July 11, 2012
Take a look in the national headlines, and you may see some familiar faces. Two Carnegie Mellon ECE research projects have made a splash in the past few months, in outlets ranging from MIT's Technology Review to CNET|TV.
Associate Professor of ECE Priya Narasimhan and her research team landed in eight press articles in the past few weeks, including MIT's Technology Review, after their AndyVision project was selected for demonstration at Research@Intel 2012 Day in late June. The project, a robotic inventory system, takes the form of an autonomous robot that can scan the aisles and shelves of a retail store to detect various retail issues of interest (e.g., out-of-stock or misplaced items) and generate an interactive store map that customers can use to browse the virtual world of the store on an in-store digital sign. The robot, a joint research initiative with the Carnegie Mellon University Store, also performs a detailed inventory and can tell store employees when an item is running low or if merchandise is out of place. According to Narasimhan, AndyVision accomplishes all of this through the real-time fusion of image-processing and machine-learning algorithms, a database of images showing the store's stock, a basic map of the store's layout, and sensors to help it navigate autonomously and to prevent it from running into things. The project is part of Narasimhan's work as head of the Intel Science and Technology Center for Embedded Computing, housed at Carnegie Mellon. AndyVision team members include Senior Systems Scientist Rajeev Gandhi; ECE post-doctoral researcher Kunal Mankodiya; ECE students Jonathan Francis, Amit Krishnan and Utsav Drolia; Carnegie Mellon Director of Campus Services Ryan Wolfe; and Angela Tumolo Neira, marketing specialist at the Carnegie Mellon University Store.
In addition to the MIT Technology Review story, AndyVision appeared in pieces on LiveScience, El Economista, Mashable, Knovel, Engadget, PC World and IEEE Spectrum. (Click on the links above to check out the stories.)
Meanwhile, on the West Coast, Assistant Research Professor Pei Zhang and his team of graduate students at Carnegie Mellon Silicon Valley played host to CNET|TV, a Web TV network specializing in original programming on tech topics operated by CBS Interactive. As part of its Always On series, reporter Molly Wood interacted with Zhang's SensorFly — a novel, low-cost, controlled-mobile aerial sensor networking platform that weighs in at around 30 grams and is about 2.5 inches long (6.5 inches in diameter). These autonomous helicopter nodes feature communication, ranging and collaborative path capabilities that Zhang says could sense survivors after disasters or enemies in urban combat situations. On the program, which you can see here, Zhang tells Wood about possible applications for the technology, and he and his grad students offer her a demonstration of SensorFly.