Abstract
This brief paper emphasizes on the experimental study of a hybrid contact model combining a traditional physical-based contact model and a data-driven error model in order to provide a more accurate description of a contact dynamics phenomenon. The physical-based contact model is employed to describe the known contact physics of a complex contact case, while the data-driven error model, which is an artificial neural network model trained from experimental data using a machine learning technique, is used to represent the inherent unmodeled factors of the contact case. A bouncing ball experiment is designed and performed to validate the model. The hybrid contact model can duplicate experimental results well, which demonstrates the feasibility and accuracy of the presented approach.