A flying robot the size of a bee will navigate and learn using a simulated insect brain running on a graphics chip small enough to carry onboard. Today’s autonomous drones rely on pre-programmed rules and struggle when faced with unexpected obstacles or unfamiliar environments. Animals like honeybees solve this problem with brains that contain only a million neurons—100,000 times fewer than a human brain—yet can learn, multitask, and adapt to novel situations within a single lifetime. This project fuses experimental neuroscience with computational modelling to build a robot controller that mimics those neural circuits, running on lightweight, energy-efficient graphics hardware designed for mobile devices. If successful, the work could transform how robots handle uncertainty in real-world settings—from search-and-rescue operations in collapsed buildings to autonomous inspection of infrastructure like power lines or pipelines. It also opens a path toward robots that learn on the fly without needing constant human reprogramming or cloud-based processing. The project is primarily fundamental science, exploring how tiny brains achieve such flexible behaviour, but the engineering goal—a flying robot with a bee’s navigational abilities—is concrete and testable.
View original technical description
What if we could design an autonomous flying robot with the navigational and learning abilities of a honeybee? Such a computationally and energy-efficient autonomous robot would represent a step-change in robotics technology, and is precisely what the 'Brains on Board' project aims to achieve. Autonomous control of mobile robots requires robustness to environmental and sensory uncertainty, and the flexibility to deal with novel environments and scenarios. Animals solve these problems through having flexible brains capable of unsupervised pattern detection and learning. Even 'small'-brained animals like bees exhibit sophisticated learning and navigation abilities using very efficient brains of only up to 1 million neurons, 100,000 times fewer than in a human brain. Crucially, these mini-brains nevertheless support high levels of multi-tasking and they are adaptable, within the lifetime of an individual, to completely novel scenarios; this is in marked contrast to typical control engineering solutions. This project will fuse computational and experimental neuroscience to develop a ground-breaking new class of highly efficient 'brain on board' robot controllers, able to exhibit adaptive behaviour while running on powerful yet lightweight General-Purpose Graphics Processing Unit hardware, now emerging for the mobile devices market. This will be demonstrated via autonomous and adaptive control of a flying robot, using an on-board computational simulation of the bee's neural circuits; an unprecedented achievement representing a step-change in robotics technology.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know