An Improved Dynamic Friction Model Using a Data-Based Approach

A significant challenge associated with the development of precision motion control systems is the identification and modeling of friction. In particular, dynamic (presliding) friction is often difficult to accurately model in both the time domain and frequency domain simultaneously. We present a data-based modification to an existing friction model, known as the Dahl Dynamic Hysteresis Model (DHM), which incorporates an empirical friction slope function to provide a more accurate representation of arbitrarily shaped hysteresis curves. This data-based approach avoids the added complexity of identifying or fitting model parameters, and can be implemented with a simple look up table. Simulation results are validated with measured friction data collected from an experimental testbed. We show that the data-based approach significantly improves the friction model accuracy in both the time and frequency domains.