Â鶹ÒùÔº

EPFL researchers are developing robotic beehive frames that help locate honey stores inside of beehives over time, without relying on cameras. The aim is to develop new observation tools to study honeybee behavior that better fit the bees' natural way to occupy space compared to current methods.

Cyril Monette is fascinated by collective behavior. Having studied how insects like cockroaches make collective decisions, the third year doctoral student in the EPFL's Mobots Laboratory is now turning his attention toward the very useful, and much less repulsive insect: the honeybee. And he's doing so with the help of robotics.

The bees inside a hive naturally cluster into a volume the shape of a ball, at the intersection of the multiple frames of honeycomb inside of the hive. This ball-shaped configuration is likely optimal for the bees, yet observation hives developed by the scientific community study only one or two isolated frames at a time, forcing bees to occupy a flat disk instead.

"The isolated observation hive was built that way to allow for automated observation thanks to cameras or direct observation," explains Monette, who would like to offer a way to study bee behavior in relation to honey stores but in an environment that respects their instincts for clustering in space.

"Unless we provide an environment that allows the bees to behave naturally, the observations yielded can only approximate how intact and strong colonies behave."

Using the thermal properties of honey

To do away with cameras altogether, researchers from Mobots have developed a thermal robotic beehive frame upon which bees can build honeycomb. The robotic frame consists of 64 temperature sensors that can measure the temperature across 10 distinct regions that can be heated up separately, and since the bees build honeycomb on both sides of the frame, that corresponds to 20 honeycomb patches per frame where bees can potentially store honey.

The idea is to first locate where the honey stores are and evaluate the amount of honey present in those 10 regions over time, and then to study bee behavior in relation to that.

"We're interested in studying the relation between bee movement, their lifecycles and correlations with honey location within the hive over time. Our first challenge has been to accurately measure the amount of honey per region, in the absence of living bees, which we successfully achieved thanks to the thermal properties of honey," explains Monette.

Because of honey's particular thermal properties, it heats up and cools down differently compared to empty honeycomb. By sending a pulse of heat at increments of either +1, +3 or +5 degrees Celsius, the researchers are able to characterize how a region of honeycomb full of honey reacts thermally and therefore deduce the amount of honey per region.

Specifically, the volume of honey is modeled according to the heating and cooling times combined with a measure of heating dynamics. These robotic frames can be combined in a hive to map the entirety of a colony's honey stocks.

Other ecological studies

"By combining ethology and robotics we can make unprecedented observations of bees in nearly fully natural conditions, revealing behaviors never observed with this precision. This allows us to challenge and improve existing hypotheses on honeybee behavior and at the same time learn techniques that can help us to protect bees," says Francesco Mondada who leads the Mobots Laboratory.

"Moreover, combining these elements with our expertise in education allows us to share these observations to a broader audience—sparking curiosity about these fascinating, lesser-known behaviors that showcase the complexity of social insects' life."

Monette and colleagues are currently conducting experiments with their robotic observation hive to study bee behavior, testing various hypotheses in situ.

"I'm intrigued by bee ethology, like the way the ball of bees expresses circadian rhythms by expanding and contracting throughout the day," explains Monette. "With our robotic observation hive, I'm hoping to map honey resources and colony demographics over time, observe how the bee ball moves throughout the winter, and provide a framework for other ecological studies such as studying the impact of heatwaves on colonies."