A new study of fruit flies may help us rethink how we teach self-driving cars to make decisions

The study, “Robustness and plasticity in Drosophila heat avoidance,” was published in the journal of Nature Communications on April 6, 2021. This research finds that fruit flies may be better drivers than autonomous cars. This study is timely, as a recent AAA survey revealed that the majority of people fear full self-driving cars.

Marco Gallio, professor in the Department of Neurobiology, and members of his lab found that drosophila (fruit flies) make decisions based on memory and continuous learning to perform basic functions. The flexible decision-making of flies is crucial to new research centered on how to program the allusive self-driving cars.

“Animal behavior, especially that of insects, is often considered largely fixed and hard-wired — like machines,” said Professor Marco Gallio. “Most people have a hard time imagining that animals as different from us as a fruit fly may possess complex brain functions, such as the ability to learn, remember or make decisions.”

In addition to their study which used plastic chambers and video recordings to map how flies reacted to boundaries and invisible barriers, the research team created an interactive simulation, called “GallioKart 1.0.” The simulation demonstrates how cars modeled after flies can “evolve” with biological inspired principles. However, it also demonstrates how regular vehicles do not “evolve,” remaining inflexible.

Watch a vehicle simulation in comparison to a fly below:

Professor Marco Gallio

Read more about this study in the Northwestern Now article: “Understanding fruit fly behavior may be next step toward autonomous vehicles.”

Learn more about Professor Marco Gallio’s lab here.