Robust Incremental Nonlinear Dynamic Inversion Control Using Angular Accelerometer Feedback

Abstract

Incremental Nonlinear Dynamic Inversion (INDI) is a robust nonlinear control tech­nique that is an adaptation of nonlinear dynamic inversion (NDI). By assuming time scale separation between fast and slow dynamics, and by using angular acceleration feedback, the inner loop of INDI directly controls angular accelerations with incremental control effector inputs. In practice, most applications rely on derivation and filtering of angular rates to obtain angular accelerations. The added steps in the feedback loop introduce transport delays and thereby reduce the tracking performance of the controller. Recent technolog­ical advancements in angular accelerometer design have made it feasible to use angular accelerometers for direct sensing of the accelerations. Simulations are done using a CS-25 type aircraft model in combination with real-world sensor characteristics and turbulence. The main contribution of this paper is the effects of using angular accelerometers in INDI control. Controller performance in the investigated configuration of angular accelerome­ ters, rate gyros and effector deflection sensors is limited by the high transport delays of the gyros. Using the angular accelerometer signals to obtain the rates, controller tracking performance is improved whilst requiring lower actuator effort. Further contributions are a study on angular accelerometer misalignment showing that INDI control rejects sensor misalignment successfully, and finally a study on the effects of using uncalibrated control effector deflection feedback showing the high sensitivity of INDI control to uncalibrated effector sensors.