Product category:
Robotics, Handling and Storage
News Release from: Labman Automation | Subject: 12m3 gantry robot
Edited by the Engineeringtalk Editorial
Team on 05 August 2004
Honeybee brain drives gantry robot
Labman's latest 12m3 gantry robot is actually controlled by a bee's brain and a camera on the robot arm.
Labman's latest 12m3 gantry robot is actually controlled by a bee's brain and a camera on the robot arm Currently the brain is a large neural network program simulating the vision and reaction components of a bee's brain, but plans are afoot to incorporate live neurons
This article was originally published on Engineeringtalk on 22 May 2001 at 8.00am (UK)
Related stories
Ultimate in belt cleaning technology claimed
Ace Arc has recently been appointed as sole UK distributor for Dirt Whacker belt cleaning equipment, claimed as the ultimate in belt cleaning technology
Automatic belt cleaner for any belt width
The Dirt Whacker belt cleaner from Ace Arc provides automatic operation, with permanent self-compensating adjustment against wear, for any belt width
Berger Lahr servomotors drive the Labman gantry frame up to 3m/s.
The commands from the bee brain are interpreted by Labman software so as to move the camera on the robot arm.
Door interlocks, strobe lights and E-stops provide a safety net (especially if the unit starts to act like a real bee in a glass box).
Various interfaces and cameras allow researchers to build better and better simulations.
The target of the project is to refine how we understand how brains function in the real world.
Starting with simple models the system will allow a "man made bee brain" real vision input in real time.
As the work progresses the gantry robot can be wheeled outside so the cameras can view clouds, wind blown tress and flowers.
Ant vision is also a well researched field and the gantry robot is built so it can track the path of an ant with a camera over a modelled terrain.
Before any neural network is allowed to command the robot it can be tested using a simulation of the vision expected from the given commands.
3D graphical models determine what the camera should actually be seeing.
This is currently how much of the work has been developed but now it is time to let them loose and see if they want to escape.
• Labman Automation: contact details and other news
• Email this article to a colleague
• Register for the free Engineeringtalk email newsletter
• Engineeringtalk Home Page
Search the Pro-Talk network of sites
