- All Journals
- Al-Qadisiyah University
- Basrah University
- Kirkuk University
- Mosul University
- The Iraqi Borad for Medical Specialization
- University of Anbar
- University of Technology
- Al Mustafa University
- Al-Zahraa University for Women
- Alnoor University
- Kunooz University
- Misan University
- Ninevah University
- Tikrit University
- University of Al-Hamdaniya
- University of Kerbala
HYBRID GENETIC ALGORITHM AND PARTICLE SWARM OPTIMIZATION WAVELET NEURAL NETWORK FOR DIRECT CONTROLLER OF ROBOT MANIPULATOR | ||
| AL-TAQANI | ||
| Article 1, Volume 24, Issue 6, 2011, Pages 51-62 | ||
| Author | ||
| Osama Kareem M | ||
| Abstract | ||
| An wavelet neural network collected some properties of neural network and also the properties of wavelet transform which used replaced of sigmoidal function. In this paper used wavelet neural network (WNN) in control of the robot arm where used the first method by direct control which contain two types of neural network one of them wavelet neural and the other adoptive linear wavelet neural which is the best because need for less number of wavelet in the hidden layers. Also training the weights of each of them by using Hybrid Genetic Algorithm and Particle Swarm Optimization (HGAPSO) and used also in the operation of training of the weight of network which controlled on robot arm instead of Gradient Decent (GD) its fast but may in local minimum . The properties of this algorithm are achieve best results where can get on least mean square error between the target signal and the output of the robot arm ,also can achieve best result with minimum number of learning cycles also can be used in complex systems which cannot derivative. . Two types of network and learning method are tested by using simulation and control on Robot arm and tested these method under different load also compared the previous methods. | ||
|
Statistics Article View: 105 PDF Download: 73 |
||