Long Short-Term Memory for Improved Transients in Neural Network Adaptive Control

In this study, we propose a novel adaptive control architecture, which provides dramatically better performance compared to conventional methods. What makes this architecture unique is the synergistic employment of a traditional, Adaptive Neural Network (ANN) controller and a Long Short-Term Memory (LSTM) network. LSTM structures, unlike the standard feed-forward neural networks, take advantage of the dependencies in an input sequence, which helps predict the evolution of an uncertainty. Through a training method we introduced, the LSTM network learns to compensate for the deficiencies of the ANN controller. This substantially improves the transient response by allowing the controller to quickly react to unexpected events. Through careful simulation studies, we demonstrate that this architecture can improve the estimation accuracy on a diverse set of unseen uncertainties. We also provide an analysis of the contributions of the ANN controller and LSTM network, identifying their individual roles in compensating low and high frequency error dynamics. This analysis provides insight into why and how the LSTM augmentation improves the system's transient response.