Title :
Artificial neuronal group method for data handling
Author :
Hu, Shengfa ; Li, Liang ; Yan, Pingfan
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing, China
Abstract :
In this paper an artificial neuronal group method for data handling (ANGMDH) is proposed. The basic unit of this method, artificial neuronal group or Ivakhnenko polynomial, is implemented by a Sigma-Pi neural network. A simulation for a two joints manipulator hand-eye coordination system shows that this method has the advantage of small samples and high approximating error. It also seldom requires knowledge about parameters of the network.
Keywords :
identification; learning (artificial intelligence); manipulators; neural nets; polynomials; Ivakhnenko polynomial; Sigma-Pi neural network; artificial neuronal group method for data handling; high approximating error; small samples; two joints manipulator hand-eye coordination system; Artificial neural networks; Automation; Control theory; Data handling; Input variables; Multi-layer neural network; Neurons; Nonlinear systems; Polynomials; System identification;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.716977
