| Project/Area Number |
08650287
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Dynamics/Control
|
|---|
| Research Institution | Yokohama National University |
|---|
Principal Investigator |
OKUMA Michio Yokohama National University Faculty of Engineering, Lecturer, 工学部, 講師 (60213667)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TODO Isao Yokohama National University Faculty of Engineering, Professor, 工学部, 教授 (70017888)
|
|---|
| Project Period (FY) |
1996 – 1997
|
| Project Status |
Completed (Fiscal Year 1997)
|
| Budget Amount *help |
¥2,300,000 (Direct Cost: ¥2,300,000)
Fiscal Year 1997: ¥400,000 (Direct Cost: ¥400,000)
Fiscal Year 1996: ¥1,900,000 (Direct Cost: ¥1,900,000)
|
|---|
| Keywords | Robotics / Visual data processing / Environment map generation / High-speed correlation processor for tracking / Vehicle robot / Route generation / Obstacle detetion / Locating position |
|---|
| Research Abstract |
For the purpose of applying robots to uncharted surroundings, a method is proposed for environment map generation by visual data processing through "Tracking Vision" The following results have been obtained :
1. Capability of the method is demonstrated by experimetns to detect and locate obstacles in the surroundings for the map generation. By combining a dynamic visual data processing unit and a vehicle robot, the method has been implemented into an experimental system at comparatively low cost. While scanning the search area, a visual sensor i.e.a CCD camera installed in the robot converts the search area scene into dynamic visual signal. The unit is composed of the Tracking Vision and a workstation. The former processes the sensor's signals and provides the latter with the output, which is analyzed quantitatively to identify obstacles. When the scanning is completed and the robot is informed of existence of the obstacles, the vehicle robot initiates action to locate the positions of the obstacles.
2. Furthermore, another method has been developed in order to generate two kinds of optimal routes, minimizing traveling distance or time employing obstacle-avoidance techniques while robots trek through areas. Additional experiments have been performed with successful results, showing beneficial utility of the environment maps in practical use. During the experiments, the robot safely trekked through an area having distributed obstacles.
|
