House-cleaning robots will surely be popular, but building one will be a daunting challenge, according to researchers.
Such a robot would need to analyze the types of messes in a house, formulate and execute a plan for room-by-room cleaning, and handle unexpected events.
Maya Cakmak, a professor at the University of Washington who earned a three-year, $400,000 grant from the National Robotics Initiative last year to research cleaning robots, says the task goes beyond getting a machine to hold a tool to some surface. "There's the angle, how much you're pushing and pressure you're applying, how fast you move it, how much you move it, and even the orientation [of the tool] relative to the dirt," she says.
Cakmak is using a technique called "programming by demonstration" to train robots in her lab. The machines learn by mimicking a researcher who shows a cleaning method for the robot's vision system.
Ilker Yildirim, an expert in computational models of cognition and perception at the Massachusetts Institute of Technology, believes machines will need to more fully understand their environment.
Cakmak says designs for homes may need to be more machine-friendly and robots may need to be hackable by end users because every house is different.
From Technology Review
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found