Communications of the ACM
Garbage-Sorting Robot Gets Its Hands Dirty
Robots could fill the role of sorting through and categorizing discarded material from construction and demolition projects for recyclers as a result of the efforts of ZenRobotics researchers.
The researchers have developed a robot that uses visual sensors, metal detectors, weight measurements, and tactile feedback to pick out large pieces of refuse and categorize them. The team has taught the Recycler robot to recognize about a dozen types of material using machine-learning software.
The Recycler, which features a robotic arm, can pull concrete, metal, or wood from a stream of material moving along a conveyor belt. The robot relies on a spectrometer to pick up unique patterns of light when it needs to recognize more ambiguous types of material, such as a piece of plywood with nails driven through it.
ZenRobotics launched a test phase in February, and reports that Recycler currently correctly identifies about half of the construction debris it encounters. The researchers say the robot eventually could be used to sort discarded household material.
From New Scientist
View Full Article - May Require Free Registration
Abstracts Copyright © 2011 Information Inc. 
, Bethesda, Maryland, USA
No entries found