TY - GEN
T1 - Non-Parametric Neuro-Adaptive Control
AU - Verginis, Christos K.
AU - Xu, Zhe
AU - Topcu, Ufuk
N1 - Publisher Copyright:
© 2023 EUCA.
PY - 2023
Y1 - 2023
N2 - We develop a learning-based algorithm for the control of autonomous systems governed by unknown, nonlinear dynamics toward trajectory tracking. Most existing algorithms either assume certain parametric forms for the unknown dynamic terms or resort to unnecessarily large control inputs in order to provide theoretical guarantees. The proposed algorithm addresses these drawbacks by integrating neuralnetwork-based learning with adaptive control. More specifically, the algorithm learns a controller, represented as a neural network, using training data that correspond to a collection of system parameters and reference trajectories. These parameters and trajectories are derived by varying the nominal parameters and the reference trajectory, respectively. It then incorporates this neural network into an online closed-form adaptive control law in such a way that the closed-loop system tracks the reference trajectory. The proposed algorithm does not use any a priori information on the unknown dynamic terms or any approximation schemes. Comparative computer simulations demonstrate the effectiveness of the proposed algorithm.
AB - We develop a learning-based algorithm for the control of autonomous systems governed by unknown, nonlinear dynamics toward trajectory tracking. Most existing algorithms either assume certain parametric forms for the unknown dynamic terms or resort to unnecessarily large control inputs in order to provide theoretical guarantees. The proposed algorithm addresses these drawbacks by integrating neuralnetwork-based learning with adaptive control. More specifically, the algorithm learns a controller, represented as a neural network, using training data that correspond to a collection of system parameters and reference trajectories. These parameters and trajectories are derived by varying the nominal parameters and the reference trajectory, respectively. It then incorporates this neural network into an online closed-form adaptive control law in such a way that the closed-loop system tracks the reference trajectory. The proposed algorithm does not use any a priori information on the unknown dynamic terms or any approximation schemes. Comparative computer simulations demonstrate the effectiveness of the proposed algorithm.
UR - http://www.scopus.com/inward/record.url?scp=85166481805&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85166481805&partnerID=8YFLogxK
U2 - 10.23919/ECC57647.2023.10178288
DO - 10.23919/ECC57647.2023.10178288
M3 - Conference contribution
AN - SCOPUS:85166481805
T3 - 2023 European Control Conference, ECC 2023
BT - 2023 European Control Conference, ECC 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 European Control Conference, ECC 2023
Y2 - 13 June 2023 through 16 June 2023
ER -