2015 Fiscal Year Final Research Report
Outdoor environment recognition reflecting actual motion experience based on image features
| Project/Area Number |
25330305
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent robotics
|
|---|
| Research Institution | Shizuoka University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
KANEKO Toru 静岡大学, 工学部, 名誉教授 (50293600)
|
|---|
| Research Collaborator |
BEKHTI Mohammed Abdessamad
MATSUMURA Kazuki
MATSUNAGA Sho
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Keywords | 移動ロボット / ナビゲーション / 屋外不整地 / 環境認識 |
|---|
| Outline of Final Research Achievements |
Autonomous robots are required to be able to act in environments which are not structured specifically for them. To meet the requirement, it is important to develop a technology which allows to adapt to unexperienced environment autonomously, instead of relying on human designers’ preparation under assumption or restriction of environment. In this research, an experience-based method for prediction of motion feature obtained by acceleration sensor was developed. The method was implemented on an outdoor autonomous robot and verified through experiment with uneven outdoor terrains. It was confirmed that it was possible to predict motion features based on information obtained by cameras attached to the robot. In addition, a motion planning scheme which allows consideration of velocity-dependent traversability assessment was proposed. We realized a navigation of a mobile robot to reach a destination while appropriately adjusting its velocity so as to avoid excessive accelerations.
|
| Free Research Field |
ロボット工学
|
