Dragonfly inspires predictive vision for driverless cars

(Credit: Don Debold)

The system, designed with autonomous vehicles in mind, takes inspiration from a neuron in dragonfly brains that anticipates movement. Known as CSTMD1, the neuron increases the insect’s response in a small ‘focus area’ just ahead of the object being tracked. If a target disappears, the focus area spreads over time, following the previous path and allowing the dragonfly’s brain to predict where the target is most likely to reappear. This ability is being replicated in a small robot in Adelaide to test its potential for machine vision.

“It is one thing for artificial systems to be able to see moving targets, but tracing movement so it can move out of the way of those things is a really important aspect to self-steering vehicles,” said Dr Steven Wiederman, from Adelaide University’s Medical School.

“What we found was the neuron in dragonflies not only predicted where a target would reappear, it also traced movement from one eye to the other – even across the brain hemispheres.”