By Andrew Nusca
Posting in Cities
A Lehigh University scientist has built a "smart" robotic wheelchair that can navigate through a crowded city on its own, with help from laser and camera sensors.
Computer science and engineering professor John Spletzer developed the wheelchair to help the disabled, such as the blind or paraplegics, interact better with their environments.
Armed with high-fidelity lasers and maps, the "smart" wheelchair navigates entirely on its own, with no human guidance or remote control. It can avoid stationery objects -- such as parking meters and light poles -- as well as unplanned objects, such as pedestrians and bicyclists.
This is Spletzer's third go at the driverless vehicle. He previously worked with researchers from the University of Pennsylvania and Lockheed Martin to convert a Toyota Prius into a robot equipped with laser and camera sensors to compete in the 2007 DARPA Grand Challenge.
He also worked with engineers from Freedom Sciences to invent the Automated Transport and Retrieval System, which enables wheelchair users to get into and out of vehicles, stow and retrieve their chairs, and drive while sitting in auto seats that meet federal safety regulations.
“Our goal now is to extend the autonomy of the wheelchair so it can navigate completely in an urban setting and take you wherever you need to go," Spletzer said in a statement.
The wheelchair works by combining sensors that detect and recognize objects and a database with maps to place them. With inspiration from Google Street View, Lehigh graduate students Chao Gao and Mike Sands created maps that robots, not humans, can track and respond to.
Nearby South Bethlehem, Pa. served as the test area (at right, a laser image of the area).
The laser used for the wheelchair is a low cost LIDAR ("Light Detection And Ranging") model. Guided by LIDAR and not GPS, the wheelchair was able to traverse a 1-kilometer route and arrive at its destination to within an accuracy of 20 centimeters.
Spletzer's moving to make the wheelchair affordable but autonomous at a distance.
"We want to develop a cost-effective solution for mobile robot localization and mapping, to go from what we have now with the ATRS, which is a low-cost system with low-level autonomy, to a system that is low-cost with high-level autonomy," he said in a statement. "To do that, we need to replace hardware, which is expensive, with software, which is cheap—once you write the code and download it."
Nov 11, 2009
Using LIDAR to cut the ties to GPS is a super idea. Cost of developing these maps is sunk immediately and never to be seen again if they also describe an engine for generating the maps from the Google resources. GPS radios are cheap and power efficient but not reliable enough at the street level on inside buildings. Cutting that string is a great idea. But will dust and urban detrius make LIDAR a problem to maintain; and will the extra sensors and processing power be a problem power drain?