In summer, anyone may have failed while trying to catch drosopgila that hover around food. Insects, including drosopgila, have a high visual ability to accurately detect approaching objects and quickly avoid them. However, what happen within the drosopgila's brain in that short moment have been found out.

Recently, a research team led by Professor Kim An-mo, major in Biomedical Engineering at Hanyang University, identified a visuomotor circuit mechanism that detects the risk of collision or predation in the brain of a flying Drosophila melanogaster and induces evasive flight.

The research team identified a circuit that directly connects from the perspective of drosopgila to the wings to create fast reflex behavior. In particular, it was revealed that certain visual cells integrate several types of risk stimuli directly related to animal survival at the initial visual stage and quickly transmit them through another nerve cell to the motor system responsible for the wing movement of drosopgila.

Detecting and avoiding dangerous objects in complex visual environments is a brain function that even mammals, including humans, perform effectively. Therefore, it is expected that future comparative studies between insects and mammals will be able to identify the general principles of neural network algorithms responsible for visual avoidance behavior.

Professor Kim said, "Currently, there are several limitations in the technology of avoiding autonomous collision of flying objects such as drones." "If we develop collision avoidance algorithms based on the data obtained through this study and implement ‘reverse engineering of the drosopgila nervous system’ that applies them to engineering systems such as drones, we will be able to develop autonomous robots that overcome various existing limitations," he said.

This study was published on January 4 in the international journal, “Current Biology”, IF=10.9 in recognition of its academic excellence in the field of Biology. Kim Hyo-sun, a Ph.D. student at Hanyang University, participated in the paper as the lead author, and Park Ha-yoon and Lee Ju-won, graduates of Hanyang University, participated as co-lead authors.