𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Feedback control of minimum-time optimal control problems using neural networks

✍ Scribed by C. J. Goh; N. J. Edwards; A. Y. Zomaya

Publisher
John Wiley and Sons
Year
1993
Tongue
English
Weight
925 KB
Volume
14
Category
Article
ISSN
0143-2087

No coin nor oath required. For personal study only.

✦ Synopsis

This paper presents an optimal feedback controller capable of driving a non-linear control system from an arbitrary initial state to a fixed final state in minimum time. The controller is based on a feedforward multilayer neural network trained repeatedly using open-loop optimal control data which densely span the field of extremals of the non-linear system. The effectiveness of the controller is clearly demonstrated by a simulation on a two-link robot manipulator. The effect of sensorlactuator noise and parameter variation is also included to confirm the robustness of the controller. KEY WORDS Optimal feedback control Neural networks Robot manipulator Minimum time

πŸ“œ SIMILAR VOLUMES

An optimal control model of neural netwo
An optimal control model of neural networks for constrained optimization problems
✍ Qiang Song; Robert P. Leland πŸ“‚ Article πŸ“… 1998 πŸ› John Wiley and Sons 🌐 English βš– 67 KB πŸ‘ 2 views

Optimization neural networks have been studied and applied in the literature, and the mechanism of such neural networks has been investigated from the point of view of optimization theory. Yet, no studies have been found by the authors to investigate the mechanism from a control systems' perspective

Optimal feedback control of time-delay s
Optimal feedback control of time-delay systems
✍ Se H. Oh; Rein Luus πŸ“‚ Article πŸ“… 1976 πŸ› American Institute of Chemical Engineers 🌐 English βš– 825 KB

## Abstract The suboptimal control policy obtained from series expansion of time‐delay terms is useful when the delays are small. The second method involving direct search on the constant gain matrix is applicable even for large delays. Both of these proposed methods are computationally simple and