Our experiments were conducted in the Wean Hall 6100 and 6200 corridors.
We placed aritificial walls to limit the environments to acyclic GVG strucutures.
We ran one very small 6 node exploration in Wean Hall 6100 and 6200 corridors which succeeded and is referred to as tree3.
The global map result of tree3 is the first figure shown above.
The second run (tree4) was an extension of tree3, and whose result is shown in the second figure above.
Unfortunately, this run failed when the robot failed to detect the strong meetpoint at the end of the 6100 corridor
(southwest tip of diagram) and began tracing back down the corridor. This issue has since been resolved.
In both runs the robot successfully accessed and traced the reduced GVG from meetpoint to meetpoint.
The dotted line shows the path of the robot, while the blue squares represent the sensed objects.
Many obstacles that the robot used to navigate are only stored temporarily in local maps.
The scale is in tenths of inches.
Go back to: H-SLAM Experiments Page | H-SLAM page
Jump to: People | Robots | Research | Events | Projects | Education | Papers | Software | Links
© Copyright 2002 Sensor Based Planning Lab, Carnegie Mellon University. All Rights Reserved.
 |
 |
 |