Sign in to view
member only content

University researchers are taking inspiration from the dragonfly to make dramatic improvements in the way driverless cars predict the flow of traffic. The researchers, from the University of Adelaide in South Australia and Lund University in Sweden, discovered a neuron in the brain of a dragonfly that anticipates movement and predicts the trajectory of their prey. These neurons are called Small Target Motion Detectors (STMD), and they increase the insect’s response in a small area just ahead of the object they are tracking. If the object disappears – if it goes behind something, for example – then the neuron is able to predict where the object should reappear.

Initially, the researchers are replicating the properties of the neurons in a small robot. And if the results are as positive as expected, the technology could then be replicated in driverless cars and other robotic vision systems. It is believed that this is the first time a target-tracking model that has been inspired by insect neurophysiology has been tested on a robot under real-world conditions.

Steven Wiederman, a research supervisor and lecturer at Adelaide’s Medical School, said: “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 things is a really important aspect to self-steering vehicles.”

There have been a number of recent advancements in the world of driverless vehicles, including a driverless and emission-free truck, and a self-driving cargo ship – which will launch in 2018. Are there any other nods to nature that driverless cars could make to improve them?