Towards Certifiable Advanced Flight Control Systems, A Sensor Based Backstepping Approach

Abstract

This paper uses singular perturbation theory in order to design a backstepping control system. Earlier applications using singular perturbation theory for control of non-affine systems exist, but here the focus is shifted to uncertain systems. Dealing with uncertainties in this way could be a major benefit for certification of advanced control laws. The reason for this is twofold. First, the backstepping construct, using Lyapunov functions, guarantees the stability of the controlled system. Second, by using measurements of the state derivatives rather than relying on perfect knowledge of the system, a major drawback of backstepping is removed: the need for adaptation to uncertain parameters or unknown model structure. As this method bears resemblance to sensor based nonlinear dynamic inversion, the control law developed in this paper will be referred to as sensor based backstepping. Preliminary results for systems with uncertainty and sensor noise look very promising and future investigation into the effects of time-delays is needed.

Hybrid Sensor-Based Backstepping Control Approach with Its Application to Fault-Tolerant Flight Control

“…Then this control law was developed further by Hovakimyan et al [15] to control a nonaffine nonlinear system. In 2011, Falkena et al [16] combined the singular perturbation theory with the backstepping technique, and developed an incremental type nonlinear backstepping control approach called the sensor based backstepping (SBB) approach. This Lyapunov function based control method can both guarantee the stability of the closed-loop system and avoid the requirement of full aerodynamic model information [16,23].…”

“…The SBB method was utilized to design a controller for the aircraft moment equations, and was also evaluated on an aerodynamic system with uncertainties and measurement noises. Indicated by [15,16], as a result of the backstepping control technique, the system stability can be guaranteed by using Lyapunov functions in this SBB approach. In addition, similar to the incremental NDI flight control scheme, the SBB control approach does not need to adapt to uncertain parameters or unknown model structure, which is essential to most model-based conventional backstepping or NDI control approaches.…”

“…In 2007, Hovakimyan et al [15] proposed an advanced controller for non-affine systems, which involves the singular perturbation theory, Tikhonov's Theorem and a backstepping strategy. Thereafter, Falkena and van Oort et al [16] extended its application and named it as SBB control approach. The SBB method was utilized to design a controller for the aircraft moment equations, and was also evaluated on an aerodynamic system with uncertainties and measurement noises.…”

“…This paper uses the SBB control approach proposed in [16], but the focus is shifted to extending its application to designing a generic attitude controller for a large civil aircraft and handling structural faults for FTFC purposes. In this paper, the SBB control law is utilized to design a body angular rate controller, while NDI control laws are adopted to design an outer loop angular controller.…”