Honeybee-inspired navigation enables efficient robot homing

What the study found: The study reports a navigation strategy called Bee-Nav, inspired by honeybee visual learning flights, that lets a robot learn a home vector and return home efficiently. The authors say it combines path integration, a way of estimating position from movement, with visual homing to correct drift.
Why the authors say this matters: The authors conclude that this strategy will be vital for resource-constrained robots that need to travel to and from a home location. They also say it offers new perspectives on insect navigation, including how visual learning shapes homing trajectories and the nature of cognitive maps.
What the researchers tested: The researchers built equivalent robotic learning flights in which a tiny neural network was trained to map omnidirectional images to a home vector based on path integration. They evaluated the approach in simulations and in real-world indoor and outdoor flights with a small drone.
What worked and what didn't: In simulation, the neural network needed training on only about 0.25–10.00% of the total flight area for realistic path integration accuracies. In experiments, the drone returned to within 0.5 m of home for 100% of 30–110 m flights and 70% of 200–600 m flights in windy conditions, using 3.4-kB and 42-kB neural networks, respectively.
What to keep in mind: The abstract does not describe detailed limitations beyond the reported conditions. The real-world success rate was lower for the longer windy flights than for the shorter flights.

Key points

• Bee-Nav is a honeybee-inspired navigation strategy for robots.
• The system uses a tiny neural network to map omnidirectional images to a home vector.
• Simulation results suggested training was needed on only about 0.25–10.00% of the total flight area.
• A small drone returned to within 0.5 m of home in 100% of 30–110 m flights and 70% of 200–600 m flights in windy conditions.
• The authors say the approach may matter for resource-constrained robots and for understanding insect navigation.