Recently, a friend of mine discovered a swarm of bees on his property. He called a local beekeeper, who came to capture the bees and take them to a safe, new hive. While she was working, she talked to us about bees.
When a colony becomes too large for its current hive, or a new queen or queens are hatched—or sometimes for no discernable reason at all—some or all the bees in the colony leave the hive and swarm, sometimes stopping at an interim location, like the bees that showed up in my friend’s front yard. Bees are the most docile while they’re swarming because they are working to protect the queen while they travel, while the scout bees are looking for a new location to live. Apparently, sometimes you can even pet them when they’re in this state, though I declined to try!
This enormous bee swarm, consisting of about 20,000 individuals, was resting in an oak tree. Click photo for more detail.
It was fascinating to watch the bees, crawling around, each with a job, knowing that the queen was in the middle of the swarm, somewhere, being nurtured by all her minions.
So, of course, it got me to thinking that there must be some technological applications as related to bees. I found a few articles that backed me up.
We all know that bee populations have been in decline in the last few years.
Along with a large data organization, The World Bee Project has launched an initiative that allows researchers in the UK to “listen” to honeybees, analyzing the acoustic data captured inside the “smart” hives, including the movement of bees’ wings and feet. Researchers are also able to analyze other precision measurements, including temperature, humidity, and honey yield, which helps them closely monitor bee colonies, detect patterns, and predict behaviors.
The data also helps conservationists and beekeepers to act to protect bee colonies that have been in decline, such as removing predators like the invasive Asian Hornet or preventing swarming at the wrong time of year.
Another way to address the declining bee population is to pollinate flowers by hand. Using a brush to apply the pollen, a person can pollinate five to ten almond trees a day, depending on the size of the trees. Tackling thousands of trees takes major manpower and budget.
This doesn’t necessarily have to be the future of pollination, however. Eijiro Miyako, a researcher at Japan’s National Institute of Advanced Industrial Science and Technology, has designed what he believes could one day be a partial solution: an insect-sized drone capable of artificial pollination. Coated with a patch of horse hair bristles and an ionic liquid gel, these wee robots can collect and transfer pollen from one plant to another.
Image may not be accurate!
Other researchers are skeptical about this new technology being effective if the bees were to die out. “There are 1 million acres of almond trees in California,” says Marla Spivak, a MacArthur Fellow and entomologist at the University of Minnesota. “Every flower needs to be pollinated to set the nut. Two million colonies of bees are trucked in to pollinate the almonds, and each colony has between ten and twenty thousand foragers. How many robots would be needed?”
The next article I found was published by the University of Washington, reporting that researchers at the university have created a sensor package that is small enough to ride aboard a bumblebee. These sensors monitor temperature, humidity, and light intensity—to the bee “backpack”. That way, the bees could collect data and log that information along with their location, and eventually compile information about the health of an entire farm.
“Drones can fly for maybe 10 or 20 minutes before they need to charge again, whereas our bees can collect data for hours,” said senior author Shyam Gollakota, an associate professor in the UW’s Paul G. Allen School of Computer Science & Engineering. “We showed for the first time that it’s possible to actually do all this computation and sensing using insects in lieu of drones.” Because insects can fly on their own, the package requires only a tiny rechargeable battery that could last for seven hours of flight and then charge while the bees are in their hive at night.
“It would be interesting to see if the bees prefer one region of the farm and visit other areas less often,” said co-author Sawyer Fuller, an assistant professor in the UW Department of Mechanical Engineering. “Alternatively, if you want to know what’s happening in a particular area, you could also program the backpack to say: ‘Hey bees, if you visit this location, take a temperature reading.'”
The latest threat to honeybees is the Varroa mite, a parasite that infests hives and sucks the blood from both bees and their young. While it rarely kills a bee outright, it can weaken it and cause young to be born similarly weak or deformed. Over time, this can lead to colony collapse.
And unless you’re looking closely, you might not even see the barely millimeter-sized mites, so infestations often go on for some time without being discovered.
Inspecting a frame in a hive
Beekeepers obviously want to avoid this. The solution has been to put a flat surface beneath a hive and pull it out every few days, inspecting all the waste, dirt and other hive junk for the tiny bodies of the mites. They’re easily missed, but until now, there was no alternative.
Machine learning to the rescue! Machine learning models are excellent at sorting through noisy data, like a surface covered in random tiny shapes and finding targets, like the shape of a dead Varroa mite. Students at the École Polytechnique Fédérale de Lausanne in Switzerland have created an image recognition system trained on images of mites that can sort through a photo, identify, and count any visible mite bodies in seconds. All the beekeeper needs to do is take a regular smartphone photo and upload it to the EPFL system.
So there’s hope, right?