· Research & technology· 日本語で読む
Robot swarms classify environmental shapes
leaderobot.com reported that a framework for swarm robots to classify environmental shapes through communication with nearby robots was also validated in real-robot experiments.
What happened
According to leaderobot.com, a research team proposed a framework called “spectral swarm robotics” that classifies environmental shapes through local communication among robots, without relying on localization or mapping. The findings were published in Science Robotics. The article reported classification accuracy of 73% in real-robot experiments using Kilobots and 100% in experiments using Pogobots.
Sources
Translated from the Japanese original: /news/leaderobot-com-science-robotics-keisai-no-gun-robotto-kenkyu-a30438. The Japanese version is authoritative if they differ.