Deep Model Reference Adaptive Control

TL;DR

This paper introduces DMRAC, a deep neural network-based adaptive control architecture that models nonlinearities while ensuring stability, outperforming previous learning-based MRAC methods in controlling complex systems.

Contribution

The paper proposes a novel deep neural network-based MRAC architecture that combines nonlinear modeling with stability guarantees, unifying and extending prior learning-based MRAC approaches.

Findings

DMRAC effectively models significant nonlinearities.

Simulation results show DMRAC subsumes previous methods.

DMRAC achieves high-performance control with stability guarantees.

Abstract

We present a new neuroadaptive architecture: Deep Neural Network based Model Reference Adaptive Control (DMRAC). Our architecture utilizes the power of deep neural network representations for modeling significant nonlinearities while marrying it with the boundedness guarantees that characterize MRAC based controllers. We demonstrate through simulations and analysis that DMRAC can subsume previously studied learning based MRAC methods, such as concurrent learning and GP-MRAC. This makes DMRAC a highly powerful architecture for high-performance control of nonlinear systems with long-term learning properties.