Volume 56, pp. 209-234, 2022.
A hybrid objective function for robustness of artificial neural networks-estimation of parameters in a mechanical system
Jan Sokolowski, Volker Schulz, Hans-Peter Beise, and Udo Schroeder
Abstract
In several studies, hybrid neural networks have proven to be more robust against noisy input data compared to plain data driven neural networks. We consider the task of estimating parameters of a mechanical vehicle model based on acceleration profiles. We introduce a convolutional neural network architecture that given sequential data, is capable to predict the parameters for a family of vehicle models that differ in the unknown parameters. This network is trained with two objective functions. The first one constitutes a more naive approach that assumes that the true parameters are known. The second objective incorporates the knowledge of the underlying dynamics and is therefore considered as hybrid approach. We show that in terms of robustness, the latter outperforms the first objective on unknown noisy input data.
Full Text (PDF) [2.2 MB], BibTeX
Key words
system identification, parameter estimation, convolutional neural networks, sequential data, prediction robustness, mathematical modelling, dynamical systems
AMS subject classifications
68T07, 93B30, 34A30
< Back