Sign In

Communications of the ACM

ACM TechNews

Tool Builds 3-D Map of Disaster Scene

View as: Print Mobile App Share:
Robin Murphy of Texas A&M Department of Computer Science and Engineering

Texas A&M College Station professor Robin Murphy says the most effective rescue robot design is slow moving, painted yellow or orange, and has lights underneath so a victim can see it approaching.

Credit: Carrie Pratt / St. Petersburg Times

Texas A&M University lab's Robin Murphy (pictured) and colleagues have developed RubbleViewer, a tool for modeling a disaster scene that they say is more efficient than drawing by hand. The RubbleViewer program is designed to upload pictures that are taken from small unmanned air vehicles (SUAVs), and use the algorithms of the PhotoSynth panorama-making software to combine the snapshots. Three-dimensional (3D) maps are built by extracting information from data points, which is like placing a blanket over a bunch of needle points, says Maarten van Zomeren, a graduate student at the Delft University of Technology in the Netherlands who helped develop the software. First responders would be able to build a topographic 3D map of an area in about a half an hour.

RubbleViewer would allow the responders to determine the location of possible survivors by clicking on a spot to annotate the map and call up the specific photos. The developers plan to integrate RubbleViewer with SUAVs as well as land-based search-and-rescue robots for an easy-to-use first responders system. The team says the system would be cheaper and more portable than helicopter-mounted lasers.

From Technology Review
View Full Article


No entries found