Proceedings of the Fuzzy System Symposium
27th Fuzzy System Symposium
Session ID : MD2-2
Conference information

Humanoid motion generation with fuzzy state knowledge
*Takaki OkuyamaNobuhiko KawashiroJunji Nishino
Author information
CONFERENCE PROCEEDINGS FREE ACCESS

Details
Abstract
We aim to generate robot's motion intuitively without solving complex mathematical formulas. We recognize the motion of humanoid robots as sequence of transition from one state to another state. We built a system OSC:Observation based State Creator. When we input 2 robot's state to the OSC, the system generates one transitive motion. Generating robot's motions is important toward a specific goal like RoboCup soccer. In this system, we can express state knowledge by multidimensional fuzzy subsets in configuration space, according to person's state knowledge called state label like "stand", "kick-state" and so on. By using these knowledge as system input, we get robot's motion intuitively.
Content from these authors
© 2011 Japan Society for Fuzzy Theory and Intelligent Informatics
Previous article Next article
feedback
Top