TY - JOUR
T1 - Saturated Finite Interval Iterative Learning for Tracking of Dynamic Systems with HNN-Structural Output
AU - Xiong, Wenjun
AU - Ho, Daniel W.C.
AU - Yu, Xinghuo
PY - 2016/7/1
Y1 - 2016/7/1
N2 - This brief investigates the interval iterative learning problem for dynamic systems with hierarchical neural network (HNN)-structural output. The first objective is to design the output of a dynamic system with HNN structure. A sufficient condition is obtained to achieve the interval tracking in a finite interval by applying iterative learning control (ILC). Then, the saturated ILC is considered into the discussed system, and a less conservative criterion is obtained to achieve the tracking in a finite interval using a network structure decomposition technique. Finally, simulation results are given to illustrate the usefulness of the developed criteria.
AB - This brief investigates the interval iterative learning problem for dynamic systems with hierarchical neural network (HNN)-structural output. The first objective is to design the output of a dynamic system with HNN structure. A sufficient condition is obtained to achieve the interval tracking in a finite interval by applying iterative learning control (ILC). Then, the saturated ILC is considered into the discussed system, and a less conservative criterion is obtained to achieve the tracking in a finite interval using a network structure decomposition technique. Finally, simulation results are given to illustrate the usefulness of the developed criteria.
KW - Hierarchical neural network (HNN) output
KW - interval iterative learning
KW - saturated iterative learning
KW - Uncertain nonlinearities.
UR - http://www.scopus.com/inward/record.url?scp=84976420338&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84976420338&origin=recordpage
U2 - 10.1109/TNNLS.2015.2448716
DO - 10.1109/TNNLS.2015.2448716
M3 - RGC 21 - Publication in refereed journal
SN - 2162-237X
VL - 27
SP - 1578
EP - 1584
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 7
M1 - 7160765
ER -