A Novel Actuator Controller: Delivering a Practical Solution to Realization of Active-Truss-Based Morphing Wings

Abstract

Abstract-A novel actuator controller is proposed in this paper for active-truss-based morphing wings (ATBMWs). An ATBMW is a new type of smart structure capable of smooth and continuous profile change, and has the potential to provide better stealth and aerodynamic performance over airfoils with discrete control surfaces. However, the sophisticated ATBMW framework and large amount of highly interacted actuators make it difficult to obtain the overall rigid-body dynamics of the wing for controller design and inconvenient to tune controllers on board. The focus of this study is thus to solve the aforementioned problems by developing an actuator-level control scheme that does not rely on the wing rigid-body dynamics and on-board tuning. The linear-quadratic-Gaussian (LQG) controller is adopted for actuator trajectory tracking, and a novel unknown-input estimator (UIE) is devised to handle un-modeled dynamics. By integrating the UIE with the LQG algorithm, a new tracking controller with enhanced tolerance to uncertainties is constructed. It is shown in simulations and experiments on an ATBMW prototype that the proposed UIE-integrated LQG controller can be designed simply using the known actuator dynamics without on-board tuning, and superior trajectory tracking of actuators was observed despite the presence of un-modeled dynamics and exogenous disturbances.

Parametric Analysis for Robust Force/Torque Tracking Control of a Virtual Stiffness-Damping System in Aeroelastic Instability Testing

“…The analysis in this paper implies that K d can also be used as a primary design parameter instead of being merely designated for finetune. This finding corrects the guideline given in [19].…”

“…Although stability robustness returns to an acceptable level with much larger L, L within this high-value range is nevertheless undesired in consideration of weak rejection of unknown inputs and increased computation load. This finding corrects the guideline provided in [19] which does not restrict the design of L. With regard to the effects of the LPF order, it is easy to see from Fig. 8 that a first-order LPF can be a better choice for relatively smaller sensitivity, with stability robustness property similar to LPFs of higher orders.…”

“…Hence, it is a force/torque tracking problem, where the reference trajectories F s (q,q) change in real time with respect to q andq in accordance with (2). To deal with transmission powerloss, the linear-quadratic-Gaussian (LQG) tracking control enhanced by unknown-input estimation (UIE) as in [19] is employed for the VSDS prototype, given the existence of an equivalent system as in (5). The controller has a structure illustrated in Fig.…”

“…In the interests of better estimation and compensation of transmission power-loss, a first-order LPF can be considered for the relatively smaller sensitivity achieved with the same K d . In [19], K d is recommended only for fine tuning the UIE-LQG controller. The analysis in this paper implies that K d can also be used as a primary design parameter instead of being merely designated for finetune.…”

“…Further improvement is made possible via techniques based on unknown-input estimation (UIE) and cancellation widely studied and applied in industrial applications, featuring enhanced tolerance to un-modeled dynamics and exogenous disturbances [14]- [21]. Therefore, in this study we investigate the linear-quadratic-Gaussian (LQG) tracking control integrated with UIE as in [19] to tackle the problem of transmission power-loss involved in the force/torque tracking of the 2-DOF VSDS for airfoil aeroelasticity testing.…”