Quantcast
Last updated on April 17, 2014 at 12:17 EDT

Finally, A Robot With The Ingenuity Of MacGyver

October 15, 2012
Image Credit: Georgia Tech

Michael Harper for redOrbit.com — Your Universe Online

MacGyver was an admittedly fictitious character who always knew what to do in a pinch. Quickly assessing his surroundings, the secret agent for the Phoenix Foundation would make use of what was available to him and break free from whatever makeshift cell he was being held in. MacGyver wasn´t a real person, of course, but it´s often the characters portrayed on screen which inspire scientific innovation. After all, would we really have iPads if it weren´t for a plethora of Science Fiction movies, such as 2001: A Space Odyssey?

Researchers from Georgia Tech have called upon the late-80s action-adventure television series for some inspiration as they plan to build a robot able to use its own ingenuity to go into a dangerous situation and rescue any who may be in harm´s way.

Aptly dubbed “MacGyver Bot,” this robot will be able to quickly asses its surroundings and know what items can be used, what items pose further danger and who needs to be rescued.

As the world of robotics progresses, these bots have been asked to perform either the most dangerous or the most monotonous of tasks. While it is much safer for robots to enter into these dangerous situations to perform search and rescue missions, these robots are currently able to perform a set list of tasks and are unable to “think” on their own two feet“¦or wheels.

“Our goal is to develop a robot that behaves like MacGyver, the television character from the 1980s who solved complex problems and escaped dangerous situations by using everyday objects and materials he found at hand,” explained Mike Stilman, the leader of the group to develop MacGyver Bot and an assistant professor in the School of Interactive Computing at Georgia Tech.

“We want to understand the basic cognitive processes that allow humans to take advantage of arbitrary objects in their environments as tools. We will achieve this by designing algorithms for robots that make tasks that are impossible for a robot alone possible for a robot with tools.”

To build this robot, Stilman will have to call upon his knowledge of navigation and robotics. In previous research, Stilman created a system by which robots could autonomously recognize different obstacles and move them out of its way as it navigated itself from one point to the next. By building upon this knowledge, Stilman says MacGyver Bot should be able to not only know how to move obstacles out of its way but also use other obstacles and other elements lying around to complete its task.

“This project is challenging because there is a critical difference between moving objects out of the way and using objects to make a way,” said Stilman in a prepared statement.

“Researchers in the robot motion planning field have traditionally used computerized vision systems to locate objects in a cluttered environment to plan collision-free paths, but these systems have not provided any information about the objects´ functions.”

Building an algorithm, Stilman and team hope to “teach” MacGyver Bot some of the basic mechanics of physics and gravity. For instance, MacGyver Bot should one day be able to understand that a simple piece of pipe could be used as a lever or that a fallen ladder could be used to build a bridge over dangerous debris.

Furthermore, this rescue bot could also learn some basic body mechanics, such as how much force should be used to open a jammed door as well as develop a plan with a series of steps to find its way out of harm.

Stilman says MacGyver Bot will be built based upon Golem Krang, the world´s strongest humanoid robot. To get development moving on MacGyver Bot, the US Navy has offered to fund the research by means of a $900,000 grant from the Office of Naval Research. The Navy hopes this robot will one day work hand in hand with service men and women in search and rescue missions.


Source: Michael Harper for redOrbit.com – Your Universe Online