A Hybrid Objective Function for Robustness of Artificial Neural Networks -- Estimation of Parameters in a Mechanical System
This work addresses robustness in neural networks for mechanical system parameter estimation, but it is incremental as it builds on existing hybrid approaches.
The paper tackles parameter estimation for a mechanical vehicle model using acceleration profiles, introducing a convolutional neural network trained with two objective functions, where a hybrid objective incorporating dynamics knowledge outperforms a naive one on noisy data, showing improved robustness.
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 is capable to predict the parameters for a family of vehicle models that differ in the unknown parameters. We introduce a convolutional neural network architecture that given sequential data predicts the parameters of the underlying data's dynamics. 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 noisy input data.