Research

Duke University researchers have demonstrated Argus, a 20-legged robot whose sea-urchin-like body gives it near-uniform motion and perception in every direction.

Researchers have developed printed magnetic sensors made from naturally sourced and lower-toxicity materials including iron, iron oxide, cellulose, and starch.

Researchers at the University of Osaka and collaborators have developed a cyborg insect system that reads heartbeat activity, neural signal features, and body movement from cockroaches.

Bosch researchers developed a decentralized collaborative mapping system inspired by ant pheromone trails and swarm intelligence. Vehicles share virtual pheromone markers to distribute mapping effort and resolve uncertainty without central coordination, translating stigmergic biological communication into autonomous transport infrastructure.

Just call these tiny autonomous construction robots “antdroids” Biomimicry is a significant driver in robot design; like ants, RAnts are too small to possess human-level cognition and memory, so if they’re going to achieve complex goals, they need to out-source, or exbody, their data-processing to their environment.

Honeybee-inspired swarm framework targets resilient planetary exploration The history of planetary exploration has long been a chronicle of singular, heroic machines.

Researchers from SUPSI and ETH Zurich developed a firefly-inspired visible-light communication system for nano-drone swarms using LEDs and convolutional neural networks for message decoding.

While the new AI companies have focussed on LLMs, Google has been spinning off projects from Demis Hassabis’ DeepMind, in AlphaGo, AlphaEvolve, and the Noble prize-winning AlphaFold.

Researchers at the University of Hong Kong demonstrated small snail-inspired modular robots that can physically attach to one another using suction-based coupling.

A recent patent landscape analysis highlights how biologically inspired swarm coordination methods are being displaced in some military and industrial drone systems by multi-agent reinforcement learning, while still influencing routing, resilience, and distributed control architectures.