| Project/Area Number |
15360130
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent mechanics/Mechanical systems
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
TSUBOUCHI Takashi University of Tsukuba, Graduate School of Systems and Information Engineering, Associate Professor, 大学院・システム情報工学研究科, 助教授 (80192649)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YUTA Shin'ichi University of Tsukuba, Vise-President, 副学長 (00092502)
OHYA Akihisa University of Tsukuba, Graduate School of Systems and Information Engineering, Associate Professor, 大学院・システム情報工学研究科, 助教授 (30241798)
OHNO Kazunori Tohoku University, Graduate School of Information Sciences, Assistant Professor, 大学院・情報科学研究科, 助 (70379486)
重松 文治 五洋建設, 技術研究所, 課長
|
|---|
| Project Period (FY) |
2003 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥15,300,000 (Direct Cost: ¥15,300,000)
Fiscal Year 2005: ¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2004: ¥4,000,000 (Direct Cost: ¥4,000,000)
Fiscal Year 2003: ¥8,300,000 (Direct Cost: ¥8,300,000)
|
|---|
| Keywords | Satellite positioning / Intelligent Robotics / SLAM / Outdoor navigation / Mobile Robot / Self-position identification / 自己位置計測 / スキャンマッチング / GPS / 自律走行 / DGPS / PTK-GPS / 屋外走行 / RTK-GPS |
|---|
| Research Abstract |
This research has revealed two facts at first as follows : 1.DGPS receiver outputs measured position data continually, however, 2.such measured position occasional ly contains large error suffering from multi path or scattering problems of radio wave from the satellites. These problems arise from roadside trees, buildings or architectural structures. Because of the second fact, it is necessary to discriminate reliable DGPS measured positions from the erroneous measurements. In this research work, a method for the discrimination based on odometry measured position is proposed. Only reliable DGPS measurements will be fused to the odometry measured position to correct cumulative error under Kalman filtering frame work. However, there remains such a problem yet that we suffer from cumulative error problem of odometry which was too much relied on when DGPS receiver continually outputs erroneous positions in a street with roadside trees in line so long.
To cope with this problem, a positioning method is proposed in this research, which utilizes GPS,LRF (Laser Range Finder), map and odometry near architectural structures. The map covers very huge area so that a novel way to build the map accurately and easily is also proposed, where 3 dimensional structure of the environment is reconstructed based on the odometry and RTK-GPS position measurements and LRF range data. After the reconstruction, a human operator specifies landmarks manual ly by using a GUI tool to present the reconstructed structure.
Finally, an experimental mobile vehicle moves along street in the universty campus for several hundreds meters where there were many buildings alongside the street successfully.
|
