Scientists at the Universities of Sheffield and Sussex are working on building accurate computer models of the honey bee brain, in particular the parts that control the sense of smell and vision. With these models, the researchers hope to build an autonomous flying robot that actually acts like a bee instead of just being programmed to carry out a set of instructions. The work would advance the world artificial intelligence as well as give us more information on the brain systems of animals.
The project, called Green Brain, could tackle a first in science -- a robot that can perform complex tasks as well as an animal. The honey bee robot would be used to find the source of certain smells or odors like a honey bee finds flowers. This type of ability would be useful in search and rescue missions, environmental monitoring and even in mechanical pollination of crops for areas where honey bee populations have dwindled.
This project fits in nicely another project we recently covered where scientists were studying the bumblebee's ability to locate a set of flowers and then find the quickest route that visited each one in just a few tries. The flight patterns were modeled and the data could be used to for things like building faster computers, plotting delivery routes and sequencing DNA.
The Green Brain project relies on some advanced hardware capable of building models of something as complex as a brain's cognitive systems donated by NVIDIA Corporation.
University of Sheffield explains:
The hardware provided by NVIDIA is based on high-performance processors called 'GPU accelerators' that generate the 3D graphics on home PCs and games consoles and power some of the world’s highest-performance supercomputers. These accelerators provide a very efficient way of performing the massive calculations needed to simulate a brain using a standard desktop PC – rather than on a large, expensive supercomputing cluster.
"Using NVIDIA's massively parallel GPU accelerators for brain models is an important goal of the project as they allow us to build faster models than ever before," explained Dr Thomas Nowotny, the leader of the Sussex team. "We expect that in many areas of science this technology will eventually replace the classic supercomputers we use today."
Because a honey bee's brain is smaller and more accessible than a larger animal's or a human's, the researchers hope to be able to build a complete and accurate model. If they're successful, it could not only lead to an autonomous honey bee robot, but could also advance other brain modeling projects.