Sensors Mag

Limping Robots

November 20, 2006 By: Melanie Martella, Sensors


E-mail Melanie Martella

You know, normally if we see something limping (or are limping ourselves), that's a bad thing. But in a recent development in the robotics world, a limping robot represents a significant advance.

When Bad Gaits are Good

Robots are complicated beasts, even the simple bug-like ones. Technologies that help robots navigate in known environments exist, but the more interesting problem is to help robots navigate and move in chaotic surroundings. Why? Because we'd really like to use robots in the great outdoors and getting robots to navigate outside is extremely difficult.

Researchers from Cornell have an article in Science about a robot with the ability to sense if it is injured and then compensate for the injury to keep moving. You can read more about it in the Seattle Times or here at the PhysOrg.com Web site.

Self-Awareness Robot Style

Essentially, the Cornell researchers (Hod Lipson, Viktor Zykov, and Josh Bongard) added a bunch of sensors and actuators throughout a starfish-shaped robot and told it to move. The robot knows what its parts are, but not how they're arranged and not how best to achieve its goal. The breakthrough comes in how the robot seeks to solve the "move forward" command. It generates a number of computer models of how it might be put together and then tests them, sending commands to its motors. Based on the results of those tests, the robot discards or amends its models. The result is that if you damage one of its arms, the robot can adapt to this change in its abilities.

To quote Hod Lipson from the PhysOrg.com article, "Most robots have a fixed model laboriously designed by human engineers. We showed, for the first time, how the model can emerge within the robot. It makes robots adaptive at a new level, because they can be given a task without requiring a model. It opens the door to a new level of machine cognition and sheds light on the age-old question of machine consciousness, which is all about internal models."

I think it's disingenuous to describe this as a robot with a sense of self—it's still a mechanical device, it's just been given more flexible programming. Do you have any pet peeves about science journalism, because anthropomorphization of devices is definitely one of mine. Scroll down to the bottom of the page and let me know what you think.


Add Comment




IIoT University


Deep Learning for Vision Using CNNs and Caffe: A Hands-on Tutorial – 9/22/16 – Cambridge, Mass


IDE






Sensors 2017 Call for Speakers


Sensors Midwest


Advertise


Subscribe



Twitter Feed

Find It Fix It Forum

Sensors invites you to join the Findit-Fixit Forum, where you can get answers to your sensing questions—concerning technologies, products, methods, applications, and services--and also offer help to your fellow engineers. The Forum covers all kinds of topics, from the basics to the extraordinary.

Join the discussion!


© Copyright 2016 Questex, LLC. All Rights Reserved. Sensorsmag. Privacy Policy | Terms of Use

If you are having technical difficulties or considerations, please contact the webmaster.