Trajectory tracking controller design using neural networks for a tiltrotor unmanned aerial vehicle

Abstract

This article presents a design of trajectory tracking controller for a tiltrotor unmanned aerial vehicle (UAV). The objective of this design is to control the tiltrotor aircraft with one fixed controller architecture for all flight modes, viz. helicopter mode, conversion mode, and aeroplane mode. The controller is implemented using two time-scale separations, and thus comprises an inner loop for attitude control and an outer loop for trajectory control. The dynamic model inversion (DMI) technique is applied to both the control loops such that the inner-loop and the outer-loop DMIs are driven using the angular equations of motion and the translational equations of motion, respectively. In addition, online adaptive neural networks are employed to compensate for the model inversion errors of the inner loop and the outer loop due to the lack of complete knowledge of tiltrotor aircraft dynamics. The pseudo-control hedging algorithm is adopted to reduce the instability problem caused by the time-delay of control loops, saturation, and rate limit of actuators. The trajectory tracking controller is evaluated using a sophisticated six-degree-of-freedom non-linear simulation programme with an approach and landing scenario including all flight modes of a smart UAV, a tiltrotor UAV.

A Multibody Model of Tilt-Rotor Aircraft Based on Kane’s Method

“…In this section, we show how the multibody model can be used for the control of Euler angles and nacelle inclination angle of the XV-15. In a common study of tilt-rotor aircraft attitude control, the control of the nacelle tilting is not considered [9][10][11][12]. Using the multibody model, the dynamic process of the nacelle inclination angle is described, and the control algorithm of the nacelle inclination angle is presented.…”

“…These control algorithms were verified on a GTRS model. Kim et al designed the trajectory tracking controller for a tilt-rotor unmanned aerial vehicle based on a neural network-augmented model inversion control method [12]. In the above research on the flight control of the tilt-rotor aircraft, active control for the nacelle tilting is not found.…”

Design of a Control System for an Organic Flight Array Based on a Neural Network Controller

“…Applying a linear controller to such nonlinear engineering models is not easy because most engineering models have nonlinearities. The dynamic equation can be expressed as follows if the state variables of the dynamic model are measurable [12]:…”

“…The dynamic model of the OFA in this process was defined based on the single ducted-fan vehicle [4]. A dynamic model of the OFA was then transformed into the affine system form to apply the dynamic model inversion technique [12]. Subsequently, the attitude control system was designed through a PD controller based on the affine system [13].…”

Multibody simulation of a generalized predictive controller for tiltrotor active aeroelastic control

“…In the same years, the application of adaptive control techniques to flight control of tiltrotor aircraft was investigated (e.g. neural network-based feedback linearization, Rysdyc and Calise [5,6]); this is an active research field [7], in view of the innumerable promising applications of vertical take-off and landingcapable high-performance unmanned vehicles. Mueller et al [8] investigated the application of a simple single-input single-output control, compared to more sophisticated linear quadratic Gaussian and GPC control strategies, to the multibody model of a semi span tiltrotor configuration.…”