On the Necessity, Scheme, and Basis of the Linear–Nonlinear Switching in Active Disturbance Rejection Control

Active disturbance rejection control (ADRC) was originally given in a nonlinear gain structure to better accommodate the dynamic uncertainties and disturbances. However, the resulting complexity in a theoretical analysis and in the parameter tuning inhibits the applications of an ADRC. ADRC was linearized and parameterized for a practical convenience. Since linear ADRC (LADRC) and nonlinear ADRC (NLADRC) each has its own advantages and disadvantages, choosing between LADRC and NLADRC is rather difficult. As a matter of fact, there is a lack of quantitative analysis and comparison between LADRC and NLADRC. This paper first gives an easy solution in the parameter tuning of the nonlinear extended state observer, followed by a quantitative analysis and comparison study on LADRC and NLADRC; then, an LADRC/NLADRC switching control (SADRC) scheme is proposed and its stability is analyzed; finally, the SADRC scheme is verified by experiment using the ball-beam platform. The proposed SADRC takes the advantage of the additional performance improvement associated with the NLADRC, but make it easier to use.