D. Schulz, M. Moors, W. Burgard, and A.B. Cremers
A Statistical Approach to Tracking Multiple Moving People with a Mobile Robot and its Application to Improved Tele-Presence.
In Proc. of the VDI-Conference Robotik 2002 (Robotik),
2002.
Abstract
Whenever mobile robots are installed in populated environments they
need appropriate techniques to determine the positions of the humans
in its surrounding. This information can be useful for a variety of
tasks such as navigation or interaction. In this paper we present
sample-based joint probabilistic data association filters as a robust
technique to estimate the positions of persons in the vicinity of a
mobile robot. Our technique uses the robot's sensors and an
appropriate motion model of walking people. We also present an
application of this approach to the 3D graphics visualization of a
tele-operation interface for a mobile robot designed to establish
tele-presence over the internet. The 3D visualization provides
real-time visual feedback of the robot's actions and of the people in
the robot's surrounding.
Download
Full paper [.pdf]
Bibtex
@InProceedings{schulz02Statistical,
author = {Schulz, D. and Moors,
M. and Burgard, W. and Cremers, A.B.},
title = {A Statistical Approach to Tracking Multiple Moving People with a Mobile Robot and its Application to Improved Tele-Presence},
booktitle = {Proc.~of the VDI-Conference Robotik 2002 (Robotik)},
year = {2002}
}