and physics-based models to generate reducedorder models from high fidelity models. We are using such models for real-time diagnosis applications. Specifically, we have developed machine learning inspired representations to generate reduced order component models that preserve, in part, the physical interpretation of the original high fidelity component models. To ensure the accuracy, scalability and numerical stability of the learning algorithms when training the reducedorder models we use optimization platforms featuring automatic differentiation. Training data is generated by simulating the highfidelity model. We showcase our approach in the context of fault diagnosis of a rail switch system. Three new model abstractions whose complexities are two orders of magnitude smaller than the complexity of the high fidelity model, both in the number of equations and simulation time are shown. The numerical experiments and results demonstrate the efficacy of the proposed hybrid modeling approach.

show abstract

Section: State Of the Art In Hybrid Modelingmentioning

confidence: 99%

Hybrid modeling: Applications in real-time diagnosis

Matei¹,

Kleer²,

Feldman³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the case where the model is represented as an ODE, we obtained good results. For example, in [19] we showed how to learn control policies for an inverted pendulum using a model predictive control approach solved using Pytorch. When dealing with DAEs though, the gradient-based optimization algorithm, when combined with direct collocation methods to approximate time derivatives, tend to converge slowly.…”

Section: Differential Programming For Gradient-based Optimizationmentioning

confidence: 99%

Classification Based Diagnosis:

et al. 2019

View full text Add to dashboard Cite

Machine learning methods based on classifiers are more robust to system complexity, but they ignore the relations that exist in the data due to the physical laws governing the behavior of the system. In this paper we discuss how (partial) knowledge about the physical system can be integrated in the machine learning process. We focus on classification based diagnosis. We show how the partially known model is integrated in the classification algorithm, and how the new algorithm differs from a machine learning classifier based on neural networks. We demonstrate that by integrating the partial system knowledge, the cross-entropy optimization problem used for learning a classifier can be expressed as a set of regression problems in terms of the parameters of the model representing the unknown behavior, followed by a simpler classifier learning. We showcase our approach when diagnosis faults for a rail switch system.

show abstract

Mapping of Standardized State Machines to Utilize Machine Learning Models in Process Control Environments

Grothoff

Kleinert

2021

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Deep Learning for Control: a non-Reinforcement Learning View

Cited by 4 publications

References 4 publications

Hybrid modeling: Applications in real-time diagnosis

Hybrid modeling: Applications in real-time diagnosis

Classification Based Diagnosis:

Mapping of Standardized State Machines to Utilize Machine Learning Models in Process Control Environments

Contact Info

Product

Resources

About