Engineers have spent years perfecting advanced sensors, radar, and GPS systems that help drones identify and reach their targets. Meanwhile, bees use five eyes, antennae, and tiny brains to travel to and from food sources miles away from their hives. Instead, if UAVs are built with vision systems inspired by a bee, they could navigate autonomously without needing the radar or GPS.
The bee-inspired drone might be closer to reality. Professor James Marshall of Sheffield University and his team are studying how bees think and fly in order to develop new navigation systems for drones. “These animals are clearly using simple and elegant strategies, honed by thousands of years of evolution.”
Reverse engineering bee brains could be the key
While on the other side of the globe, Interface Focus brings together biologists and engineers to discuss a topic that’s relatively straightforward and, well, pretty empirically cool. “It’s completely focused on how bees fly and how that can help us build flying robots,” said Lentink, assistant professor of mechanical engineering at Stanford. He is a part of the team working on the project.
Moreover, in a presentation to the American Association for the Advancement of Science conference in Seattle, Professor James Marshall demonstrated how he and his team are reverse engineering bee brains to create a working prototype of the stated concept. He explained: “Bees are really consummate visual navigators. They can navigate a complex 3D environment with minimal learning very robustly, using only a million neurons in a cubic millimeter of brain.” The project has a £4.8 million grant from UK Research and Innovation. The team is planning to commercialize the technology. They already have established a company called Opteran Technologies that is working towards raising £ 1.5 million in seed funding.
Bee brains are better than AI
The team is carrying out research particularly in two areas — to work out how honeybees and bumblebees can reliably navigate over several kilometers, learning the features that will enable them to return to their hives. The brains of bees work very differently compared to human-engineered AI systems. When bees fly out in search of nectar they optimize their distance from the nest and can learn extremely fast in new environments, unlike AI. To understand how bees learn and optimize distances, researchers needed to track these insects. Thus, they attached tiny radar transponders on the bees. Instead of building bee-sized drone swarms, researchers are intently studying how bees function and are instead putting brains on 250 and 600-gram drones. Drones with brains of their own would be much faster compared to the ones that rely on ground-based pilots or machine learning AI.
Bee-inspired drones could have endless possibilities, to begin with. Using this automated navigation drones can easily excel at civic surveillance, inspections, and even drone deliveries. With an onboard brain, drones will be able to analyze and ‘think’ around complex barriers. The successful testing and implementation of this concept will pave the way for an era of truly intelligent and autonomous drones.