1999 Fiscal Year Final Research Report Summary
Study on a Robot with a World Model
| Project/Area Number |
10650255
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent mechanics/Mechanical systems
|
|---|
| Research Institution | Univ. of Electro-Communications |
|---|
Principal Investigator |
TAKASE Kunikatsu Univ. of Electro-Communications Graduate School of Information Systems, Professor, 大学院・情報システム学研究科, 教授 (10270927)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
OKA Tetsushi Univ. of Electro-Communications Graduate School of Information Systems, Assistant Professor, 大学院・情報システム学研究科, 助手 (00282921)
|
|---|
| Project Period (FY) |
1998 – 1999
|
| Keywords | World Model / Autonomous Robot / Mark-based / 3D vision / Delivery Robot / Geometrical Model / Barcode / Path planning |
|---|
| Research Abstract |
3D Mark-based Vision : 3D Mark-based Vision has been studied, and a new method to use marks with topological features has been proposed to be implemented. Through the recognition of three marks attached to an object by a stereo vision, the position and orientation of the object has been successfully obtained. Furthermore a compact mark has been designed which enables the identification of an object among several thousands objects by integrating barcodes.
Construction of World Model : A method to construct a world model based on the above-mentioned 3D environment recognition has been studied. As a practical method to afford real-time recognition and motion planning, a combination of global model and local model has been proposed. In the scheme, the global model is used for a global planning and sub-goal setting, while the local planning for a frequent path check to the sub-goal. The method is implemented. And effectiveness of it has been shown experimentally.
Wide Area Visual Monitoring System : In order to monitor mobile robots in a wide area, a vision system which has a wide scope has been presented. Infrared LEDs are attached to on the top of robots to detect the robots reliably even in a long distance. Using this method, a goods delivery robots system in a building has been developed. Very robust navigating ability of the system has been verified experimentally.
|
Research Products
(8 results)
