Abstract
In making neural networks learn nonlinear relations effectively, it is desired to have appropriate trainingsets. In the proposed method, after a certain number of iterations, input-output pairs having worseerrors are extracted from the original training set and form a new temporary set. From the followingiteration, the temporary set is applied to the neural networks instead of the original set. In this case,only pairs with worse errors are used for updating the weights until the mean value of errors decreasesto a desired level. Once the learning is conducted using the temporary set, the original set is appliedagain instead of the temporary set. The effectiveness of the proposed approach is demonstrated throughsimulations using kinematic models of a leg module with a serial link structure and an industrial robot.
| Original language | English |
|---|---|
| Title of host publication | Handbook of Research on Biomimetics and Biomedical Robotics |
| Publisher | IGI Global |
| Pages | 205-227 |
| Number of pages | 23 |
| ISBN (Electronic) | 9781522529941 |
| ISBN (Print) | 1522529934, 9781522529934 |
| DOIs | |
| Publication status | Published - Dec 15 2017 |
Fingerprint
ASJC Scopus subject areas
- Engineering(all)
- Computer Science(all)
- Chemical Engineering(all)
- Biochemistry, Genetics and Molecular Biology(all)
Cite this
Neural networks to solve nonlinear inverse kinematic problems. / Nataga, Fusaomi; Habib, Maki K.; Watanabe, Keigo.
Handbook of Research on Biomimetics and Biomedical Robotics. IGI Global, 2017. p. 205-227.Research output: Chapter in Book/Report/Conference proceeding › Chapter
}
TY - CHAP
T1 - Neural networks to solve nonlinear inverse kinematic problems
AU - Nataga, Fusaomi
AU - Habib, Maki K.
AU - Watanabe, Keigo
PY - 2017/12/15
Y1 - 2017/12/15
N2 - In making neural networks learn nonlinear relations effectively, it is desired to have appropriate trainingsets. In the proposed method, after a certain number of iterations, input-output pairs having worseerrors are extracted from the original training set and form a new temporary set. From the followingiteration, the temporary set is applied to the neural networks instead of the original set. In this case,only pairs with worse errors are used for updating the weights until the mean value of errors decreasesto a desired level. Once the learning is conducted using the temporary set, the original set is appliedagain instead of the temporary set. The effectiveness of the proposed approach is demonstrated throughsimulations using kinematic models of a leg module with a serial link structure and an industrial robot.
AB - In making neural networks learn nonlinear relations effectively, it is desired to have appropriate trainingsets. In the proposed method, after a certain number of iterations, input-output pairs having worseerrors are extracted from the original training set and form a new temporary set. From the followingiteration, the temporary set is applied to the neural networks instead of the original set. In this case,only pairs with worse errors are used for updating the weights until the mean value of errors decreasesto a desired level. Once the learning is conducted using the temporary set, the original set is appliedagain instead of the temporary set. The effectiveness of the proposed approach is demonstrated throughsimulations using kinematic models of a leg module with a serial link structure and an industrial robot.
UR - http://www.scopus.com/inward/record.url?scp=85045723249&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85045723249&partnerID=8YFLogxK
U2 - 10.4018/978-1-5225-2993-4.ch009
DO - 10.4018/978-1-5225-2993-4.ch009
M3 - Chapter
AN - SCOPUS:85045723249
SN - 1522529934
SN - 9781522529934
SP - 205
EP - 227
BT - Handbook of Research on Biomimetics and Biomedical Robotics
PB - IGI Global
ER -