Emergence of Bipedal locomotion with Dynamically-Rearranging Neural Networks.
|
Accession number;02A0585400
|
| Title;Emergence of Bipedal locomotion with Dynamically-Rearranging Neural Networks. |
| Author;
FUJII AKINOBU
(Nagoya Univ., Graduate School of Engineering, JPN)
ISHIGURO AKIO
(Nagoya Univ., Graduate School of Engineering, JPN)
AOKI TAKESHI
(Nagoya Munic. Ind. Res. Inst.)
EGGENBERGER P
(Atr)
|
Journal Title;Nippon Kikai Gakkai Robotikusu, Mekatoronikusu Koenkai Koen Ronbunshu
|
Journal Code:L0318A
|
ISSN:
|
|
VOL.2001;NO.Pt.3;PAGE.2A1.E4(1)-2A1.E4(2)(2001)
|
| Figure&Table&Reference;FIG.4, REF.2 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;Since there exists highly complicated interaction dynamics, it is in general extremely difficult to design controllers for legged robots. So far various methods have been proposed with the concept of neural circuits so-called Central Pattern Generators(CPG). In contrast to these approaches in this article we use a polymorphic neural circuit instead, allowing to dynamically changing its properties according to the current situation in real time. To do so, we introduce the concept of neuromodulation with a diffusion-reaction mechanism of neuromodulators. As there exists no theory about how such dynamic neural networks can be created, the evolutionary approach is the method of choice to explore the interaction among the neuromodulators, receptors, synapses and neurons. We apply this neural network to the control of a 3-D biped robot which is intrinsically unstable. In this article, we show some simulation results and provide some interesting points derived from the obtained results. (author abst.) |
|
|
|
Related Articles;
|
|
