The VIPER Front Page
Visual Position Estimation:
Estimating Position from Outdoor Images
Carlos Ernesto Guestrin
The Basic Idea
Viper is a "smart" teleoperation interface which analyzes
images sent by a mobile robot in space missions and helps the human teleoperator
on Earth. Teleoperation of mobile robots is a difficult and stressing task;
it is well-known that remote drivers get lost easily, despite having maps
and visible landmarks. Our goal is to reduce the cognitive load on teleoperators
by providing cues that help prevent them getting lost and disoriented.
The basic idea, illustrated by the figure below, follows these steps:
- the system receives the images from the rover;
- features extracted from the images and from a digital elevation map
of the region are matched to estimate the rover's position;
- visual cues are overlayed on the images and on the map to aid the user
on teleoperation tasks.
We call the system VIPER, for VIsual
More information about VIPER
- Detailed information on the current implementation
is available (still in construction!).
- Papers related to the system:
and Pose Estimation for Teleoperation of Lunar Rovers,
is the most recent paper about the system, presented at the
International Conference on Robotics
and Automation, New Mexico, 1997.
Estimation from Outdoor Visual Landmarks for Teleoperation of
is a concise, and older, description of VIPER and our earlier
developments (paper in the
Third IEEE Workshop on Applications of Computer Vision).
- Depth from
describes some ideas developed when working with VIPER,
in this case the possibility of recovering depth from
scattering in outdoor environments (paper in the
IEEE Conference on Computer Vision
and Pattern Recognition, Puerto Rico, 1997.
and images that we have used to test the system.
We have not been able to find much information online about outdoor
position estimation, but if you find something, please
let us know. There are many excellent printed papers about outdoor
localization, about positioning and navigation for space rovers, etc. A
(very) brief sample is given in the
our earlier system description.
- Professor Thompson
and his group at University of Utah
have done a great job at analyzing the process by which humans attain localization
"from views of outdoor environments and maps representing topographic
information". Their data, and related papers, are available.
- To test any outdoor localization system, chances are that you will
need Earth topographic maps. USGS
(United States Geological Survey) is the source for this information
in the United States.
This work has been conducted at the Robotics Institute at the
School of Computer Science, Carnegie Mellon University. It has been partially
funded by NASA; Fabio Cozman has a scholarship from CNPq (Brazil). We thank
these four organizations for all their support.