Adaptive friction compensation using extended Kalman-Bucy filter friction estimation: a comparative study

This paper presents an adaptive friction compensation method that does not require a structured nonlinear friction model. An extended Kalman-Bucy filter (EKBF) relies on an accurate model of system rigid-body dynamics and measured motion to estimate an external friction torque. The friction estimate is used in a traditional friction compensator to cancel friction effects. The paper motivates and describes estimation method and reports results of friction compensation experiments. In addition, the EKBF compensator is compared with other friction compensation strategies, both model-based and non-model based. Experimental results for each compensator are presented for position tracking with velocity reversals. When motion is dominated by static and Stribeck friction, the EKBF-based compensator consistently provides more than fifty percent RMS tracking error reduction over a well-tuned PID compensator, and equal or superior tracking performance over other compensation methods