Active disturbance rejection control based robust output feedback autopilot design for airbreathing hypersonic vehicles. (March 2018)

• Record Type: Journal Article
• Authors: Tian, Jiayi; Zhang, Shifeng; Zhang, Yinhui; Li, Tong
• Is Part Of: ISA transactions. Volume 74(2018)
• Page Start: 45
• Page End: 59
• Publication Date: 2018-03
• Subjects: Nonlinear uncertain system -- Output feedback control -- Active disturbance rejection control (ADRC) -- Feedback linearization (FL) -- Equivalent input disturbance (EID) -- Airbreathing hypersonic vehicle (AHV); Engineering instruments -- Periodicals; Engineering instruments; Periodicals; Electronic journals; 629.805
• DOI: 10.1016/j.isatra.2018.01.002
• Languages: English
• ISSNs: 0019-0578

Abstract

Abstract: Since motion control plant ( y ( n ) = f ( ⋅ ) + d ) was repeatedly used to exemplify how active disturbance rejection control (ADRC) works when it was proposed, the integral chain system subject to matched disturbances is always regarded as a canonical form and even misconstrued as the only form that ADRC is applicable to. In this paper, a systematic approach is first presented to apply ADRC to a generic nonlinear uncertain system with mismatched disturbances and a robust output feedback autopilot for an airbreathing hypersonic vehicle (AHV) is devised based on that. The key idea is to employ the feedback linearization (FL) and equivalent input disturbance (EID) technique to decouple nonlinear uncertain system into several subsystems in canonical form, thus it would be much easy to directly design classical/improved linear/nonlinear ADRC controller for each subsystem. It is noticed that all disturbances are taken into account when implementing FL rather than just omitting that in previous research, which greatly enhances controllers' robustness against external disturbances. For autopilot design, ADRC strategy enables precise tracking for velocity and altitude reference command in the presence of severe parametric perturbations and atmospheric disturbances only using measurable output information. Bounded-input-bounded-output (BIBO) stable is analyzed for closed-loop system. To illustrate the feasibility and superiority of this novel design, a series of comparative simulations with some prominent and representative methods are carried out on a benchmark longitudinal AHV model. Highlights: A method is proposed to illustrate how to apply ADRC to a wider kind of nonlinear uncertain systems. The nominal control performance can be retained for this method only using measurable output information. The approach provides a simple and general way to identify the real "total disturbance" imposed on controlled output.