Intelligent Interface for 3-D Shape Manipulation Using Various-Shaped Elastic Objects as Input Tools
Grant-in-Aid for Scientific Research (B).
|Allocation Type||Single-year Grants|
Intelligent mechanics/Mechanical systems
|Research Institution||The University of Tokyo|
MURAKAMI Tamotsu The University of Tokyo, Graduate School of Engineering, Associate Professor, 大学院・工学系研究科, 助教授 (20212251)
|Project Period (FY)
1998 – 2000
Completed(Fiscal Year 2000)
|Budget Amount *help
¥5,600,000 (Direct Cost : ¥5,600,000)
Fiscal Year 2000 : ¥800,000 (Direct Cost : ¥800,000)
Fiscal Year 1999 : ¥2,200,000 (Direct Cost : ¥2,200,000)
Fiscal Year 1998 : ¥2,600,000 (Direct Cost : ¥2,600,000)
|Keywords||3-D geometric operation / Direct and intuitive input tool / Haptic interface / Operation integration / Human interface / Computer-aided design / Virtual reality / Neural network / 形状モデリング / コンピュータ・グラフィックス / 3次元形状測定|
The purpose of this research is to propose a passive haptic interface for manipulating 3-D geometric model in CAD and CG directly and intuitively by bending and twisting real elastic objects as input tools by bare hands. The results are summarized as follows.
(1) A passive haptic input tool made of elastic and conductive foam rubber lump to measure 3-D deformation on the tool by electrical resistance change is proposed.
(2) 3-D deformation measurement and calculation method applicable to various-shaped foam rubber input tool is developed using neural network learning.
(3) Free-form deformation technique, originally developed for parallelepiped shape, is extended to various shape by introducing a simplified control volume which approximates the object shape with a set of parallelepiped control volumes.
(4) To measure and calculate 3-D deformation on various shaped foam rubber input tool with a limited learning ability of neural network, the number and allocation of electrical terminals embodied to foam rubber lump must be minimized and optimized. By identifying input layer units, which correspond to electrical terminals, of a neural network with less importance during learning process and removing them, a method of minimizing and optimizing the number and allocation of electrical terminals is developed.
(5) An input tool consisting of foam rubber block, electromagnetic six-degree-of-freedom tracker, buttons and slide volumes are proposed to integrate multiple 3-D geometric operations, such as deformation, position/orientation control and parameter controls, for 3-D CAD and CG with a single tool.
(6) By quantitatively comparing the proposed tool and conventional mouse-keyboard interface for necessary time and number of actions for specific task, the effectiveness and possibility of our approach are confirmed.
Research Output (18results)