Research team develops biomimetic vision system inspired by praying mantis eyes.
Self-driving cars sometimes crash because their visual systems struggle to process static or slow-moving objects in 3D space, similar to how many insects see the world. However, praying mantises have a unique advantage – their compound eyes provide binocular vision with depth perception in 3D space.
To overcome limitations in machine learning, researchers at the University of Virginia School of Engineering and Applied Science developed artificial compound eyes that mimic praying mantis eyes. These innovative “eyes” use microlenses, photodiodes, and flexible semiconductor materials to emulate the convex shapes and faceted positions within mantis eyes.
The team’s biomimetic system provides a wide field of view, superior depth perception, and precise spatial awareness in real-time, essential for applications like low-power vehicles, drones, self-driving vehicles, robotic assembly, surveillance systems, smart home devices, and more.
One notable achievement is the potential reduction in power consumption by over 400 times compared to traditional visual systems. The system also processes visual information on the edge, nearly eliminating data transfer costs while minimizing energy usage.
The key to this breakthrough lies in the integration of flexible semiconductor materials, conformal devices that preserve exact angles, an in-sensor memory component, and unique post-processing algorithms. The sensor array continuously monitors changes in the scene, identifying which pixels have changed and encoding this information into smaller data sets for processing.
This biomimetic approach mirrors how insects perceive the world through visual cues, differentiating pixels between scenes to understand motion and spatial data. The seamless fusion of advanced materials and algorithms enables real-time, efficient, and accurate 3D spatiotemporal perception.
Source: https://techxplore.com/news/2024-07-team-biomimetic-vision-based-praying.html