Fruit fly inspires AI chip to help drones avoid obstacles, save power

An NTHU team has developed an AI chip that follows the streamlined function of a fruit fly optic nerve. A major limitation for aerial drones is the tradeoff between weight and battery capacity, which limits their range and usefulness for applications such as agriculture and infrastructure inspection. To address this challenge, a multidisciplinary team at National Tsing Hua University in Hsinchu, Taiwan, has developed an artificial intelligence processor that mimics the optical nerves of a fruit fly. This AI chip enables unmanned aerial vehicles (UAVs) to automatically avoid obstacles while staying in an "ultra-power-saving mode," said the researchers. The team was led by professors Tang Kea-tiong of the Department of Electrical Engineering and Lo Chung-chuan of the Department of Life Sciences at National Tsing Hua University (NTHU).