acm-header
Sign In

Communications of the ACM

ACM TechNews

New System Allows Robots to Continuously Map Their Environment


View as: Print Mobile App Share:
PR2 robot

The MIT researchers used at PR2 robot, developed by Willow Garage, with a Microsoft's Kinect sensor to test their system.

Credit: Hordur Johannsson

Massachusetts Institute of Technology (MIT) researchers have developed a system that enables robots to create and continuously update a three-dimensional (3D) map of their environment using a low-cost camera.

MIT's Maurice Fallon says the technique, based on simultaneous localization and mapping technology, enables robots to constantly update a map as they learn new information over time. As a robot travels through an unexplored environment, the camera scans the surroundings, building a 3D model of the walls of the room and the objects within it. When the robot passes through the same area again, the system compares the scene with the previous images it has taken until it finds a match.

The system estimates the robot's motion using on-board sensors that measure the distance the wheels have rotated. Fallon notes that by combining the visual information with the motion data, the system can eliminate errors that might occur if the system relied on the sensors alone.

The researchers tested the system on a robotic wheelchair and a portable sensor suite worn by a human volunteer. The researchers found the system could locate itself within a 3D map of its surroundings while traveling at up to 1.5 meters a second.

From MIT News 
View Full Article

Abstracts Copyright © 2012 Information Inc. External Link, Bethesda, Maryland, USA 


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account