Adaptive friction compensation using neural network approximations

Abstract

We present a new compensation technique for a friction model, which captures problematic friction effects such as Stribeck effects, hysteresis, stick-slip limit cycling, pre-sliding displacement and rising static friction. The proposed control utilizes a PD control structure and an adaptive estimate of the friction force. Specifically, a radial basis function (RBF) is used to compensate the effects of the unknown nonlinearly occurring Stribeck parameter in the friction model. The main analytical result is a stability theorem for the proposed compensator which can achieve regional stability of the closed-loop system. Furthermore, we show that the transient performance of the resulting adaptive system is analytically quantified. To support the theoretical concepts, we present dynamic simulations for the proposed control scheme.

Friction modelling and compensation for motion control using hybrid neural network models

“…A RBF approach was proposed in Du and Nair (1999) where the center points and variances of the Gaussian functions had to be chosen a priori. A similar RBF based adaptive scheme was also proposed in Huang et al (2000) with the same drawbacks. A reinforcement learning algorithm using functional link models for friction modelling was reported in Kim and Lewis (2000).…”

Adaptive Friction Compensation: Modular Design with Passive Identifier

“…(7), observer (11), and update law (13) has a globally uniformly stable equilibrium at the origin z ¼ 0; ϵ z ¼ 0; kc ¼ 0 , and…”

Precise angular speed control of permanent magnet DC motors in presence of high modeling uncertainties via sliding mode observer-based model reference adaptive algorithm

“…Some examples of these controllers are adaptive controller [25,26], state feedback regulation [27] and robust control [28]. Fuzzy and neural networks algorithms have been also utilized in this field of research to control servo actuators with regards to reducing the demerits of friction by employing estimation techniques [29][30][31]. Along with diverse types of adaptive controllers, model reference adaptive controllers (MRAC) have addressed the problem of friction compensation more effectively [32,25] and can be an appropriate choice for industrial applications.…”