Will RoboBees Out-Swarm the Real Ones?

Robotic pollinators are on the horizon, a prospect that deeply worries environmentalists.

May 1 2014, 8:10pm
Image: Greenpeace/YouTube

Imagine a future where robotic bees have taken over the role of the real ones—that's the cautionary tale conjured in Greenpeace's latest bit of activist fiction. The video may seem winkingly over-the-top in the advocacy group's trademark style, but it actually raises a fairly pertinent question: Are we okay with the notion that robo-bees could one day pollinate our flora? Because Harvard researchers think we'll live to see the day when we will. 

Admittedly, I was feeling slightly panicked after watching Greenpeace’s PSA—it depicts a place where bees have died out and have been replaced by swarms of robotic “NewBees”—so I reached out to the environmental org to learn more about their concerns. 

“The concept of robo-bees fulfilling the role of natural pollinators is science fiction and should stay science fiction,” Isabelle Phillipe told me over email. “It would be foolish to rely on a technological fix to the problem of bee decline due to industrial agriculture when there is no guarantee that such a fix would not make the problems worse.”

That, of course, is a matter of opinion. Over at Harvard’s School of Engineering and Applied Sciences (SEAS), researchers are busy developing ‘RoboBees,’ a technology they believe has many benefits for society, including pollination.  

“It could impact everything from crop pollination to search and rescue—and that’s science, not fiction,” explains a SEAS video.

Colony Collapse Disorder has decimated the population of bees necessary to pollinate crucial food crops. In a paper titled “Programming Micro-Aerial Vehicle Swarms with Karma” Harvard researchers Karthik Dantu and Bryan Kate explain how RoboBees could one day autonomously pollinate crops—specifically, alfalfa—using a system they’ve dubbed “Karma.”

In the system the team has devised, RoboBees would be dispatched as drones from a computer in the field which acts as a kind of hive. The hive can process the RoboBees’ data, and administrate their actions when dispatched into the rows of crops to be pollinated and monitored.

The system is relatively simple because the drones themselves only perform a limited number of functions that don’t have to be coordinated in real-time. The application programmer only needs to describe how a RoboBee should behave when it’s dispatched, and the little fellas take care of the rest themselves.

The SEAS team recently took their RoboBees for their first controlled flight. Though the little robo-critters don’t look much like real bees, they hover and flit about on translucent wings just like the real thing.

Despite the similarities between the robotic bees being developed at SEAS and and Greenpeace’s imagined scenario, Wood and  see see their work not as an arbiter of the apocalypse, but a wealth of potential innovation.

Robotic bees are great guinea pigs for exploring ways to use applied robotics—tackling fundamental questions in materials science, fluid mechanics, controls, circuit design, manufacturing, and computer science.

“To do something useful with a bunch of simple robots, we look to bee colonies as inspiration,” Wood told me. “How do they communicate? How do they share resources? How do they make decisions? The answers to these and other questions provide a great starting point for this project.”

Insects are ideal models for roboticists. Image: SEAS

There are many potential uses: “Most are similar to applications for other autonomous robots such as search and rescue and hazardous environment exploration,” Woods said. The SEAS website lists military surveillance and weather mapping as a couple examples.

“One of the potential applications of our work with small robots might someday be to artificially pollinate crops. However, we are at least 20 years away from that possibility,” said Wood. “Furthermore, even if these robots are able to be used for pollination, it should only be as a stopgap measure while the actual causes of [Colony Collapse Disorder] are determined and a solution is implemented to restore natural pollinators.”

Faced with full-on colony collapse disorder, we may very well turn to the RoboBees one day. But let’s listen to critics like Greenpeace, too and search out alternate solutions as well. Let's work on saving the real bees, maybe, before we enlist swarms of robot pollinators.