2007 Fiscal Year Final Research Report Summary
Environment Recognition Method Based on Vision-Haptics-Motion Fusion with a Humanoid Robot
| Project/Area Number |
17300068
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | National Institute of Advanced Industrial Science and Technology |
|---|
Principal Investigator |
YOKOI Kazuhito National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology,Intelligent Systems Research Instit, Research Group Leader (40358304)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NOBORIO Hiroshi Osaka Electro-Communication University, Faculty of Information Science and Arts, Professor (10198616)
YOSHIDA Eiichi National Institute of Advanced Industrial Science and Technology, Intelligent Systems Research Institute, Senior Researcher (30358329)
KITA Nobuyuki National Institute of Advanced Industrial Science and Technology, Intelligent Systems Research Institute, Senior Researcher (90356874)
|
|---|
| Project Period (FY) |
2005 – 2007
|
| Keywords | Intelligent Robotics / Humanoid Robot / Motion Planning |
|---|
| Research Abstract |
The long term target of our humanoid research is to improve the capabilities of humanoid robot in order to perform a larger set of tasks in normal environments. This work is part of the undergoing investigation in integrating vision, haptics, and motion in an efficient way to increase humanoid's autonomy, interactivity and communication. In tins research, we developed several new technologies:
1. A human-inspired method to implement a basic visio-haptic behavior serving in preliminary knowledge acquisition of manipulable objects
2. The first implementation of real-time 3D SLAM for a humanoid robot which is able to demonstrate loop closing
3. A method of generating a reaching motion of a humanoid robot
4. A 3D object recognition method based on spin-images for a humanoid robot having a stereoscopic vision system which is able quickly to compute multi-resolution and interpolating spin-images
5. A visual search behavior of a 3D object in a 3D environment performed by a humanoid robot
6. A method of an object reconstruction for a visual search by a humanoid robot
7. A method to improving a performance of single camera SLAM by using fish-eye view, and
8. A method of generating full body motion for a humanoid robot to achieve camera pose given by a Next Best View algorithm.
The effectiveness of all the methods was confirmed by numerical simulations with OpenHRP simulator and experiments with a real humanoid robot HRP-2.
|
Research Products
(18 results)
