Under-Used Statistical Tool Leads to Better Robotic Eyes

Under-Used Statistical Tool Leads to Better Robotic Eyes
Oct-10-13
An algorithm based on a neglected statistical construct has brought researchers a step closer to creating robotic eyes able to determine the orientation of an object.

Graduate student Jared Glover developed the new eyes using a statistical construct known as the Bingham distribution. The Bingham distribution is based on probabilistically, making it particularly useful in situations where information is unreliable or incomplete.

In tests, Glover’s algorithm based on the Bingham distribution offered an improvement of more than 50 percent over conventional alternatives. Robots using the system identified single objects correctly 84 percent of the time and stacked objects, which are more difficult to recognize, 73 percent of the time.

Under-Used Statistical Tool Leads to Better Robotic Eyes


More Info about this Invention:

[DVICE.COM]
[MIT]
Share on      

Add Comment

Comments


Please include me on all news...
Posted by Kam Leong on October 14, 2013

Add your Comment:

[LOGIN FIRST] if you're already a member.

fields are required.



Note: Your name will appear at the bottom of your comment.