3D Ball tracking in a Humanoid Robot using Depth Map

Title
3D Ball tracking in a Humanoid Robot using Depth Map
Authors
김강건박가람유범재
Keywords
3D Ball tracking; Humanoid Robot; Extended Kalman Filter; Depth Map
Issue Date
2010-11
Publisher
The 7th International Conference on Ubiquitous Robots and Ambient Intelligence
Citation
, 249-251
Abstract
In this paper, a system to catch a flying ball by using a humanoid robot is proposed. Various researches has been conducted on the 3D ball tracking, however, there are few studies on the humanoid robot catches the ball using images acquired by a sensor mounted on the head of the humanoid robot. This is because there are some limitations on the detection of the ball due to the performance of a mounted camera on the head. The goal of this paper is to present an effective way to overcome these restrictions in the humanoid robot. We are able to get 3D data sets as depth maps using time-of-flight camera. In order to detect the ball, we use difference images. Extended Kalman Filter (EKF) is used to track the flying ball and predict the target position. The air drag is also considered. Measured target 3D position and velocity information are sent to the humanoid robot and this is able to catch the flying ball.
URI
http://pubs.kist.re.kr/handle/201004/38917
Appears in Collections:
KIST Publication > Conference Paper
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE