A Novel Data mining method for next-generation robots

Research Project

Project/Area Number	20500181
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Single-year Grants
Section	一般
Research Field	Perception information processing/Intelligent robotics
Research Institution	Wakayama University
Principal Investigator	NAKAMURA Takayuki Wakayama University, システム工学部, 准教授 (50291969)
Co-Investigator(Kenkyū-buntansha)	WADA Toshikazu 和歌山大学, システム工学部, 教授 (00231035)
Project Period (FY)	2008 – 2010
Project Status	Completed (Fiscal Year 2010)
Budget Amount *help	¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000) Fiscal Year 2010: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000) Fiscal Year 2009: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000) Fiscal Year 2008: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Keywords	知能ロボット / データマイニング / Kernel PCA / RBF / 視覚センサ搭載ロボット / isoMDS / NMDS / GPLVM / MDS / FastMap / PivotMDS / Glimmer
Research Abstract	In this study, we challenge development of the novel modeling technique that is essential to a next-generation robot. We assume that our modeling technique has two functions. Function 1 detects the data which are short for modeling, and generates new data necessary for modeling by operating a robot. Function 2 is a function to choose only data necessary for modeling from a large amount of sampled data. When we controlled a 3-DOF manipulator using visual information, we examined the method based on the Kernel PCA method which can compress high-dimensional learning data using a non-linear mapping function to the low-dimensional data.