Application of Nonlinear Active Disturbance Rejection Control to One-DOF Electrostatic Actuators

In this paper, a non-linear active disturbance rejection controller (NADRC) is designed and originally applied to a non-linear one-degree-of-freedom electrostatic actuator (ESA). The imperfections of micro-fabrication and micro-packaging result in the modeling uncertainties of ESA. In addition, the ESA is inherently unstable due to pull-in phenomenon. So our control goal is to overcome the pull-in instability of ESA and achieve 99.99% of its full travel range despite of the presence of uncertainty. The NADRC consists of an extended state observer (ESO) and a feedback controller. Two kinds of ESOs are developed in this paper. They are high gain ESO (HG ESO) and the ESO with Fal nonlinearity (FAL ESO). The NADRC is independent of accurate model information, and therefore is a suitable controller for the uncertain ESA. The NADRCs with two different ESOs are simulated on the nonlinear ESA. A comparison study is conducted between both ESOs in terms of control performance and stability. The simulation results demonstrate the controllers with both ESOs reach our control goal successfully. While the NADRC with HG ESO generates larger control effort than the one with FAl ESO, the former is more robust against parameter variations and disturbance than the later.