Mobile Operations Performed by Mobile Manipulators on Irregular Terrain - Torque Compensation Using Neural Networks for Disturbance Torques Produced by Irregular Terrain -

1998 ◽  
Vol 10 (5) ◽  
pp. 377-386 ◽  
Author(s):  
Mamoru Minami ◽  
◽  
Masatoshi Hatano ◽  
Toshiyuki Asakura ◽  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Adaptive Control ◽  
Control System ◽  
Horizontal Plane ◽  
Mobile Manipulator ◽  
Mobile Manipulators ◽  
Adaptive Control System ◽  
Control Performance ◽  
Irregular Terrain

In the present study, we propose a control system for mobile operations of mobile manipulators traveling on irregular terrain. Irregularities exist even in structures such as man-made floors of factories and buildings. Since the hand of a mobile manipulator is often required to operate precisely while traveling on irregular terrain and it is subject to disturbance torques caused by traveling on terrain, a method for decreasing control errors caused by disturbances due to terrain must be considered. In the present paper, an adaptive control system including a compensator that uses a neural network, i.e., a neuro adaptive control system, is proposed. In addition, we discuss the control performance of the proposed control system, and show that the control system can decrease control errors occurring on irregular terrain to the levels of errors that occur while traveling on a horizontal plane.

Combination of Neural Networks and State Vector Feedback Adaptive Control (SVFAC) Technique to Control the Gas Turbine Combustor

2006 ◽  
Author(s):  
Mahmood Lahroodi ◽  
A. A. Mozafari
Keyword(s):  
Neural Networks ◽  
Adaptive Control ◽  
Control System ◽  
Gas Turbine ◽  
Control Method ◽  
Control Performance ◽  
Gas Turbine Combustor ◽  
Model Network ◽  
Neural Network Controller ◽  
Nonlinear Plants

Neural networks have been applied very successfully in the identification and control of dynamic systems. When designing a control system to ensure the safe and automatic operation of the gas turbine combustor, it is necessary to be able to predict temperature and pressure levels and outlet flow rate throughout the gas turbine combustor to use them for selection of control parameters. This paper describes a nonlinear SVFAC controller scheme for gas turbine combustor. In order to achieve the satisfied control performance, we have to consider the affection of nonlinear factors contained in controller. The neural network controller learns to produce the input selected by the optimization process. The controller is adaptively trained to force the plant output to track a reference output. Proposed Adaptive control system configuration uses two neural networks: a controller network and a model network. The model network is used to predict the effect of controller changes on plant output, which allows the updating of controller parameters. This paper presents the new adaptive SFVC controller using neural networks with compensation for nonlinear plants. The control performance of designed controller is compared with inverse control method and results have shown that the proposed method has good performance for nonlinear plants such as gas turbine combustor.

Adaptive control of a robot using neural networks

Robotica ◽  
1994 ◽  
Vol 12 (6) ◽  
pp. 553-561 ◽  
Author(s):  
D. T. Pham ◽  
S. J. Oh
Keyword(s):  
Neural Networks ◽  
Adaptive Control ◽  
Control System ◽  
Inverse Dynamics ◽  
Dynamic Interaction ◽  
Variable Load ◽  
Time Varying ◽  
Adaptive Control System ◽  
The Third ◽  
Articulated Robot

SummaryThis paper describes an adaptive control system for an articulated robot with n joints carrying a variable load. The robot is a complex nonlinear time-varying MIMO plant with dynamic interaction between its inputs and outputs. However, the design of the control system is relatively straightforward and does not require any prior knowledge about the plant. This is because the control system is based on using neural networks which can capture the dynamic characteristics of the plant automatically. Three neural networks are employed in total, the first to learn the dynamics of the robot, the second to model its inverse dynamics and the third, a copy of the second neural network, to control the robot.

NEW METHOD OF NEURON DESIGN BASED ON DISCRETE Z-FUNCTION TO ADAPT THE CHANGE OF INTEGRATED VEHICLE STABILITY CONTROL ORDER

2009 ◽  
Vol 08 (03) ◽  
pp. 253-285 ◽  
Author(s):  
M. HARLY ◽  
I. N. SUTANTRA ◽  
H. P. MAURIDHI
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Back Propagation ◽  
Activation Function ◽  
Zero Order ◽  
Vehicle Stability ◽  
Adaptive Control System ◽  
Control Performance ◽  
Fixed Order ◽  
Hidden Layer

Fixed order neural networks (FONN), such as high order neural network (HONN), in which its architecture is developed from zero order of activation function and joint weight, regulates only the number of weight and their value. As a result, this network only produces a fixed order model or control level. These obstacles, which affect preceeding architectures, have been performing finite ability to adapt uncertainty character of real world plant, such as driving dynamics and its desired control performance. This paper introduces a new concept of neural network neuron. In this matter, exploiting discrete z-function builds new neuron activation. Instead of zero order joint weight matrices, the discrete z-function weight matrix will be provided to realize uncertainty or undetermined real word plant and desired adaptive control system that their order has probably been changing. Instead of using bias, an initial condition value is developed. Neural networks using new neurons is called Varied Order Neural Network (VONN). For optimization process, updating order, coefficient and initial value of node activation function uses GA; while updating joint weight, it applies both back propagation (combined LSE-gauss Newton) and NPSO. To estimate the number of hidden layer, constructive back propagation (CBP) was also applied. Thorough simulation was conducted to compare the control performance between FONN and MONN. In order to control, vehicle stability was equipped by electronics stability program (ESP), electronics four wheel steering (4-EWS), and active suspension (AS). 2000, 4000, 6000, 8000 data that are from TODS, a hidden layer, 3 input nodes, 3 output nodes were provided to train and test the network of both the uncertainty model and its adaptive control system. The result of simulation, therefore, shows that stability parameter such as yaw rate error, vehicle side slip error, and rolling angle error produces better performance control in the form of smaller performance index using FDNN than those using MONN.

Research of Model-Free Adaptive Control System Based on Neural Network

2014 ◽  
Vol 608-609 ◽  
pp. 484-488
Author(s):  
Ze Min Liu
Keyword(s):  
Neural Network ◽  
Adaptive Control ◽  
Control System ◽  
Nonlinear Problems ◽  
Adaptive Control System ◽  
Industrial Control ◽  
Bp Network ◽  
Model Free ◽  
The Neural Network ◽  
Positive Effect

With the development of industry, the control system is more and more complex. For the nonlinear problems which can’t be solved by the traditional linear control system used now, it uses the model-free adaptive control system based on the neural network to effectively solve them. In this paper, it firstly makes a detailed analysis on the neural network, describing the neuron, the BP network and the training of neural network; then talks about the model-free adaptive control system, analyzing the structure, characteristics and algorithm of the system; and finally gives the core code of the model-free adaptive control system of the neural network. This paper provides positive effect to the industrial control staff and artificial intelligence researchers.

Sign in / Sign up

Export Citation Format

Share Document