Gaussian neural network for direct adaptive control of robotic systems

2002 ◽  
Author(s):  
R.T. Mizerek
Keyword(s):  
Neural Network ◽  
Adaptive Control ◽  
Robotic Systems ◽  
Direct Adaptive Control

Direct Adaptive Control of Underwater Vehicles using Neural Networks

2003 ◽  
Vol 9 (5) ◽  
pp. 605-619 ◽  
Author(s):  
Myung-Hyun Kim ◽  
Daniel J. Inman
Keyword(s):  
Neural Network ◽  
Adaptive Control ◽  
Network Architecture ◽  
Explicit Knowledge ◽  
Tracking Error ◽  
Network Control ◽  
Lyapunov Theory ◽  
Direct Adaptive Control ◽  
Neural Network Controller ◽  
Comparison Of The Results

A direct adaptive neural network controller is developed for a model of an underwater vehicle. A radial basis neural network and a multilayer neural network are used in the closed-loop to approximate the nonlinear vehicle dynamics. No prior off-line training phase and no explicit knowledge of the structure of the plant are required, and this scheme exploits the advantages of both neural network control and adaptive control. A control law and a stable on-line adaptive law are derived using the Lyapunov theory, and the convergence of the tracking error to zero and the boundedness of signals are guaranteed. A comparison of the results with different neural network architecture is made, and the performance of the controller is demonstrated by computer simulations.

scholarly journals Direct adaptive control using feedforward neural networks

2003 ◽  
Vol 14 (4) ◽  
pp. 348-358 ◽  
Author(s):  
Daniel Oliveira Cajueiro ◽  
Elder Moreira Hemerly
Keyword(s):  
Neural Network ◽  
Adaptive Control ◽  
Feedforward Neural Networks ◽  
Control Input ◽  
Neural Adaptive Control ◽  
Direct Adaptive Control ◽  
Network Training ◽  
The Neural Network ◽  
Lyapunov’S Second Method ◽  
Kalman Filter Algorithm

This paper proposes a new scheme for direct neural adaptive control that works efficiently employing only one neural network, used for simultaneously identifying and controlling the plant. The idea behind this structure of adaptive control is to compensate the control input obtained by a conventional feedback controller. The neural network training process is carried out by using two different techniques: backpropagation and extended Kalman filter algorithm. Additionally, the convergence of the identification error is investigated by Lyapunov's second method. The performance of the proposed scheme is evaluated via simulations and a real time application.

Sign in / Sign up

Export Citation Format

Share Document