TY - JOUR
T1 - Adaptive Tracking Control of State Constraint Systems Based on Differential Neural Networks
T2 - A Barrier Lyapunov Function Approach
AU - Fuentes-Aguilar, Rita Q.
AU - Chairez, Isaac
N1 - Publisher Copyright:
© 2012 IEEE.
PY - 2020/12
Y1 - 2020/12
N2 - The aim of this article is to investigate the trajectory tracking problem of systems with uncertain models and state restrictions using differential neural networks (DNNs). The adaptive control design considers the design of a nonparametric identifier based on a class of continuous artificial neural networks (ANNs). The design of adaptive controllers used the estimated weights on the identifier structure yielding a compensating structure and a linear correction element on the tracking error. The stability of both the identification and tracking errors, considering the DNN, uses a barrier Lyapunov function (BLF) that grow to infinity whenever its arguments approach some finite limits for the state satisfying some predefined ellipsoid bounds. The analysis guarantees the semi-globally uniformly ultimately bounded (SGUUB) solution for the tracking error, which implies the achievement of an invariant set. The suggested controller produces closed-loop bounded signals. This article also presents the comparison between the tracking states forced by the adaptive controller estimated with the DNN based on BLF and quadratic Lyapunov functions as well. The effectiveness of the proposal is demonstrated with a numerical example and an implementation in a real plant (mass-spring system). This comparison confirmed the superiority of the suggested controller based on the BLF using the estimates of the upper bounds for the system states.
AB - The aim of this article is to investigate the trajectory tracking problem of systems with uncertain models and state restrictions using differential neural networks (DNNs). The adaptive control design considers the design of a nonparametric identifier based on a class of continuous artificial neural networks (ANNs). The design of adaptive controllers used the estimated weights on the identifier structure yielding a compensating structure and a linear correction element on the tracking error. The stability of both the identification and tracking errors, considering the DNN, uses a barrier Lyapunov function (BLF) that grow to infinity whenever its arguments approach some finite limits for the state satisfying some predefined ellipsoid bounds. The analysis guarantees the semi-globally uniformly ultimately bounded (SGUUB) solution for the tracking error, which implies the achievement of an invariant set. The suggested controller produces closed-loop bounded signals. This article also presents the comparison between the tracking states forced by the adaptive controller estimated with the DNN based on BLF and quadratic Lyapunov functions as well. The effectiveness of the proposal is demonstrated with a numerical example and an implementation in a real plant (mass-spring system). This comparison confirmed the superiority of the suggested controller based on the BLF using the estimates of the upper bounds for the system states.
KW - Barrier Lyapunov functions (BLFs)
KW - differential neural network (DNNs)
KW - neuro-identification
KW - state constraint systems
KW - trajectory tracking
UR - http://www.scopus.com/inward/record.url?scp=85079871174&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2020.2966914
DO - 10.1109/TNNLS.2020.2966914
M3 - Artículo
C2 - 32078564
SN - 2162-237X
VL - 31
SP - 5390
EP - 5401
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 12
M1 - 9000797
ER -