The main objective was to analyse the effects of rotor dynamics, and in particular lead-lag and flapping dynamics, on incremental backstepping (IBS). The main focus in this research was on the angular rate controller of a rotorcraft, since it required an idealised rotorcraft mode
...
The main objective was to analyse the effects of rotor dynamics, and in particular lead-lag and flapping dynamics, on incremental backstepping (IBS). The main focus in this research was on the angular rate controller of a rotorcraft, since it required an idealised rotorcraft model to establish the IBS controller. The idealised rotorcraft model is established by means of the method of residualised dynamics, which assumes steady-state condition of the rotor dynamics. The idealised rotorcraft model would assure that the time-scale separation (TSS) condition is not, or at least less-likely, violated. However, the use of an idealised rotorcraft model for establishing the IBS control law would require a countermeasure when there exists a significant difference between the idealised and actual rotorcraft model. The problem lies therein that actuator and angular acceleration measurements are required for the IBS control law of the angular rate subsystem, but do come from the idealised and actual rotorcraft model respectively. This causes synchronisation issues. Based on findings it can be stated, that accounting for flapping dynamics when establishing the idealised rotorcraft model, is a requisite. However, accounting for lead-lag dynamics when establishing the idealised rotorcraft model, can be considered redundant, as it does not yield a significant difference with respect to the idealised model based upon steady-state flapping dynamics. In other words, an idealised rotorcraft model based upon steady-state flap-lag dynamics is almost the same as an idealised rotorcraft model based upon steady-state flap dynamics wherein considering the natural modes of motion and open-loop frequency response. From this it followed that the controller is robust to uncertainties in lead-lag dynamics, because the IBS control law and synchronisation can be established without requiring any knowledge of lead-lag dynamics. This is however, in great contrast with flapping dynamics, because sufficient knowledge is required of flapping dynamics when establishing both synchronisation filter as well as the IBS control law. Especially the control effectiveness matrix in the synchronisation filter, must be well-known, otherwise performance can be significantly harmed. This shows that there is some lack in controller robustness, when establishing the IBS controller based upon an idealised rotorcraft model.