A Sparse Neural Network Approach to Model Reference Adaptive Control with Hypersonic Flight Applications

Abstract

Neural network-based model reference adaptive control (MRAC) is an effective architecture used in the flight control community to combat significant uncertainties where the structure of the uncertainty is unknown. In our previous work, a novel adaptive control architecture called sparse neural network (SNN) was developed in order to improve long-term learning and transient performance of flight vehicles with persistent uncertainties which lie in various regions throughout the operating regime. The SNN is designed to operate with small learning rates in order to avoid high-frequency oscillations and utilizes only a small number of active neurons in the adaptive controller in order to reduce the computational burden on the processor. In this paper, we enhance the SNN architecture by developing an innovative adaptive control term which is used to mitigate a control effectiveness matrix. Furthermore, we design a robust control term and a strict dwell time condition in order to ensure stability while switching between segments. We demonstrate the effectiveness of the SNN approach by controlling a sophisticated hypersonic vehicle with flexible body effects.

Improved model reference‐based adaptive nonlinear dynamic inversion for fault‐tolerant flight control

“…The latter adjusts the controller parameters and structure in real-time depending on the former FDD results to ensure that the fault system achieves acceptable dynamics. The typical design methods for AFTCS include the interacting multiple model, [16][17][18] feedback linearization, 3,19,20 model predictive control, 21,22 Eigen structure assignment, 23 Model Reference Adaptive Control (MRAC), [24][25][26][27][28][29] and so on. With the development of the control theory, AFTC techniques have been successfully applied to the aviation system.…”

“…With the development of the control theory, AFTC techniques have been successfully applied to the aviation system. [19][20][21][22][23][24][25][26][27] MRAC, a representative adaptive FTC approach, excels in coping with uncertainties and faults in the system. In contrast to the fixed-gain controller, MRAC has a flexible structure and can adjust the control gain in real-time with disturbance influence changes, thereby weakening the disturbance influence.…”

Asymmetric integral BLF based state‐constrained flight control using NN and DOB

“…To address the model uncertainties, robust control strategies including the sliding mode control 5‐7 and H control 8 are investigated on HFVs. Moreover, adaptive laws and intelligent learning are applied in References 9‐15 to achieve online identification. In References 16‐19, the learning performance is improved by mining the internal information of the HFV dynamics.…”

Investigating Data-Driven Systems as Digital Twins: Numerical Behavior of Ho–Kalman Method for Order Estimation

“…Applications of adaptive control can be found literally everywhere. From domestic applications [11], to engineering [17] and manufacturing [8], it is highly evident that adaptive control is very useful. Indicatively, recently, identification techniqes had been used to model a system response originating from Partial Differential Equations [18] and attempting to control it in an empirical, yet adaptive way.…”