RESUMO
This survey presents various methods of improving the overall estimation quality in the class of extended state observers (ESO), which estimate not only the conventional states of the system, but the acting disturbance as well. This type of observers is crucial in forming the active disturbance rejection control structure (ADRC), where the precision of online perturbation reconstruction and cancellation directly influences the robustness of the closed-loop control system. Various aspects of the observer-based disturbance estimation/rejection loop are covered by this work and divided into three categories, related with observer: structure, tuning, and working conditions. The survey is dedicated to researchers and practitioners who are interested in increasing the efficiency of their ADRC-based governing schemes.
RESUMO
The paper presents an application of a special case of an Active Disturbance Rejection Controller (ADRC) in governing a proper realization of basic limb rehabilitation trainings. The experimental study is performed on a model of a flexible joint manipulator, whose behavior resembles a real robotic rehabilitation device. The multidimensional character of the considered assisting mechanism makes it a nontrivial modeling and control problem. However, by the use of the ADRC approach, the modeling uncertainty in the plant is partially decoupled from the system, which increases the robustness of the whole control framework against both internal and external disturbances.